Sun Unveils RAID-Less Storage Appliance

pisadinho writes "eWEEK's Chris Preimesberger explains how Sun Microsystems has completely discarded RAID volume management in its new Amber Road storage boxes, released today. Because it uses the Zettabyte File System, the Amber Road has eliminated the use of RAID arrays, RAID controllers and volume management software — meaning that it's very fast and easy to use."

Comments

Sun shoots, and... well, you already know.

[AKAImBatman]

Pricing:
Sun Storage 7110: $10,995 for 2TB;
Sun Storage 7210 starts at $34,995 for 11.5TB;
Sun Storage 7410: Single node version starts at $57,490 for 12TB;
cluster version (with two server nodes) starts at $89,490 for 12TB.

Correct me if I'm wrong, but doesn't charging enterprise prices for simplified hardware that relies on commodity software solutions, kind of defeat the point?

Unless I'm misunderstanding this hardware, the entire idea is to move data safety away from hardware redundancy toward software-driven duplication. In that way, the data is safe from failure in the same way that GoogleFS protects against individual machine failures. The only difference is that Google probably doesn't pay $11,000 for 2TB of storage. :-/

One of these days, I really will understand why Sun regularly shoots themselves in the foot. Until then, I suppose I must trust them to somehow find a customer who's willing to pay exorbitant prices for an otherwise good idea. (i.e. I'd really love to see Sun bring Google-style reliability from unreliability to the market.)

BTW, here's the link to Sun's marketing on this:
http://www.sun.com/storage/disk_systems/unified_storage/index.jsp

It's actually pretty cool tech. Sun could own the market if they just understood how the market views pricing and features.

Re:Sun shoots, and... well, you already know.

[poot_rootbeer]

I suppose I must trust them to somehow find a customer who's willing to pay exorbitant prices for an otherwise good idea.

Have you worked with any of Sun's customers recently? I believe P.T. Barnum was involved in the development of their business strategy.

Re:Sun shoots, and... well, you already know.

[jmorris42]

> Correct me if I'm wrong, but doesn't charging enterprise prices for simplified hardware
> that relies on commodity software solutions, kind of defeat the point?

Yea, that is amazing. Ya could put in a pair of 1U servers with RAID1 on each for a fraction of that pricetag. Use any of a number of ways to make the two units cluster, including using OpenSolaris and you get everything they are selling except the pretty front end for about half the sticker, Go SCSI/SAS on all of the drives in 2U machines if you want to spend about what they are charging and still come out with a redundant cluster.

Re:Sun shoots, and... well, you already know.

[darkjedi521]

Some of that is the custom gear that goes into making those beasts. Yes, it might eliminate the hardware raid card, but in the case of the 7210, the hardware to drive 48 SATA drives and not saturate the bus still isn't cheap. Plus hotswap everything, and the price quickly rises to something close to what Sun is charging. I use 4 x4500s at work for a single cluster, and they are a hell of a lot cheaper for that capacity than the traditional rack of fiber arrays/raid controllers/etc. The 4 of them cost me what another vendor wanted for half the raw storage (and far less usable storage).

Re:Sun shoots, and... well, you already know.

[Archangel Michael]

I could build a setup that would be way more powerful and less costly and more storage for way less.

What kind of drugs are these people on? two TB for 10K ???? ARE YOU NUTS SUN????

I could probably build this using SSD for less. SHEESH

Re:Sun shoots, and... well, you already know.

[AKAImBatman]

the hardware to drive 48 SATA drives and not saturate the bus still isn't cheap.

If you're driving 48 SATA drives on one bus, you're:

A) Not looking at the minimum 11.5TB layout
B) Not paying $35,000
C) Not a small-business customer

Which brings me back to: Sun is promising to target the small business and yet totally missed the mark. This is Enterprise hardware.

Re:Sun shoots, and... well, you already know.

[chrisj_0]

I don't think you could get the IOPs (or anywhere near) out of a pair of off the shelf 1u servers that they're advertising. I just checked dells website, their new AX4-5i (iscsi SAN) starts at over $14,000 and that only includes the 4x 750GB vault drives. Add 4x 1TB SATA drives (at $1,100 each) in a RAID 10 and you still wouldn't get the IOPs that Sun is talking about. This product looks to try and take a market share from the FC SAN vendors, not companies that want their in house geek to build a "cluster storage solution".

Re:Sun shoots, and... well, you already know.

[Wesley Felter]

the hardware to drive 48 SATA drives and not saturate the bus still isn't cheap.

Actually, SAS HBAs and JBODs (which is what Sun is using) are cheap; that's why it's odd that Sun is charging so much. For example, the 7210 is the same hardware as the X4540 yet it appears to cost much more.

Re:Sun shoots, and... well, you already know.

[rainer_d]

I don't think it's that expensive.
I use Promise's VtrakJ610s at work (16x1TB SATA), and it cost about half that - but I still need a server for it (DL385 in our case). And I need to fit the disks myself (16x4 countersunk screws...) into the ultra-cheap harddrive containers.
A MSA70 full of SAS-disks (25) costs 10k, IIRC - but you need a server, HBAs etc.
I'm soooooo sick of the "I could build one for XXX% less using YYY"-comments.
Please, all the winers: go and start your own company selling and supporting storage-systems.
Good night and good luck....

Re:Sun shoots, and... well, you already know.

[Famanoran]

With the same level of assurance that the solution will operate, first time - every time?

With the same level of confidence that Some Vendor will bend over backwards to fix it if it doesn't work?

Will your solution be as well tested and engineered?

It's not like you can just grab 3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

Your "home brew" solution will not meet any of the objectives Sun are achieving with this product. Your spindle count will suck, so concurrent access will be slow. You will probably be limited to one of iSCSI, CIFS, NFS or WebDAV, I doubt your solution would have all - and if it does, the integration will suck.

Will your solution have the diagnostic tools that Sun can provide? Oh wait, you don't have the millions of dollars to invest in engineering quality diagnostics, right from disk analysis (Sector scanning, remapping, etc) through to performance related faults? Well, then your solution will suck. What about snap-cloning?

In short, yes - storage is cheap. You can grab large drives very cheaply and put together something that works. That does not mean it will be good. Production quality storage is expensive, and for good bloody reason.

As for doing this using SSD storage, that's just ridiculous. 2048GB of storage would be at least 16 128GB SSD disks - this is not counting any disks for redundancy (i.e, raid-5/6 parity), or hot-spares. Assuming 2 drives for RAID-6 parity and 2 hot swaps, you'd need 20 SSD disks - with 10 grand, you're expecting to pay $500 per disk - and no other hardware, i.e, motherboard, case, cooling (more important than you think), etc.

So, until you have a clue about designing production quality storage systems, please refrain from making statements you have no clue about, you're only serving to confuse those people who are actually interested in what this product has to offer them. Keep to building crappy 3 or 4 disk RAID-5 systems using extremely large drives for storing your music, movies and pr0n on, but don't ever ever ever ever think about using those in any situation where your financial livelihood depends on that data.

Re:Sun shoots, and... well, you already know.

[edsousa]

You all must be on drugs.. DRAM, SSD & HDD. You got all the software you will ever need. 75% less energy than same capacity solutions.
Do you ever use an EMC Clariion? Did you check those insane prices? The CX4-120 costs around $4000 and the software for 1 user another $4000 (prices vary)
The folks at Sun are not stupid, specially when it is HPC.
And BTW, storage space isn't the most important thing. Have you ever wondered why Google keeps offering more space for GMail? They need huge amounts of IOPS (Input/Output operations Per Second) and the standard way is adding more and more HDDs.
Or Sun style, using an approach that I will consider when my enterprise starts.

Re:Sun shoots, and... well, you already know.

[Ngarrang]

I could build a setup that would be way more powerful and less costly and more storage for way less.

What kind of drugs are these people on? two TB for 10K ???? ARE YOU NUTS SUN????

I could probably build this using SSD for less. SHEESH

Promise VTrack 16-drive array...$4,500
(16) Seagate 1TB SATA300 @ $130 each...$2,080

$6,580 for 16TB of disk space connected to one server.

Then you install OpenSolaris and install ZFS.

Re:Sun shoots, and... well, you already know.

[sexconker]

It's SUN.
They're goal is to stay relevant, their strategy is to make headlines.

It's like a cross between a child acting up for attention and an emo cutting themselves.

Re:Sun shoots, and... well, you already know.

[Archangel Michael]

Okay perhaps not with SSD.

I've built 1.5 TB systems over a year ago, RAID for $1200. FULL including swappable drives, gig ethernet and plenty 2 GB of ram to cache the system. They are FAST and reliable. They worked FIRST TIME, and have worked exactly perfectly for 1.5 years. Drives fail, we have hot swaps available. While not quite 2 TB, they are also 1.5 years old now, and I'll be replacing them in another 1.5 years with bigger drive systems.

And these, will be spares and lower priority sytems when I update them with newer stuff in a year or two.

Expensive Technology for the sake of all that other stuff you listed is just silly. It is exactly why SUN doesn't get it, and why some pointy hair boss is buys the bs.

Production quality storage means that it works for the time needed. I've actually had WORSE reliability from Name brand "Server" quality stuff. We've got HP Proliant Servers in production, and at least THREE from three different lots have all failed due to MOBO Failures. While they do send out a tech to replace the MOBO, it is really really annoying to have to tell people that the server is down because the MOBO failed. And all the great diagnostic tools HP has on those servers didn't predict nor would they fix the errors.

You can build it 1/2 as much then you can easily have two on hand, in case one dies.

Re:Sun shoots, and... well, you already know.

[AKAImBatman]

With the same level of assurance that the solution will operate, first time - every time?

Sure.

With the same level of confidence that Some Vendor will bend over backwards to fix it if it doesn't work?

Heck, I'll even throw in the same vendor!

Will your solution be as well tested and engineered?

Even better. It will have had the same testing and engineering, PLUS a pre-existing history of operating in the marketplace.

I give you, the Sun Fire X4500 Server:

12TB (48x250GB) - $23,995.00
24TB (48 x 500GB) - $34,995.00
48TB (48 x 1TB) - $61,995.00

Let us compare with Sun's new line, shall we?

11.5 TB (46 x 250GB) - $34,995.00
22.5 TB (45 x 500GB) - $71,995.00
44.0 TB (44 x 1TB) - $117,995.00

So... twice the price for the same storage? To steal a line from a very famous "programmer":

Brillant

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

With the same level of assurance that the solution will operate, first time - every time?

Where are you getting your Sun hardware?

Re:Sun shoots, and... well, you already know.

[Fweeky]

You forgot the SSD's for ZFS secondary ARC cache. Oh, and the server.

Re:Sun shoots, and... well, you already know.

[Guspaz]

You can build a single server containing 12TB powered by ZFS and RAIDZ for about $3000-3500 Canadian, including hot-swap drive bays. And the drives.

Sure, there's probably a lot of redundancy in these Sun boxes, but if they're relying on ZFS/RAIDZ to provide much of the reliability, and you build your $3500 box (which is housed in a mid-tower case with 9 drive bays) using OpenSolaris, you're most of the way there. At that point, you've got the data reliability, you just might not have quite the same uptime.

The thing is, for the prices they charge for even a single node ($57,490), I could build half a dozen of my commodity boxes and replicate the data between them. And it would still cost less. And probably perform faster; I could also cluster with redundancy for less than they charge, and they're using commodity 7200RPM SATA drives.

Re:Sun shoots, and... well, you already know.

[Architect_sasyr]

We've got HP Proliant Servers in production

Some things you should keep to yourself no matter how bad it operates ;)

This is a real case of quality, support and "bling" factor. To use a (bad) car analogy: There is no need to buy a Mercedes when you can own a Nissan for half the price and it has exactly the same features (it may even be more powerful in some cases). However anyone can drive a Nissan (or can afford to), so there is a certain bling factor to driving the Mercedes. Just like there is a hell of a "bling" factor to owning Sun equipment as opposed to the "hack job" we can all put together. Personally I would prefer to spend twice as much and know that it's no longer my problem, even if it crashes, but that's just the opinion of one Network Admin.

Completely off hand: I've never had a mobo fail in any server, IBM or Dell based.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

You are aware that those prices are for SSDs, not for spinning disks, right?

Re:Sun shoots, and... well, you already know.

[jebrew]

Just because I'm too lazy: What are the IOPs they're advertising?

I'm curious because I've seen off the shelf AMD 2U servers hit over 1M IOPs...can they do better?

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

It is SSD.

Re:Sun shoots, and... well, you already know.

[PopeRatzo]

I suppose I must trust them to somehow find a customer who's willing to pay exorbitant prices for an otherwise good idea.

It works for Apple.

Re:Sun shoots, and... well, you already know.

[SuperQ]

Yea, I recently built an 8T server. It cost me about $5000, and has no raid controller, and uses linux software raid.

If I really wanted, I could buy a second $5000 server and do DRBD between them to have 2x redundancy.

Re:Sun shoots, and... well, you already know.

[Godji]

the entire idea is to move data safety away from hardware redundancy toward software-driven duplication

You are exactly right. When you pay the exorbitant price, you pay for great hardware, the development of great software (which you could have gotten for free), the convenience of a prepackaged solution, and for the hardware and software support.

Should anything happen to these machines, you can always get your data back. If you can't afford another set of machines like these, simply plug the drives into anything that runs Solaris (or generally ZFS), and you have your data.

Just because it's open doesn't mean it has to be cheap. But Solaris is open source, so if you don't like these, roll your own and support it yourself.

Re:Sun shoots, and... well, you already know.

[Fweeky]

• Sun X4240, cheap from third party reseller not necessarily in the best of configurations: ~£2700.
• 2 cheap and cheerful SSD's for ARC second level cache: £300. £1200 if you want ones with decent write performance.
• 14*146GB 2.5" 10kRPM SAS disks: £2100.

Even if you put it together and test it using slave labour, you're not getting much change from $11k.

Sure, you could just plonk three 1.5T Seagates in there, shove a RAIDZ over it and call it a day, and that would be fine for some uses, but it's not really something you buy a storage appliance for, is it?

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Look at EMC prices for Hulk or Maui or Centerra, which seems to be what this competes against. Or perhaps Isilon.

The price per GB is actually pretty reasonable for list prices. A lot of enterprise customers pay more than twice that if they got a deal, and more than six times that if they didn't ($30/GB anyone?). I bet if you were a customer with leverage you could get it down to less than $1/GB on this gear.

I know you can get a couple of 1TB drives via mail order and put them in a linux box, but that is not the same market, and I suspect you know that. Size isn't the only storage metric people care about. You can get a single box with 12 1TB SAS or SATA drives in it from several vendors.

Can you serve iSCSI or FC out of GoogleFS? (Solaris will enable FC target mode via COMSTAR, something I'm watching closely.) How's the GoogleFS performance with an RDBMS?

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

I've had a lot more off-the-shelf success with HP and Sun server gear (the non-x86/x86_64 stuff) than with your typical server-class HP, Dell, etc.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

You're not driving 48 SATA drives on one bus. The x4500 has 6 disk controllers each driving 8 disks. There appear to be 3 PCI buses, each running 2 controllers. I'm not deep into the magic of storage, but I can at least putty into one of our x4500s and take a look at what's going on before I start talking.

Re:Sun shoots, and... well, you already know.

[segedunum]

It's not like you can just grab 3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

That's exactly what Google and many others do, and they spend their money, and significantly less than this, on managing that storage effectively. It works. When it boils down to it, you can have all the exorbitantly expensive and brilliant 'enterprise ready' tools you want but the bottom line is you need redundancy - and that's pretty much it.

Your "home brew" solution will not meet any of the objectives Sun are achieving with this product.

Sun say they are targeting small businesses, and they have lost already with this poor showing. They have advanced no further than when they stiffed all the Cobalt Cube customers and withdrew the product, who then went out and bought Windows SBS servers ;-). If you think people are going to jack them in for this then you need a stiff drink.

Your spindle count will suck, so concurrent access will be slow.

Ahhh, shit. I'm heart broken. What I'd like to know is how a small business will handle a behemoth like that, how they'll fund the electricity for all those drives and who'll manage it all. I expect that will be an ongoing cost to Sun support ;-).

Keep to building crappy 3 or 4 disk RAID-5 systems using extremely large drives for storing your music, movies and pr0n on, but don't ever ever ever ever think about using those in any situation where your financial livelihood depends on that data.

I have news for you. People have been doing it for years, and the reason why Sun's business has gone down the toilet to commodity servers, Linux and Windows, especially with small businesses, for the past ten years is exactly for this reason.

Sun need to stop pretending that they can package up some commodity shit with some features very, very, very, very, very few need (and is waaaaaaaaaaaaay outside their target market) and label it as 'enterprise ready', which they think justifies an exorbitant price tag and ongoing support. They lost with this strategy with x86 and Solaris where they tried to protect SPARC, they lost with the exodus from SPARC after the dot com boom and they will keep on losing.

Re:Sun shoots, and... well, you already know.

[iamme9182]

from a sun blog so take it with a grain of salt, but here are benchmarks http://blogs.sun.com/hotnets/entry/analyzing_the_sun_storage_7000

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

You are aware those prices are for spinning disks with a tiny amount of SSD cache, right?

Re:Sun shoots, and... well, you already know.

[runningduck]

It seems others disagree with you about 2TB SSD systems.

http://www.superssd.com/products/ramsan-500/

Ironically, these devices are less than an order of magnitude more than the Sun storage.

Re:Sun shoots, and... well, you already know.

[jebrew]

I've got to say, from an IOPs perspective, that's extremely disappointing. ~24k IO/s max @ 8k. By their own chart they're getting about 233 IO/s per disk. I can get 1500 on a USB key on my desktop @ 8k. Granted, that's only running in 1 thread (they're using 128 which comes with some overhead).

I dunno, it just seems really low. I expected something more in the 200k-300k range.

Am I missing something critical? Are they running over a 100Mb network?

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

I do storage using HP DL380s hooked to some Promise VTJ310SDs (12 disks in 2U trumps 16 in 3), and my math says this Sun thing is kind of expensive unless you're really going to be maxing out the speed.

Using my storage cost spreadsheet, it looks like I'd expect to pay about $40k for roughly the same capabilities that Sun is selling for $72k.

That said, it's not that big of a difference all things considered. We probably used another $20k in man-hours during the initial configuration and performance testing.

Re:Sun shoots, and... well, you already know.

[AKAImBatman]

How does 2TB of Solid State Memory compare to 2TB (14 x 146GB) of spinning disks? Apples to oranges, my friend. Apples to oranges.

(FWIW, you need to spend about $71 grand (!) to get a mere 18GB of SSD. Almost like a cracker-jack prize or something. :-P)

Re:Sun shoots, and... well, you already know.

[kandresen]

Just a comment about the 48 disk setup; it is not always about getting the most space, but often about getting fastest response time. In this case the important factor is the amount of spindles. 11.5TB divided on 48 disks would be ~240GB a disk. Many companies would want 48 70GB disks as they are not in need of more space, only faster response times.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

I'd take your comment more seriously if I couldn't actually buy one for XXX% less from exactly the same vendor.

Seriously if you don't think it's that expensive you deserve to get screwed. 5 minutes on Sun's own website will probably save you tens of thousands of dollars with their other products.

Re:Sun shoots, and... well, you already know.

[dfn_deux]

Look at what netapp charges for similarly sized boxes and I'm sure you'll see this is probably at a similar price parity for size/performance. Netapp's wafl-fs is basically doing the same thing as ZFS (see netapp's lawsuit against Sun for more details). The addition of upgradeable cache ram, (something netapp only introduced to their product line with the 3040 addon cache card), and SSD based zfs disk cache should really make this system haul ass and explains a good portion of the pricing model.

At the enterprise level the purchase price of the hardware is nothing compared to the support contract cost, which is what I think will either make or break this for sun. I know that if my netapps take a shit I will have new hardware onsite inside 4 hours, my faith in sun delivering the same level of support is similar to my faith in the tooth fairy.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

No it's not.

Re:Sun shoots, and... well, you already know.

[the_B0fh]

You're not reading what you quoted. The hardware to *DRIVE* the HDs are not cheap. You're not talking about using ONE PCI bus for the entire server, for example.

Re:Sun shoots, and... well, you already know.

[pod]

And I bet those are also pre-discount prices (~ 60%) that anyone that seriously buys Sun hardware would get. At that point, it's entirely price competitive with FC attached storage units from, say, Hitachi.

Re:Sun shoots, and... well, you already know.

[hlge]

You're missing the point, look at what Netapp and EMC are charging for a similar solution. So on the low end take the S7110 at 10,995 and compare that to the FAS250 at 42,000. OK the FAS250 gives you 4TB, but you are stuck with a slow MIPS CPU and 512MB RAM where the 7110 gives you a a fast QC AMD CPU and 8GB, then add some data services and you start to get a real price difference :) And it gets even more interesting when you start look at some mid range to high end configurations.

Re:Sun shoots, and... well, you already know.

[canuck57]

Will your solution be as well tested and engineered?

Sun doesn't do much engineering here. Grab a bunch of SATA drives engineered by someone else, put them on a SATA chipset engineered by someone else, skip adding RAID to cut engineering costs and you have a storage array. Someone in China crimps the cables and packs it in a box, one metal and then one cardboard.

But I could go to my local PC enthusiast store buy a pair of 800W or 1200W power supplies, a mobo with quad core, 8GB of RAM, 6 SATA ports built in, 2 x 12 port SATA add ons, 30 1.5TB Seagate drives. Then load Linux (or Solaris) for 45TB of the good stuff and RAID it the way you want. Cost, a fraction of Sun box per TB. BTW, 1.5TB of Seagate SATA is under $200 right now. I am sure I would get a discount if I said I wanted weekly orders of 300 to build 10 per week. Oh, add 2 x 1000BT for bonded Ethernet, need that I/O.

For just 2TB, you really only need 3 1.5TB Seagate drives, use on board ICH9R or ICR19R RAID 5, on the cheap. Might even be able to do that for not much more than $1000 with 8GB of RAM cache. Ooops, Solaris doesn't support many ICH9 configs, use Linux.

Re:Sun shoots, and... well, you already know.

[Tubal-Cain]

HP

Re:Sun shoots, and... well, you already know.

[Wesley Felter]

An x86 server motherboard is also cheap.

Re:Sun shoots, and... well, you already know.

[the_B0fh]

It's not like you can just grab 3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

That's exactly what Google and many others do, and they spend their money, and significantly less than this, on managing that storage effectively. It works. When it boils down to it, you can have all the exorbitantly expensive and brilliant 'enterprise ready' tools you want but the bottom line is you need redundancy - and that's pretty much it.

You're trolling, right? You spend money for reliability either on hardware engineering, or software engineering. If you're doing things on the scale Google is, obviously it saves you more money to do software engineering, since that can be replicated easily on a large scale. For most companies, even "enterprises", hardware reliability gives you a better bang for the buck, because you can't bloody afford multiple data centers.

Re:Sun shoots, and... well, you already know.

[Gumbercules!!]

I used to work for the largest Sun reseller in Western Australia and they loaned us early versions of this very equipment, configured by their own engineers. It was shyte. Utter shyte.

ZFS performed *way* below expectations and redundancy was atrocious. Removing disks to see what happened brought about totally random results of data loss.

In the end, I felt it was a better choice to go with a white box PC with Windows software RAID on the cheapest disks money could find. I was very unimpressed.

I'll want to see some serious demonstrations that this is not just marketing madness before I ever believe a word out of Sun again.

Re:Sun shoots, and... well, you already know.

[myifolder]

Dell sucks. I have had more problems with Dell than HP and would never buy another Dell product. HP all the way. I do agree about SUN products being very well build and so is IBM. But I will put it to you like this, every thing breaks and nothing is 100% and everyone like different things. That is why Dell is still in business today and the same goes for HP and IBM. One thing for sure this SUN line is way too much for me.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Yes, you just don't get it. We just paid NetApp $50,000 for 6TB, and it's worth the price. So $10,000 for 2TB of enterprise disk doesn't sound like that bad of a deal.

If you can build an enterprise level SAN that supports fiber channel at 4Gb I/O speeds for less, go into business. You'd have lots of customers.

Re:Sun shoots, and... well, you already know.

[jedidiah]

For most companies, the "cheap crap" is more than adequate.
They simply don't have the money to spend and they get by
perfectly well despite not having drunk the cool-aid.

Also, overpriced hardware with the label "enterprise"
plastered on it is not going to do anything to prevent
the need for a multiple data centers. Overpriced enterprise
hardware and multiple data centers solve different problems.

Re:Sun shoots, and... well, you already know.

[the_B0fh]

If you can't understand what I wrote, go back to trolling newegg please, kthx hand

Re:Sun shoots, and... well, you already know.

[zerocool^]

yeah, actually, I'm thinking about this, and I'm pretty sure you could get a 1-U box that fits 4 HDDs. Call it $300 for the case/power supply, get a cheap quad-core core2 from intel, 2GB of ram, and a motherboard with 2x gig-e nics, and then get an LSI SATA 150-6D or similar raid card + cache battery for about, say, $1200? Then stick 4x 1TB hard disks in a raid 1+0. I'm confident that you can get all the parts for less than $2000. Stick CentOS on it, bond the nics together, and run ISCSI or NFS or whatever. Bam, $2K for 2TB hotswap high performance RAID on a 2Gb/s interface in 1U.

It won't be SAS, which might be where the SUN price tag is coming from. But, meh, to me the cost doesn't justify the increased MTBF and read speed for most applications.

~X

Re:Sun shoots, and... well, you already know.

[funkboy]

With the same level of assurance that the solution will operate, first time - every time?

With the same level of confidence that Some Vendor will bend over backwards to fix it if it doesn't work?

Over on Dell's website, I just configured a 2u Powervault NF500 with dual quad xeons, 2TB worth of 15k rpm SAS drives and their excellent LSI-based PERC controllers, 10GigE, and every other expensive option I could find (including 24x7x4 support) and it came out to about $8500 (including the M$ tax).

So, despite your claims, Sun is once again overpriced to the tune of about 30%.

Re:Sun shoots, and... well, you already know.

[Thundersnatch]

You, sir, are never allowed anywhere near my data centers!

Re:Sun shoots, and... well, you already know.

[Hyppy]

People avoid SATA in high-IO environments for a reason.

Re:Sun shoots, and... well, you already know.

[Hyppy]

Oh, and the SCSI/SAS drives. SATA doesn't handle multiple IOs very well as it turns out, even with NCQ. If all you're doing is running a fileserver, fine, SATA will do. Databases or virtualization? SAS/SCSI is the only way to go.

Re:Sun shoots, and... well, you already know.

[Hyppy]

Boo on me for not RTFAing.

Re:Sun shoots, and... well, you already know.

[Hyppy]

And if I bothered to RTFA, I wouldn't feel like such a moron right now. Move along, nothing to see here.

Re:Sun shoots, and... well, you already know.

[tabrisnet]

11,000USD is _cheap_ compared to NetApp. And I have it on good authority that (at least for corporate, but not necessarily for production) that Google uses NetApp for storage. They have attempted multiple times to produce a better solution, but last I knew (about 6 months ago) they still hadn't found a good solution.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

At EDS, we only get a measly 55%. Who are you fucking to get 60%?

Re:Sun shoots, and... well, you already know.

[afidel]

I have 110 HP servers in production, I have had ONE motherboard replacement in the last 2.5 years. Your anecdote means nothing, only statistically significant samples do, thanks for playing.

Re:Sun shoots, and... well, you already know.

[Score Whore]

Is that 2U 1M IOPS unit racked right next to your 1U, 64K core, 1024 TeraHertz system?

FYI, a loaded HDS 9990V (i.e. hundreds of spindles and multiple gigabytes of cache) manages to provide 200,000 IOPS (SPC-1). Even Texas Memory Systems RamSAN 400 (i.e. SDRAM) can only make 400,000 IOPS. Hell, it was only a couple of weeks ago that TMS was announcing that they sold a RamSAN-5000, which is the only storage device I've ever seen specced to 1,000,000 IOPS. And it's 10 different RAM cached, flash backed units.

Re:Sun shoots, and... well, you already know.

[afidel]

Sun Gold or Platinum contracts are great, especially if you are a large customer with onsite techs. You pay out or every orifice for them but they are pretty good. Of course I prefer HP 6HR call to repair contracts where they have a binding obligation to have you up and running in 6 hours, not just have a tech onsite in 4 like most support.

Re:Sun shoots, and... well, you already know.

[jcnnghm]

Small Businesses are businesses that make under $25M/year by definition. I can imagine small businesses being in the market for inexpensive, high throughput, SANs.

Re:Sun shoots, and... well, you already know.

[dbIII]

3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

It really depends on what they are used for and where the data came from or can be recreated from. Sometimes you just need a lot of space and want just a bit more reliability than a stripe set but won't lose much sleep over losing two disks and the entire array. I still wouldn't do it because it would be a lot slower than a few more smaller drives making up the same array.

You only really need to mirror the stuff that is inconvenient to recover from backup. There are of course multiple levels of "production quality storage systems" depending on how they are used.

Re:Sun shoots, and... well, you already know.

[itzdandy]

With zfs and dtrace, you can build a nice small business storage box.

Keys are ECC ram, quality components from trusted vendors, and a good spindle count to catch bad drives.

Many SATA drives in a 16 drive array with a few hot spares and the ability to swap in drives as need running on Solaris with ZFS is a good setup. You dont need to spend $12,000 on this, you can spend half or 1/3 that in a small business environment.

Re:Sun shoots, and... well, you already know.

[Kent Recal]

Well, that's a nice little department file server that you specced out there.
Sun targets a *slightly* different market with their device (think: databases, mission critical, pink slips).

Re:Sun shoots, and... well, you already know.

[Score Whore]

Nice. How's the replication work on that rig you just built? And how many IOPS you getting? And how quickly will your vendor bring replacement hardware to you? How many filesystem snapshots can you take with your fancy ICH9 supporting linux? You gonna back that up over NDMP? How's the thin provisioning working out for you there? How much data you pushing through those two slots? Where's the other 2 gig ethernet ports. You got hot swappable power supplies there? After you're done stuffing all that gear onto your mobo, how many pcie slots you got left for future growth?

No offense, but try and get some clue as to what it takes to have a commodity class storage appliance.

Re:Sun shoots, and... well, you already know.

[dosguru]

This must have been what my Sun sales guy was talking about a month ago at lunch when he had a product that he described that was very similar to this, but he couldn't even give me the full details. I wanted to buy about a PB of then then and there, but we'll see what my other vendors come back with as counters. The new HDS arrays are changing to SAS from FC and have excellent virutalization on them. I've been pushing for 10K SATA2 drives to really shake up enterprise level storage. I'd love to have these mixed into the ~6PB of disk I help design and plan for.

IOPS and repose time are the keys in storage, not the actual cost or specific technology. Just like the space shuttle main computers, it doesn't matter if they cost a lot or are old. They have to be right 100% of the time and do what they need to do in the time allowed.

Re:Sun shoots, and... well, you already know.

[Sentry21]

Correct me if I'm wrong, but doesn't charging enterprise prices for simplified hardware that relies on commodity software solutions, kind of defeat the point?

The difference resolves itself when you pick up the phone, dial a number, and say 'It's broken, come fix it' and you're back up within a few hours and you've saved yourself a few hundred grand in downtime.

Re:Sun shoots, and... well, you already know.

[remmelt]

> RAID-5 [...] That's exactly what Google and many others do

If I recall correctly, they don't even do that. They just have three drives in each machine with a straight copy of the main drive on each. I can't find the source right now, but there was a PDF a while back.

Re:Sun shoots, and... well, you already know.

[TooMuchToDo]

As an ex-owner of a Mercedes SLK350, let me congratulate you on an excellent Slashdot car analogy. My Mercedes had "bling" factor, but for what I paid I could have afforded two reliable cars. I ended up handing it back when the lease ended and bought a Camry Hybrid.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Fuck me. That's classic.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Hmmmm HP is the only brand of server I've never had a mobo failure in, and I've delt with a lot. Dell is pretty good also, but IBM can't make a reliable server to save themselves.

I guess everyone has their own story.

Re:Sun shoots, and... well, you already know.

[this great guy]

The X4540 is even better to prove your point as it is cheaper than the X4500:
12TB (48x250GB) - $21,995.00
and is virtually identical to the new 7210 box (config with 48x250GB) that sells for $34,995.00. Therefore proving that the same hardware is sold at a 60% markup ! Someone mod the parent up, he laid out the perfect counter-argument to the GP.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Sun customers? Take a look at some IBM customers. My boss paid more than 15 times those amounts for an IBM storage system just a couple of years ago.

Re:Sun shoots, and... well, you already know.

[Fweeky]

the bottom line is you need redundancy - and that's pretty much it.

Um, well, good for you if that's all you need, and Sun will surely be happy to sell you something appropriate for that too, but for some of us, we kind of need that 2TB to do more than 300 IOPS.

Re:Sun shoots, and... well, you already know.

[ShakaUVM]

>>Production quality storage is expensive, and for good bloody reason.

Sure, they like their profit margins.

It really depends on your concept of "production-quality". I have a couple small businesses, and our servers have been running quite well using a RAID0+1 setup that's been going strong since 2004. It'd have been RAID5, but the mobo support for RAID5 is sketchy.

As someone who writes a check out of his own damn pocket to pay for every piece of equipment his company buys, I can tell you that there's nothing a $10,000+ appliance can do that would make me fork over that much money. I'd rather get a new car, thank you very much.

Re:Sun shoots, and... well, you already know.

[segedunum]

Sun targets a *slightly* different market with their device (think: databases, mission critical, pink slips).

Which is non-existent. Sun is trying to target small business with this which is laughable, and many running the sort of workloads you described switched to commodity hardware and storage years ago. To assume that they will then switch back is even more funny, but then, Sun is always game for a laugh.

Re:Sun shoots, and... well, you already know.

[segedunum]

You're trolling, right?

No, but Sun is with this overpriced bucket of bolts, right back to where we were with them a few years ago. They just don't get it.

For most companies, even "enterprises", hardware reliability gives you a better bang for the buck, because you can't bloody afford multiple data centers.

I wish people would stop using that stupid phrase 'bang for the buck'. Repeating it isn't going to make this value for money, as it's all about Sun working out ways of charging what they have always charged and pretending the last ten years never happened. This provides barely a whimper, let along a bang. The vast majority, and the people who Sun say they are targeting with this, don't need to spend money on this or on 'data centres' although when you get beyond a certain size you will need a data centre regardless. You're going to need a data centre to store these damn things when you have them though.

Re:Sun shoots, and... well, you already know.

[segedunum]

Sun's strategy: Over engineer the fuck out of a machine where people won't use most of what is in it, and use that as justification to charge an exorbitant price.

Re:Sun shoots, and... well, you already know.

[UnanimousCoward]

Dude, if you can really build such cheaper, reliable, fast storage systems, you're set for life since, as has been mentioned, many professionals rely on such. Call every photographer you can find in the "Yellow Pages" to start...

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

With the same level of assurance that the solution will operate, first time - every time?

With the same level of confidence that Some Vendor will bend over backwards to fix it if it doesn't work?

Will your solution be as well tested and engineered?

It's not like you can just grab 3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

Your "home brew" solution will not meet any of the objectives Sun are achieving with this product. Your spindle count will suck, so concurrent access will be slow. You will probably be limited to one of iSCSI, CIFS, NFS or WebDAV, I doubt your solution would have all - and if it does, the integration will suck.

Will your solution have the diagnostic tools that Sun can provide? Oh wait, you don't have the millions of dollars to invest in engineering quality diagnostics, right from disk analysis (Sector scanning, remapping, etc) through to performance related faults? Well, then your solution will suck. What about snap-cloning?

In short, yes - storage is cheap. You can grab large drives very cheaply and put together something that works. That does not mean it will be good. Production quality storage is expensive, and for good bloody reason.

As for doing this using SSD storage, that's just ridiculous. 2048GB of storage would be at least 16 128GB SSD disks - this is not counting any disks for redundancy (i.e, raid-5/6 parity), or hot-spares. Assuming 2 drives for RAID-6 parity and 2 hot swaps, you'd need 20 SSD disks - with 10 grand, you're expecting to pay $500 per disk - and no other hardware, i.e, motherboard, case, cooling (more important than you think), etc.

So, until you have a clue about designing production quality storage systems, please refrain from making statements you have no clue about, you're only serving to confuse those people who are actually interested in what this product has to offer them. Keep to building crappy 3 or 4 disk RAID-5 systems using extremely large drives for storing your music, movies and pr0n on, but don't ever ever ever ever think about using those in any situation where your financial livelihood depends on that data.

With the same level of assurance that the solution will operate, first time - every time?

With the same level of confidence that Some Vendor will bend over backwards to fix it if it doesn't work?

Will your solution be as well tested and engineered?

It's not like you can just grab 3 1TB SATA drives, throw them into RAID-5 and say that you've got 2TB of production ready storage. Well, you can, but you'd be an idiot.

Your "home brew" solution will not meet any of the objectives Sun are achieving with this product. Your spindle count will suck, so concurrent access will be slow. You will probably be limited to one of iSCSI, CIFS, NFS or WebDAV, I doubt your solution would have all - and if it does, the integration will suck.

Will your solution have the diagnostic tools that Sun can provide? Oh wait, you don't have the millions of dollars to invest in engineering quality diagnostics, right from disk analysis (Sector scanning, remapping, etc) through to performance related faults? Well, then your solution will suck. What about snap-cloning?

In short, yes - storage is cheap. You can grab large drives very cheaply and put together something that works. That does not mean it will be good. Production quality storage is expensive, and for good bloody reason.

As for doing this using SSD storage, that's just ridiculous. 2048GB of storage would be at least 16 128GB SSD disks - this is not counting any disks for redundancy (i.e, raid-5/6 parity), or hot-spares. Assuming 2 drives for RAID-6 parity and 2 hot swaps, you'd need 20 SSD disks - with 10 grand, you're expecting to pay $500 per disk - and no other hardware, i.e, motherboard, case, cooling (more important than you think), etc.

So, until you have a clue about designing pr

Re:Sun shoots, and... well, you already know.

[Matheus]

Yeah.. on a lark I just went to Dell's enterprise site and built the following machine:
2x Quad Core Xeon 3.16Ghz
Solaris :)
16GB Memory
6 400GB 10K SAS Drives in RAID5 (about 1.8T)
(the rest of the 2950III features)
$13422 (and I'm sure I could get a few grand lobbed off with my rep)

SO for a couple grand more than (or the same price as) Sun's 2T array I get a powerful server as well?

I don't see the benefit of features for price.. their solution would have to have MUCH better performance and reliability for me to choose that apple over this orange.

Re:Sun shoots, and... well, you already know.

[jabuzz]

That's because IBM rebadge some OEM stuff (mostly MSI from what I can make out)

That said their pSeries stuff is reliable, though very expensive.

Re:Sun shoots, and... well, you already know.

[jabuzz]

Even better take a X4500 and add some software to turn it into a VTL and watch the price go through the roof. UK academic pricing is like 18k GBP for a 48GB X4500, with VTL software it is 80k GBP...

Re:Sun shoots, and... well, you already know.

[jabuzz]

Thin provisioning is a nasty hack for Windows boxes, or something that does not have the equivalent of LVM.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Your ID is appropriate, score_whore.

The Sun appliance doesn't come with replication and one or quad port ethernet are cheap. rsync can sync and other more sophisticated options exist. In fact, if you rip these appliances apart, they will not be so different.

I suppose part of the issue is pulling teeth to get 100GB of storage from a enterprise storage array that management says is costing $10K when you know 1.5TB is $199. We often micromanage enterprise storage to a point that it costs far too much.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

That's exactly what Google and many others do, and they spend their money, and significantly less than this, on managing that storage effectively. It works.

Google does that for their various web apps. Because there is no "right" answer for search. They can lose data and it doesn't really matter. They also can afford to save pennies on hard drives and let the rest of the world suffer their needing to power up six times as many drives as would be necessary if they purchased higher quality equipment.

BTW, their payroll system doesn't run on GoogleFS. Their ERP system doesn't run on GoogleFS. They're accounting doesn't run on GoogleFS.

Re:Sun shoots, and... well, you already know.

[raptor21]

That's exactly what Google and many others do, and they spend their money, and significantly less than this, on managing that storage effectively. It works. When it boils down to it, you can have all the exorbitantly expensive and brilliant 'enterprise ready' tools you want but the bottom line is you need redundancy - and that's pretty much
it.

Google is not just any company. They build their own systems and don't just take off the shelf crap and use them anymore. If Google sold the hardware they made internally they could easily compete with Dell and Sun.

Go look at the Jobs @Google page and figure out why they need device driver developers and ASIC designers.

No body else does what Google does. They have 20K employees and a few thousand of those working on their Systems and Storage needs.

Most companies can't afford to hire so many people to run thier Data Centers.

Ahhh, shit. I'm heart broken. What I'd like to know is how a small business will handle a behemoth like that, how they'll fund the electricity for all those drives and who'll manage it all. I expect that will be an ongoing cost to Sun support ;-).

Is this supposed to be some logical follow up to a real question?

I have news for you. People have been doing it for years, and the reason why Sun's business has gone down the toilet to commodity servers, Linux and Windows, especially with small businesses, for the past ten years is exactly for this reason.

People have been buying EMC, NetAPP, HP and IBM's expensive hardware for years as well and in higher volumes than those that do homebrew solutions. What is your point there?

Re:Sun shoots, and... well, you already know.

[Kent Recal]

Well, last time I checked this market did exist and wasn't small either. It's the market of small and medium sized businesses that simply can't afford to run mission critical stuff on commodity hardware. Building a near equal storage device from commodity parts may be cheaper (albeit not as cheap as you think), maintaining and supporting it is not. Especially when you're talking multiple boxes of the same or similar ilk - ever tried to re-buy *identical* commodity parts after half a year? Ever tried to debug subtile differences in failure modes of different RAID controller BIOS revisions? Ever tried to set up coherent monitoring, retention and backup on top of such a commodity mix?

Like it or not, a large part of SUN's business consists of cleaning up the mess that people with your mindset leave behind. They jump in when your cheap'n'easy commodity setup falls apart or just won't scale anymore.

Re:Sun shoots, and... well, you already know.

[Archangel Michael]

Actually, the problem I had with the MOBOs were wide spread, and I wasn't the only one having them. I spent nearly a week on the first one with HP trying to diagnose the problem, all the while they knew, or should have known, what the problem was.

So, count yourself lucky as you escaped the problem.

Nice try though.

Re:Sun shoots, and... well, you already know.

[Anonymous Coward]

Have you seen a Nissan GTR?

On Slashdot, all car analogies (which obviously end up with 'you shouldn't buy BMW or Mercedes') think the only variable to consider in a car is how 'powerful' they are, which obviously I take to mean the bhp figure advertised.

And yes, they all look the same.

All other considerations, such as handling, ride, comfort, break feel, chassis dynamics, weight... are all thrown out.

Nice... only if engineering was that simple that you had one variable to play with...

Sometimes it's so frustrating to be a tech nerd and a car fanatic.

Re:Sun shoots, and... well, you already know.

[Znork]

I've worked in 'Enterprise' IT for fifteen years. Ten years ago it was nice to bring junked equipment home. These days I bring things I junk at home to work.

'Enterprise' hardware has gone from being expensive and significantly better to being just expensive. It's not the enterprise that's driving the state of the art anymore.

These days when I hear a mention of 'mission critical' I tend to file it in the same folder as 'enLarge your DATA'; sales pitches aimed at insecure managers. For the price you'll get swindled into paying for (some barely tested) 'high-end' products you can usually afford 10x redundancy. If not 100x (and I've seen worse than that).

So Sun is targeting a slightly different *wallet* with their device. That doesn't change the fact that most would get better performance/security/space/price by simply, like the GP suggested, throwing together commodity hardware.

Re:Sun shoots, and... well, you already know.

[Znork]

Sounds like you need to read up on your consumer hardware; it's not 1997 anymore. There are cheap modern consumer grade motherboards that come with 3xPCIe(2.0) 16x slots. That's a combined bus bandwidth of 24GB per second. Far more than enough to eat 48 SATA drives.

Re:Sun shoots, and... well, you already know.

[the_B0fh]

Oh great. Another moron who thinks a physical slot == a full bus. That's why have hacks around the damned IRQs - the original IBM PC had 8 physical slots too, but, you know what, it's damned funny how 8 IRQ signals wasn't enough.

I bet you think you will get 100% utilization on all the CPUs if you can stick 100 cpus in a box too, right?

Re:Sun shoots, and... well, you already know.

[pod]

lol, 60% after discount.

Re:Sun shoots, and... well, you already know.

[Znork]

40 PCIe 2.0 lanes (which are the cheapo chipsets capacity) are more than enough to drive 48 disks. You do realize that PCIe is comprised of switched serial lanes, right? There is no bus contention on PCIe lanes. Each lane is one 500MB/sec serial bus in it's own right.

You really should read up some on modern hardware. You sortof sound like you've been in hibernation for a decade. And woken up cranky ;).

Re:Sun shoots, and... well, you already know.

[the_B0fh]

Damn, you're right. Heh. I should know that. I claim being half asleep when I posted. Or something.

Re:Sun shoots, and... well, you already know.

[segedunum]

Is this supposed to be some logical follow up to a real question?

Errr, yer. Most businesses don't stay away at night thinking about spindle size, and managing those 48 disks is going to more than outweigh the benefits.

People have been buying EMC, NetAPP, HP and IBM's expensive hardware for years as well and in higher volumes than those that do homebrew solutions. What is your point there?

Sun isn't making it cheaper and is flying right over the heads of small businesses.

Re:Sun shoots, and... well, you already know.

[R Sync'r]

Yep, you can take Sun hardware, blast a Nexenta image on it, and have your Solaris / ZFS wrapped in Ubuntu. And if you want the UI and search and other things (most of which we have not used) you can get NexentaStor for a fraction of the 7xxx. But wait, you can also run it on non-Sun hardware and manage the packages all via APT irrespective of the hardware and Sun could have done all this themselves but did NOT. Why?

Re:Sun shoots, and... well, you already know.

[badkarmadayaccount]

These are luxury cars, they don't have any of that except for comfort. I'll take my (non-existent) Toyota sports car over your german glorified trailor any day.

Re:Sun shoots, and... well, you already know.

[badkarmadayaccount]

not troll. just say'n

Re:Sun shoots, and... well, you already know.

[Thundersnatch]

Yea, I recently built an 8T server. It cost me about $5000, and has no raid controller, and uses linux software raid.

And you suffer from the famous "RAID-5 write hole". Hello silent data corruption. And good luck during a rebuild if when you have a latent sector error.

If I really wanted, I could buy a second $5000 server and do DRBD between them to have 2x redundancy.

DRBD is not instant-failover clustering and load balancing, it is simply block-level replication. The vast majority of applications - especially databases - would undoubtedly corrupt data during a node failure on your proposed setup. If all you're storing is MP3s, fine, but if you have lots of apps banging away at your storage, Linux software RAID and DRBD don't cut it.

We have some real "enterprise-class" clustered storage systems built on Linux from LeftHand. In testing we've ripped the plugs on one of the five nodes (plain HP DL320S's rebadged) and every application just kept on trucking with no corruption or downtime. And 10K random 8K IOPS, 1/3 of which are writes. Nothing you can build yourself with Linux can currently match those capabilites, so we paid $100K for it.

Google gets away with cheap storage using very sophisticated software, and by controlling the entire application stack to tolerate storage and node failures. You can't buy Google's software, and even if you could you couldn't run your app of choice on BigTable without a lot of coding. But you can get similar capability by paying money to LeftHand, IBRIX, NetApp and other "enterprise" clustered storage vendors.

Looks great.. but

[Pengo]

$12k for 2TB of storage is hardly a small IT shop pricetag.

I'm not a OpenSolaris expert, but for under $600 you can build a 4 TB pooled ZFS server on relatively low cost hardware. Hell, even $1k you can get a nice rack mount with 4 1TB sata2 drives from Dell, throw opensolaris on it and your up and running.

With that said, linux REALLY needs ZFS , and not in userspace.

Re:Looks great.. but

[Unknown Relic]

While that may be true, you're hardly comparing apples to apples. The entry level 2TB model has 14 146GB 10K RPM SAS drives.

You'd still be able to whitebox it for a lot cheaper, but not 4x1TB SATA cheap.

It was also mentioned on the pre-announcement discussion that some people at Sun wanted to price it lower, but internally the powers that be didn't want their hardware to look "cheap". As such, prices went up. The good news is that supposedly the VARs will have some room to play with on the pricing. Not the most straight forward way to go on Sun's part, but the actual prices may be a fair bit lower at the end of the day. Hell, you can get it at 20% off from Sun directly through their try and buy program.

Re:Looks great.. but

[jcnnghm]

Will that $600 box be using 14 146 GB 10k RPM SAS disks?

These boxes aren't about providing stupid storage, their about providing massive I/O throughput. The larger boxes scale to 44TB and 576TB respectively. This also automatically moves frequently accessed data to flash drives (and RAM) for even faster I/O.

These are absolutely monstrous compared to anything you could build for $600. There seems to be quite a bit of custom hardware to power this setup.

Re:Looks great.. but

[JesseMcDonald]

With that said, linux REALLY needs ZFS , and not in userspace.

Due to deliberate licensing issues we won't have native ZFS in Linux any time soon. However, BtrFS should be merging into the mainline kernel soon enough (~2.6.29), and it includes most of ZFS's features plus a few of its own: storage pools, checksumming, mutable snapshots, built-in extent-level striping and mirroring, etc. It even supports in-place, reversible conversion from ext3 via a copy-on-write snapshot.

Re:Looks great.. but

[Fweeky]

Second level ARC is standard in recent ZFS; you could just plonk some X25-M's in your X4240, attach a disk shelf to it, configure ZFS to use the SSD's as secondary ARC for it, and pretty much have something like what Sun are selling.

You know, just with less vendor support, and more effort involved in building, configuring, tuning and testing. If you come out of it with change from $10k, you probably earned it with the effort you put in.

Re:Looks great.. but

[Anonymous Coward]

Checksumming is in ZFS. Some of the other features you mention are also in ZFS.

Re:Looks great.. but

[E-Lad]

Yes yes yes, you can do that with just $1000 and a afternoon at Fry's or browsing Newegg, right?

Everyone's missing the point here, and a lot of what is being said could be applied (just as wrongly) to NetApp... after all, those are just x86 boxes running a BSD kernel.

The special sauce here is not so much the underlying OpenSolaris OS (which does provid the IO and services such as CIFS, NFS, iSCSI, data replication, and so on) but the Fishworks software put on top of it. Built-in failover clustering, the integrated web GUI and CLI... if you weren't paying attention to the console during boot, you might not even have a clue that's it's OpenSolaris underneath... which is one of the marks of a good appliance OS... easy to manage and the idiosyncrasies of the underlying OS is sufficiently abstracted away.

You don't need to be a Solaris admin to use this, just like you don't need to know about BSD to run a NetApp. The difference here is that this takes pretty high-end x86 hardware and does better than NetApp, for cheap. Ever see a support contract for any of the NetApp filers? I guarantee you'll spin a 180 on your heals and pretend you never saw the number.

Re:Looks great.. but

[boner]

You should use a mix of SLC and MLC. MLC for the frequent read, infrequent write, SLC for the frequent write.

There is more underneath the covers than meets the eye.

Re:Looks great.. but

[Anonymous Coward]

Hmmm, BtrFS features?

storage pools - check, yes, they're in ZFS
checksumming - check, part of ZFS
mutable snapshots - check, can create clones of snapshots in ZFS that you can write to
built-in extent level stripinng & mirroring - check, in ZFS.

Please familiarise yourself with ZFS more before proclaiming that feature "X" is in BtrFS and not in ZFS.

I'll grant you that ZFS doesn't do the ext3 thing, but why would it want to ? :)

(someone please mod the parent appropriately... it's very very wrong.)

Re:Looks great.. but

[Anonymous Coward]

Tux3 also looks promising:
http://tux3.org/

Re:Looks great.. but

[JesseMcDonald]

Perhaps I wasn't completely clear. I am fully aware that ZFS supports all these features; I meant only that these ZFS features are also in BtrFS. BtrFS has some additional features not in ZFS, and visa-versa; there is a more extensive list on the BtrFS Wiki.

Among all the features of BtrFS the most significant is native Linux compatibility, which ZFS does not have and is not likely to acquire.

Re:Looks great.. but

[JesseMcDonald]

All of the features I mentioned are in ZFS. The GP wanted a filesystem with the functionality of ZFS, so those are the points I emphasized. BtrFS has other features which can be found on its homepage.

Re:Looks great.. but

[Anonymous Coward]

People will easily parse it as

it includes most of ZFS's features plus (a few of its own: storage pools, checksumming, mutable snapshots, built-in extent-level striping and mirroring, etc).

instead of

it includes (most of ZFS's features plus a few of its own): storage pools, checksumming, mutable snapshots, built-in extent-level striping and mirroring, etc.

Re:Looks great.. but

[this great guy]

Due to deliberate licensing issues we won't have native ZFS in Linux any time soon.

It's funny how this viewpoint is always the one promoted on slashdot. One could argue that the Linux GPL is the problem. FreeBSD and Mac OS X had no problem integrating ZFS into their code precisely because the ZFS license (CDDL) allowed it.

Re:Looks great.. but

[JesseMcDonald]

The Linux kernel has been GPLv2 for a long time. If Sun wanted ZFS support in Linux they could have made it work easily. The license they chose, and the patents they hold, allow them to call it "open source" while avoiding the possibility of native support in the most popular open-source alternative to their own Solaris operating system. I don't think that's a coincidence.

A much more detailed debate on the LKML is summarized on KernelTrap in support of my position. About a third of the way down the page you'll find this quote: "there are senior Sun programmers who have on record stated that one of the reasons why Sun picked the CDDL was precisely because it was incompatible with GPL and Sun fears Linux."

Re:Looks great.. but

[powerlord]

... The license they chose, and the patents they hold, allow them to call it "open source" while avoiding the possibility of native support in the most popular open-source alternative to their own Solaris operating system. I don't think that's a coincidence. ...

(emphesis mine)

I agree with what you're saying, but remember two things:

1) Its the license they chose. The developed the code, and they can choose how they license it.

2) I object to enclosing open source in quotes. The CDDL IS in fact Open Source.

The fact that something is incompatible with GPL (which is an aspect of the CDDL inherited from the MPL it was based on), does not mean its not Open Source, anymore than the fact that you can't freely use GPL code with proprietary code without opening up the proprietary code within certain guidelines means that GPL code is less open than BSD code.

Files licensed under the CDDL can be combined with files licensed under other licenses, whether open source or proprietary[1]. The Free Software Foundation considers it a free license incompatible with the GNU General Public License (GPL).[2] The incompatibility arises from a complex interaction of several clauses that the CDDL inherited from the MPL.[3] The CDDL was submitted for approval to the Open Source Initiative on December 1, 2004 and approved as an open source license in mid January 2005.

http://en.wikipedia.org/wiki/CDDL

There are different licenses with different goals.

Re:Looks great.. but

[JesseMcDonald]

This seems to be a day for misinterpretation. I did not mean to imply that the CDDL is not open source; I was simply setting the term off from the rest of the sentence, e.g. "someone says, 'ZFS is open source'" or, equivalently, "someone says that this project is 'open source.'" In other words, what quote marks used to mean before they were co-opted to indicate sarcasm.

Regardless of Sun's intentions, the effect is that they get the good PR associated with releasing code under an open-source license while keeping their most popular open-source competitor, Linux, from simply porting over the interesting bits. Most open-source project aim for technical and legal compatibility; thus the GPL and similar standard licenses. That Sun apparently does not share these goals bodes ill for those inclined to treat them as just another open-source contributor.

Re:Looks great.. but

[powerlord]

Misinterpret what you wrote. Its difficult to track the nuances of writing, especially in an internet age where multiple different conventions abound.

Technically they are still in competition, and if they see Linux as a competitor, so much the better. I am sure the GPL community will come up with an alternative to ZFS as soon as they can. The fact that BSD and OS X can incorporate ZFS though, does imply a certain amount of "legal compatibility" on the part of CDDL though. OS X is even a commercial product, that could arguably be considered a competitor against Sun's workstations.

True, I wouldn't treat them as just another contributor, but since they've based their license on MPL, I assume all the work of the Mozilla community should be similarly viewed?

Re:Looks great.. but

[powerlord]

Misinterpret what you wrote. ...

Somehow the beginning of my message got cut off (thats what I get for not hitting "preview"). Should have started:

"Sorry to misinterpret what you wrote."

Re:Looks great.. but

[JesseMcDonald]

Don't worry about the misinterpretation. In retrospect it's easy to see how you could draw that conclusion, just as it was the other two times it happened to me today. :-)

The fact that BSD and OS X can incorporate ZFS though, does imply a certain amount of "legal compatibility" on the part of CDDL though.

True. However, Linux is one of the most obvious targets for any code transplants from Solaris, and the one that presents the greatest threat to Sun. My main point is that Linux's use of the GPL isn't the problem; indications are that Sun didn't want Solaris features to end up in Linux, and chose their license accordingly.

... since they've based their license on MPL, I assume all the work of the Mozilla community should be similarly viewed?

Your example supports my argument. You'll note that the Mozilla code is all tri-licensed, MPL/GPL/LGPL, in the interests of maximizing compatibility with other open-source projects.

Sun replace RAID with RAID

[mollymoo]

What a stupid and misleading title. You can, and I suspect most people will, use RAID with these boxes. RAID-Z more than likely, though other types of RAID are possible too. It is not a RAID-less box, it's a box without a dedicated RAID controller.

Re:Sun replace RAID with RAID

[Purity Of Essence]

"Sun replace RAID with RAID"

No, they replaced it with "RAD"; they took the "I" right out of it.

Re:Sun replace RAID with RAID

[seriv]

I could easily be wildly off here, but I am pretty sure I spotted a raid controller in the picture of 7110. It was in the position of the raid controller on other sun 2U boxes, and it looked like a raid controller, so.....

Re:Sun replace RAID with RAID

[Namlak]

No, they replaced it with "RAD"; they took the "I" right out of it.

It's all fun and games until someone loses an "I".

How well does it integrate in a M$ environment?

[C3ntaur]

I remember Sun's 52xx NAS storage line was a non-starter for many because it didn't have a lot of the competition's (NTAP) features that made it Just Work with Active Directory, CIFS, etc. I wonder if this is still the case?

Re:How well does it integrate in a M$ environment?

[mistshadow]

It supports active directory, and user mapping between AD and LDAP. The CIFS stack is in-kernel.

No RAID?

[wonkavader]

"All of the new unified storage systems include comprehensive data services at no extra cost, Fowler said. These include snapshots/cloning, restores, mirroring, RAID-5, RAID-6, replication, active-active clustering, compression, thin provisioning, CIFS (Common Internet File System), NFS (Network File System), iSCSI, HTTP/FTP and WebDAV (Web-based Distributed Authoring and Versioning)."

Note that this system includes "RAID".

Re:No RAID?

[Anonymous Coward]

"All of the new unified storage systems include comprehensive data services at no extra cost, Fowler said. These include snapshots/cloning, restores, mirroring, RAID-5, RAID-6, replication, active-active clustering, compression, thin provisioning, CIFS (Common Internet File System), NFS (Network File System), iSCSI, HTTP/FTP and WebDAV (Web-based Distributed Authoring and Versioning)."

Note that this system includes "RAID".

(overheard in the Sun IT break room)

"You know that fucking clueless Marketing guy? Yeah, he asked me to write up something for the new RAID-free array. Heh, I hooked him up."

*Zetta?*

[Anonymous Coward]

So zetta slow!

Sun's storage strategy

[thanasakis]

Considering that they've purchased MySQL, StorageTec and Cluster File Systems (of Lustre fame), developed ZFS, implemented CIFS in OpenSolaris from scratch (not Samba based), participated in NFSv4 and constructed the thumper, these machines hardly come as a surprise.

For the last two years, almost all their moves are targeted towards one goal: Enter the storage market from a non-conventional angle. They want to do it unconventionally, because they know that storage more than anything else is becoming The commodity and today's toys won't cut it. Plus, at this point, all the mainstream storage vendors have difficulty tapping the low end. They may be able to sell their expensive products to clients with deep pockets, but for small businesses it's a different story. No to mention that they are unwilling to reinvent themselves. OTOH with all these inventions Sun may be trying to do what it did with workstations when it started in the 80s, start low and increase. Remains to be seen whether they can pull it.

Re:Sun's storage strategy

[Jeremy Allison - Sam]

Sun CIFS isn't reimplemented from scratch by Sun, it was code they got from their Procom acquisition. It remains to be seen if putting a CIFS server into an otherwise stable kernel is a good idea or not :-).

Jeremy.

Re:Sun's storage strategy

[nine-times]

Plus, at this point, all the mainstream storage vendors have difficulty tapping the low end. They may be able to sell their expensive products to clients with deep pockets, but for small businesses it's a different story.

This doesn't seem like the "low end" for small business to me. Someone up the page quoted that their cheapest model is $11k for 2TB. You should be able to get >10TB of disk space for that price.

I'm not trying to say that Sun is a bad value. You might get some really great features for all that extra money. I wouldn't know because it's not worth investigating at those prices. There's no way I could justify spending $11k for 2TB.

Re:Sun's storage strategy

[saleenS281]

I'm a huge fan of Sun, I'm on the zfs mailing list, I run it at home.

That said, I can buy an S-series from NetApp for under 10,000 with more storage and features, TODAY.

Re:Sun's storage strategy

[Anonymous Coward]

Show me the small business that is going to lay out even the most "affordable" 10k for storage in today's economy.

They'll go with cheap ass disks jury rigged together by the local tech expert, and for the most part will be happy.

As long as they're backing shit up regularly, why do they need this expensive mofo ass system?

Unless they're doing intensive I/O, this Sun system is overkill.

Like most everything Sun does, they cannot do low-end. It's not in their DNA.

Not RAID = unRAID

[Anonymous Coward]

Somebody needs to introduce them to unRAID ... parity protection without striping.

12TB for $2,000. (System plus 13 1TB drives (12 for data plus 1 for parity))

What Everyone is Missing

[jcnnghm]

This system will intelligently move the data around to put frequently accessed bits on the SSDs. This is a lot more than a 2u server with a few TB drives in a raid 10.

Re:What Everyone is Missing

[Anpheus]

What you're missing is that SSDs typically have that built in, and that ideally you want a file system that doesn't worry about large-scale contiguity because it results in unnecessary reads and writes.

Re:What Everyone is Missing

[Anonymous Coward]

This man knows what he is talking about. Listen people, if you can go build something larger, better, faster than what groups of engineers at Sun have quite possibly working on for a long time, then put your money where your mouth is! Either do that or get off Slashdot and go back to configuring your 10 node windows domain.

Congrats to Sun, Looks like a great product.

Re:What Everyone is Missing

[Anonymous Coward]

SSD's typically have the ability to copy frequently accessed data form other disks?

Re:What Everyone is Missing

[jcnnghm]

SSD drives can usually detect they are part of a cluster of traditional disks, and can request that data to be written to the traditional disks in the cluster be written to them instead, to improve I/O?

Re:What Everyone is Missing

[Anonymous Coward]

Hadoop is significantly more disruptive than Yet Another Storage box. It also scales larger and was written in less time than the 10 years it took Bonwick's team to build ZFS.

Re:What Everyone is Missing

[Anpheus]

I apologize if I jumped the gun, but that wasn't clear from his post.

Re:What Everyone is Missing

[boner]

Not according to the Intel X25-E specs...

Care to provide some proof? I don't believe for a second that an SSD drive can intelligently copy bits from a hard-drive. It would violate the way storage drivers work. The system does this, not the drive.

Re:What Everyone is Missing

[SiggyTheViking]

Read this.
Sun rocks.
Real engineering here.

Zettabyte?

[ethan0]

ZFS doesn't stand for zettabyte anything. "The name originally stood for "Zettabyte File System", but is now an orphan acronym." from wikipedia, sourced from http://blogs.sun.com/bonwick/entry/you_say_zeta_i_say .

and of course "RAID Array" is lovelily redundant phrasing.

Re:Zettabyte?

[Barny]

It could be a cheap, redundant, RAID array of drives :P

Re:Zettabyte?

[ShieldW0lf]

But zettabyte wasn't perfect, actually. We (we were a team by now) found that when you call it the zettabyte filesystem, you have to explain what a zettabyte is, and by then the elevator has reached the top floor and all people know is that you're doing large capacity. Which is true, but it's not the main point.

Should have just called it the Zinc File System. Then when people ask, tell em it's cause Zinc is good for you.

Re:Zettabyte?

[ethan0]

Then you have to explain why you're abbreviating Zinc as Z, not Zn. ZnFS doesn't have quite the same appeal, and even if you did that you'd have to explain why it's called Zinc Flouride Sulfide.

look at Sun x4500

[luguvalium2]

1] The filesystem is called ZFS not Zettabyte

2] They appear to be twice as expensive as storage solutions that sun already sells.

Compare:
  Sun Storage 7210, Option 3, $117,995, 44Tb

with
  Sun Fire X4500 Server, Config 4, $61,995, 48Tb
 

Re:look at Sun x4500

[BrainInAJar]

For data warehousing sure, but if performance matters as well as reliability look at the cost per IOPS and these things come out as damned cheap

Re:look at Sun x4500

[drsmithy]

While it certainly doesn't make up the ~$50k difference on its own, the 7210 _does_ come with 64G of RAM (vs 16G) and a pair of 18G SSDs. They're not completely identical.

Re:look at Sun x4500

[this great guy]

An additional 48GB of RAM and 18GB of SSD are not worth $50k. Seriously... Registered ECC DDR2 RAM is about $50/GB and SSD about $20/GB (gross overestimation): they are really worth 48*50 + 18*20 = $2760... Sun is charging 18x more !

Also the 7210 with 48x250GB is strictly equivalent to the X4540 with 48x250GB (same amount of RAM, no SSD), but is $13k more expensive... for the same box ! That is outrageous.

For the love of FSM...

[More_Cowbell]

People, please stop trying to compare a couple of drives from Newegg tossed in a chassis as a similar product for thousands less, simply because you have the same storage capacity.
That's not even apples and oranges, it's more like apples and redwoods.
Last I checked Netapp was still charging $10,000 per TB! Do you really think there is no reason for this?

Re:For the love of FSM...

[Anonymous Coward]

When Jonathan Schwartz talks about commodity economics, people are going to compare his prices against whitebox prices. He asked for it.

Re:For the love of FSM...

[QuasiEvil]

I hate to say it, but for the small business market, they should be compared. If you're selling a 2TB redundant storage device to a small business without a huge IT department, then you're competing against what can be built from commodity parts (aka, crap from Newegg + Linux + RAID) because often cost, not performance, is the defining factor.

Re:For the love of FSM...

[nabsltd]

Last I checked Netapp was still charging $10,000 per TB! Do you really think there is no reason for this?

Of course there's a reason for this...the CEO needs a new private jet.

Really, the profit in the storage arena is insane. Part of that because the marketing convinces management they need 50,000 IOPS to run what is essentially a flat-file database. Add in the "reliability" and "support" (which generally means that if you don't mind being told that a replacement part won't be in for a week and as long as you follow standard backup strategies, you probably won't actually lose any data), and it's pretty much a license to print money.

You really can do a lot of this stuff much cheaper, and even Sun know this. As other people have pointed out, Sun already sells much cheaper solutions to the exact same problem.

Re:For the love of FSM...

[this great guy]

Last I checked Netapp was still charging $10,000 per TB! Do you really think there is no reason for this?

Absolutely. The storage industry has always been a particularly profitable market precisely because buyers think these prices are justified. IMHO it's caused by a lack of competition in the high-end market (competiton is much more fierce in the entry-level and mid-range market). As someone pointed out, Google built a highly reliable platform and is certaintly not paying $10k/TB. Read about Google FS.

A friend of mine who works for a storage vendor that should remain nameless fully admits that they can sell some systems at 4-5x their OEM prices only because their customers, like you, think that "it must be worth it".

Look at the diff between config #2 ($35k, 32GB RAM) and #3 ($42k, 64GB RAM) of the Sun Thumper x4540 (48 drives in 4U): http://shop.sun.com/is-bin/INTERSHOP.enfinity/WFS/Sun_NorthAmerica-Sun_Store_US-Site/en_US/-/USD/ViewStandardCatalog-Browse?CategoryName=SF_X4540&CategoryDomainName=Sun_NorthAmerica-Sun_Store_US-SunCatalog Sun charges $7000 for an additional 32 GB of DDR2 registered ECC RAM, when in reality an increase from 32 GB to 64 GB should only costs 64 GB * $46/GB (price for 4GB modules) - 32 * $20/GB (price for 2GB modules) = $2304. So Sun is charging 3x the OEM price for a RAM upgrade. (Before you ask, I know the prices off the top of my head because I recently built a box with 32GB RAM).

Look at, as pointed out by another poster, the diff between a Thumper x4540 with 48 x 250GB SATA drives at $22k and a Sun Storage 7210 with 48 x 250GB SATA drives at $35k. This is a $13k price difference for virtually the same box. Beside even the x4540 is overpriced (another friend of mine bought one with the smallest drives and replaced them with 1TB from Dell to end up saving $20k to his employer).

So, no this practice of overpricing in the storage industry is not justified ! I have been saving tens of thousand of dollars by buying and building my own servers from newegg and literally "throwing a few drives in 2U chassis" (and throwing ZFS on top of it) and guess what ? It works pretty well ! Obviously not everybody can do it (don't build your servers if you don't know the diff between 1.8V and 2.0V RAM, or the effect of a 75W vs. 120W processor in a 1U chassis, etc), but if you can do it, you will save non-negligible amounts of money to your employer. This is important both for small startups who have to run as efficiently as possible, as well as Fortune500 like Google who have to scale efficiently. But sure, if you have the money and don't mind getting ripped off by storage vendors, go ahead, it's your money :) I don't deny they sell high-quality hardware, it's just unnecessarily expensive.

Re:For the love of FSM...

[lucifuge31337]

I just bought a 3TB S550 for $6400, including support. You can certainly pay more and get more redundancy/speed, but don't write off NetApp for the low end. So fat this thing is great. I'm using it as an iSCSI target for a VMWare LUN as well as NFS storage.

Re:For the love of FSM...

[NoBozo99]

Yep! Sun is aim this at customers and potential customers of NetApp. I'd be worried if I were NetApp.

Re:For the love of FSM...

[More_Cowbell]

I was not going to reply to any of the people that wrote responses like yours, but you obviously put a lot of effort into this, so here goes:

and guess what ? It works pretty well !

I could pick apart your individual points, but I do not really want to take the time, and I don't NEED to. This sentence shows me that you have no idea (no offense meant, really) what these types of storage devices are needed for. My current employer uses devices like like Netapp (and a few others in the same category) to back up thousands of client servers, 24/7/365. The backups literally run nonstop all day every day for clients that pay us ungodly amounts of money for this service.

YOU CAN NOT DO THIS WITH SHIT PASTED TOGETHER FROM PARTS AT NEWEGG.

The throughput, read/write speeds, redundancy, and lets face it, reputation are just not there. Is it overpriced? More than likely, but if your frankenbox dies, do you have a tech from Newegg on site within hours to bring it back online? 99.999% uptime is unacceptable; for our clients, for us. We are not alone. This is why they charge those prices, if you don't like it, go somewhere else- (as you obviously have) more power to you!

Re:For the love of FSM...

[this great guy]

YOU CAN NOT DO THIS WITH SHIT PASTED TOGETHER FROM PARTS AT NEWEGG.

Of course you can. The lack of technical details in your post proves you have no idea what it takes to build servers comparable to what storage vendors sell, and what is exactly the value that they offer when you buy these at over $1200/TB.

Take a concrete example: Thumper X4540. Sun use the same Opteron 2300 series processors you can buy on newegg. They use the same nVidia chipsets MCP55 and IO55 you find on many motherboards from newegg. They use the same LSI SAS 1068e chips you can find on the Supermicro card AOC-USAS-L8i recommended by Sun ZFS engineers themselves. They use the same enterprise-class SATA drives (Seagate IIRC) you can buy from newegg. They use DDR2 modules based on the same OEM RAM chips used by many memory vendors selling on newegg. And all the technology is the same: same PCI-E links, same 10GbE technology, same SATA links, etc. The only major custom-designed electronic component in the X4540 is the motherboard (mostly because they had to connect the 3 PCI-E slots to an IO55 chipset linked to an independent HT link because the sequential throughput of the 48 disks already use about 70% of the throughput of the two other 2000MT/s HT links: 48 * 120 MB/s / (2 * 4000 MB/s) = 0.72). But this custom motherboard is a pretty much unique requirement for the X4540 -- 99% of the storage servers on the market are far from being able to saturate 2 HT links as almost nobody else packs 48 drives in 4U.

The value of what Sun sells is twofold. They validate the platform as a whole: compatibility of the components, thermal/power characteristics, basic assessment of the failure rate of components, etc (something you may not know how to do, but system integrators such as me have been doing for years successfully). And they offer a bunch of "enterprise" features that are irrelevant to this discussion about performance and reliability of the I/O subsystem (mechanical design of the chassis, drive bays, IPMI and other monitoring features, etc).

This myth that storage vendors have access to "superior" technology has to stop. As always, I'll conclude by citing the example of Google that started by building their datacenters out of OEM parts with motherboards and drives literally velcro'd together. Perfect example of good system integrators hired to do the job in-house instead of buying from storage vendors...

Re:For the love of FSM...

[More_Cowbell]

I can't believe I'm responding to you again, as I usually avoid arguing with people on here. You are certainly entitled to your opinion, you just happen to be entirely missing my point.

The lack of technical details in your post proves you have no idea what it takes to build servers comparable to what storage vendors sell...

No, it proves it is a waste of my time to research and post them here.

As always, I'll conclude by citing the example of Google that started by building their datacenters out of OEM parts with motherboards and drives literally velcro'd together. Perfect example of good system integrators hired to do the job in-house instead of buying from storage vendors...

Dude. Just, dude. I will spend 30 seconds on this one for you, since you seem to be laboring under some large delusional mindset here.
http://wiki.answers.com/Q/How_many_servers_does_Google_have (no idea of the date: 700,000)
http://gotads.blogspot.com/2006/06/how-many-servers-does-google-have.html (6/27/2006 : 450,000 - 500,000)
http://www.techworld.com/green-it/features/index.cfm?featureid=3487 (June 26, 07: 500,000+)
http://news.cnet.com/8301-10784_3-9955184-7.html ..."clusters of 1,800 servers are pretty routine, if not exactly ho-hum. In each cluster's first year, it's typical that 1,000 individual machine failures will occur; thousands of hard drive failures will occur; one power distribution unit will fail, bringing down 500 to 1,000 machines for about 6 hours; 20 racks will fail, each time causing 40 to 80 machines to vanish from the network; 5 racks will "go wonky," with half their network packets missing in action; and the cluster will have to be rewired once, affecting 5 percent of the machines at any given moment over a 2-day span, Dean said. And there's about a 50 percent chance that the cluster will overheat, taking down most of the servers in less than 5 minutes and taking 1 to 2 days to recover."

So, give or take a FEW HUNDRED THOUSAND servers, you think this is a cost effective solution for a company with less than a $98.03 Billion Market cap?
Respond if you really must, but I'm really getting the feeling you are just Trolling, so I think I'm all set here.

Re:For the love of FSM...

[Anonymous Coward]

Ahem. Who is going to shell out this money during one of the biggest global recessions (possibly a global depression)? I'm sorry, but Google's Sergey Brin said it right (as quoted by Dr. Eric Schmidt a few weeks ago at Bloomberg News in NYC , especially during a recession: Scarcity Builds Clarity ... COTS will TRUMP Sun's new offerings for small businesses especially when CREDIT and CASH is hard to come by! Sun doesn't yet understand that economic times have changed (but maybe they will wake up if their stock price trends toward $0 like GM's).

Re:For the love of FSM...

[this great guy]

So, give or take a FEW HUNDRED THOUSAND servers, you think this is a cost effective solution for a company with less than a $98.03 Billion Market cap?

Yes: when they started, they had ZERO server and a non-existent market cap. It was cost effective for them to do so. And it still is.

no RAID or volume management?

[Anonymous Coward]

Actually...ZFS includes volume management...and while it's not conventional RAID, it still is RAID, just the better ZFS equivalents (i.e. RAID-Z instead of RAID-5).

RAID-Less how???

[mkcmkc]

• It doesn't use Redundant storage?
• The storage isn't an Array? (Meaning what? That it's composed of non-uniform parts?)
• It's not Inexpensive?
• It's not Disk-based?

The third one I believe--the rest I'm skeptical about...

Re:RAID-Less how???

[raynet]

Maybe instead of an array of disks it is a hash or maybe even a linked list.

Re:RAID-Less how???

[LotsOfPhil]

In this ($1000s/TB) case, the I stands for independent.

Re:RAID-Less how???

[TheRaven64]

It's redundant, it's an array, and it contains disks, but with 140GB SAS disks, inexpensive is definitely not an adjective that applies. It's a RAEP solution, which sounds like it's probably illegal.

Re:RAID-Less how???

[Anonymous Coward]

Umm. What? Care to comment a little more on that one?

Re:RAID-Less how???

[Culture20]

It's a RAEP solution, which sounds like it's probably illegal.

A RAEP-ier solution would be even worse. http://en.wikipedia.org/wiki/The_Xtacles

Re:RAID-Less how???

[Anonymous Coward]

Redundant Array of Expensive Disks.

Re:RAID-Less how???

[Anonymous Coward]

If you spent even one moment educating yourself instead of immediately posting your worthless opinion to slashdot you would realise that ZFS has raid capability built into it. Think linux mdadm that isn't a completely unreliable piece of junk.

Re:RAID-Less how???

[Anonymous Coward]

Inexpensive... Try independent!

Inexpensive these days is not an appellation I would apply to RAID. See Wikipedia.

Re:RAID-Less how???

[Ant P.]

Redundant Array of Erroneous Posts

Re:RAID-Less how???

[the_B0fh]

Oh great. Another revisionist. RAID is Inexpensive, quit fucking around with the terminology.

Re:RAID-Less how???

[Trogre]

Okay, I've still got a couple of points left. Just let me mod you up now.

Re:RAID-Less how???

[Trogre]

Crap.

Re:RAID-Less how???

[initialE]

The same way you can have computers around the world that are international business machines, but you can't call them IBM computers. RAID is a branding term as much as a redundancy system. If you're using something that's noticeably different, you need to say so.

Re:RAID-Less how???

[Anonymous Coward]

The "I" can also stand for "independent". For that matter, RAID-0 isn't Redundant, it's a scheme to distribute data to reduce the drives' apparent latency (transferring data with Drive A while waiting for a seek on Drive B).

Re:RAID-Less how???

[mkcmkc]

Not sure what you mean by "branding term", but your last sentence is my point: their difference is not apparent.

Re:RAID-Less how???

[the_B0fh]

To the mucking foron who modded me down - go read the original paper on it. But if you could read, you wouldn't have modded me down.

Re:RAID-Less how???

[thexile]

Data is stored on array of punch cards instead of disk.

Re:RAID-Less how???

[mkcmkc]

Never overestimate the bandwidth of a pickup truck full of punch cards. (unless it's moving at relativistic speeds)

DOWNLOAD THE STORAGE SIMULATOR!

[Anonymous Coward]

enough whining by people who really dont know what they are talking about (save those of us who do)

Check out the simulator that Sun built for you to try it in a VMWare instance:

https://cds.sun.com/is-bin/INTERSHOP.enfinity/WFS/CDS-CDS_SMI-Site/en_US/-/USD/ViewProductDetail-Start?ProductRef=SunStorageSim-1.0-OTH-G-F@CDS-CDS_SMI

Re:DOWNLOAD THE STORAGE SIMULATOR!

[nabsltd]

For anyone interested, when I registered to allow me to do the download, it took the Sun server 30 seconds from the time I hit the "Register" button until it returned the result page.

I guess they aren't using the bad-ass storage they are trying to sell.

The machine is filled with SSD, not hard drives.

[Anonymous Coward]

Did anyone bother in this entire thread to read the full article plus material?

The machines are filled with SSD, this is not about spinning drives.

Go back and rethink your arguments once you do the math.

Others call this "software RAID"...

[gweihir]

I have been unsing this on Linux for nearly a decade. Marketing again selling something old as "new"?

Re:Others call this "software RAID"...

[Anonymous Coward]

you have not been using this for nearly a decade. Come back to the discussion when you have a clue about what Sun has released.

Re:Others call this "software RAID"...

[the_B0fh]

I'm not sure which part of ZFS is considered "old". Glad RAID is working out for you. Be happy that you have not hit any of the issues that RAID has.

But we're in the terabyte size drive age now. If all you can do is raid, your data is going to go bye bye. You need ZFS. Go google for the intro paper they wrote on why use ZFS.

RAID-less? Not so! RTFA!

[DragonWriter]

FTFA:

All of the new unified storage systems include comprehensive data services at no extra cost, Fowler said. These include snapshots/cloning, restores, mirroring, optional RAID-5, optional RAID-6, replication, active-active clustering, compression, thin provisioning, CIFS (Common Internet File System), NFS (Network File System), iSCSI, HTTP/FTP and WebDAV (Web-based Distributed Authoring and Versioning).

So, these RAID-less devices all include optional RAID-5 and optional RAID-6?

Putting the RAID as part of the fully integrated hardware-software black box may be a convenience, and the particular way it is implemented may save money, configuration, reduce failure points, or provide some other benefit, but what it doesn't do is make the box "RAID-less" in any reasonable sense.

DL180/185

[paugq]

Four easy steps to dead-cheap SAN/NAS storage:

1. Buy HP DL180 or DL185 servers, with a Smart Array P800 RAID card
2. Buy 12 to 14 1TB or even 1.5TB hard disks from Seagate and trays in eBay (I've bought in the past from SCSITray and there are other sellers)
3. Install Solaris 10 or Nexenta OS, set up ZFS
4. Sun goes bankrupt

Re:DL180/185

[rainer_d]

This is BS. Clearly.
You have certainly never done this yourself.
First of all, the P800 is a PoS for anything but the included RAID5 or 6 (we haven't even tested RAID6, IIRC).
It has a maximum number of logical disks it can create and you will most likely have to reboot the server and go into Array Manager to setup another "array" (single disk). You can't use the RAID on the card, because ZFS wants to control the disks themself, without a RAID-controller in between (and ideally no Cache-RAM).
My co-worker's been there, done that, got the t-shirt.
So, you've got to buy one of these: http://www.sun.com/storagetek/storage_networking/hba/sas/specs.xml and connect a good external JBOD chassis to it (MSA70 e.g.). For SATA, we use Promise VtrakJ610s (which are not good, but cheap...).
Then it works.
But you will miss some features, like the red or yellow light when a disk is dead (so you'll have to count...).
And of course, you also don't get all the integration-work SUN has done with their new filers, all the statistics, all the health-checks, the GUIs.
In the end, you end-up a bit cheaper, but with a lot more labour and no support and no warrantee from anybody (best-effort only support from HP and SUN for Solaris on the DLxxx,)

Re:DL180/185

[paugq]

I've done with the DL320s, P400 and Linux, as Solaris did not work on the DL320s at the moment. It does certainly work.

Re:DL180/185

[rainer_d]

Linux does not have ZFS.

Re:DL180/185

[level4]

You have certainly never done this yourself.

Or maybe he's still on his 1st or 2nd home-built RAID. In my experience techs tend to build maybe 3 or 4, and be around a while to see the problems, before they say "fuck this" and buy from a vendor.

Home built RAID is great for home use and as a learning experience. Personally, I'd go with a Enhance UltraStor RS16-JS SAS and an Areca ARC-1680ix, but that's just me and because I use macs at home. And I don't really mind it taking 16 hours to rebuild. Oh no, I can't stream my movies at full speed today. Try telling your boss that your ecommerce site will be 10x slower and timing out all over the place for the next day or two at christmastime but that's cool because you saved a few grand last year building the storage array yourself and see the delighted look on his face.

You would have to be nuts to try and pull the "save money by building it myself!" trick in a heavy production environment. You might get away with it for a while, maybe with one or two units. What happens when you have 12? 24? 48? Practically a full time job just babysitting the damn things. And what is your strategy for when, not if, but when the power fails?

I'm actually pretty interested in these Suns. The price isn't *that* bad when you consider that ZFS, in theory, doesn't have to rebuild and should get itself into an inconsistent state less. Those two features alone are pretty much worth the markup IMO. And furthermore, everyone knows Sun RRPs are just there to make the discounts they then offer you look better. No-one pays more than 50-80% of that if you buy more than a couple of things at a time and you have a decent VAR.

Re:DL180/185

[Wesley Felter]

When you lose a disk ZFS has to rebuild and unless your system was over-engineered to begin with, there will be performance degradation. But now there's a pretty graph showing you how big the degradation is.

Re:DL180/185

[SuperQ]

I did a benchmark comparison of the HP P400 in a DL185 vs a simple LSI 1068E SAS jbod controller. (HP branded of course) HP said it wouldn't work with the 1068E in the 12x DL185 expander setup, but of course it only took me about 10min of looking at the LSI website to find the Initiator Target firmware and get it fixed.

Basically my benchmark showed that in all cases the JBOD setup with linux software raid was 10-50% faster than the P400 controller.

They didn't ship a BBU for the P400 so it wouldn't do any write caching in the RAID5 test. The write speed was 2% the speed of linux software RAID, so I basically ignored that test.

Eventually I gave up on the DL158 and bought a Supermicro based box (SC836 chassis, great box) for a bit less, and a lot less HP bullshit.

Re:DL180/185

[the_B0fh]

It does, using Fuse. If OP used ZFS via Fuse for production boxes, he's a bigger moron than I thought possible.

Re:DL180/185

[the_B0fh]

And furthermore, everyone knows Sun RRPs are just there to make the discounts they then offer you look better. No-one pays more than 50-80% of that if you buy more than a couple of things at a time and you have a decent VAR.

Oh man, you haven't seen IBM work that game. There's retail. Then there's preferred which is about 1/2 X. Then there's the "just because we love you discount" bringing things down to about 2/5 X. Still above reasonable, but, hey, senior management never got fired for buying IBM right?

Then, next year, you find out what the annual support and maintenance cost. 20%. Of the freaking *RETAIL* price.

Re:DL180/185

[level4]

Guh. Sorry. I'm tired, and re-reading my comment the english is well-formed but the concepts are jumbled nonsense. Let me try again, by your leave...

Yes, it's unavoidable to rebuild when you lose a disk, and there will be a performance hit unless you go for full on 100% redundancy, and not many companies can afford to do that with a lot of data.

ZFS offers a number of benefits, though, in the event of drive failure-triggered rebuild, in that it basically knows where the data is and only bothers with that. A hardware controller has no idea what's data and what is blank space and so just redoes everything. In theory, assuming the MB/s of rebuild is the same, a ZFS rebuild of a half-full array should take half the time of a traditional controller.

It is also much more intelligent about *what* it rebuilds, starting at the top and then descending down the FS tree, marking it as known good along the way. This means that if a second drive fails halfway through the resync, instead of a catastrophic failure you still have the data up to the point of failure.

I can't remember where I read that; maybe here: http://blogs.sun.com/bonwick/entry/smokin_mirrors

But I didn't even want to talk about drive-failure rebuilding, what I actually wanted to say that ZFS is, in theory, less likely to get itself into an inconsistent state in the case of power fluctuations, controller RAM failures, drive failures w/ pending writes, that kind of thing. That's the kind of rebuild I meant - after some kind of catastrophic failure. I should probably have said "integrity checking" though.

By design, ZFS never holds critical data in memory only and so at least in theory should always be consistent on-disk. Basically it shouldn't need to fsck. That is a giant advantage to me, if it turns out to be as good in reality as it sounds on paper. Of course, that also has a lot to do with the capabilities of the FS proper, but removing the evil, evil HW controllers from the picture can only be a plus.

I don't know why, but RAID controllers are the most unreliable pieces of hardware I have ever known, besides the drives themselves (but at least they are consistent and expected to fail). Get a few of them together and something WILL go wrong, more often than not in a horrible and unexpected way. When some RAM goes bad in a HW RAID controller you are in for a whole lot of subtle, silent-error-prone fun. Anything that gets the HW controllers out of the picture is a win for me.

And don't even mention the batteries in HW raid controllers. They are the wrong solution to the power failure problem, especially since it's always after a failure that a disk will decide it's had enough of spinning and would just like to sit still for a while, thank you very much. Drive failure with pending writes! Exactly the words every administrator wants to hear. Almost as good as power failure with pending writes. Combine the two (highly likely!) for maximum LULZ. Ok, this is turning into a rant, I better stop.

Anyway, thanks for the corrections. My original comment (and probably this one) came across as a confused mess upon re-reading .. sorry .. will sleep now : )

Re:DL180/185

[paugq]

I'm using ext3, of course. ZFS on FUSE is nowhere near production quality. The original idea was to use Zumastor but it is not production-ready, either.

Re:DL180/185

[Princeofcups]

Guh. Sorry. I'm tired, and re-reading my comment the english is well-formed but the concepts are jumbled nonsense. Let me try again, by your leave...

Yes, it's unavoidable to rebuild when you lose a disk, and there will be a performance hit unless you go for full on 100% redundancy, and not many companies can afford to do that with a lot of data.

Actually, ALL the companies that I've worked for (from startup to Fortune 100) do full 100% redundancy (RAID 0+1) for all their filesystems (on Sun Solaris). Not some, ALL. OS, straight mirrored. Application space, straight mirrored. Database, straight mirrored. Disk is cheap compared to software cost. Have you priced Oracle Enterprise lately?

Re:DL180/185

[paugq]

ext3 in RAID-6, BTW. But with the configuration I posted in the first post of this thread, you can build a very cheap SAN/NAS. It does work. You can get support from Sun for the software (Sun sells Solaris support contracts) and HP for the hardware. Obviously, you won't have the nice GUI, but it's TEN times cheaper per TB.

Re:DL180/185

[level4]

Ah, yes, I was mainly thinking about non-mission-critical data, for example vast amounts of user-uploaded data for web sites.

You would have to be utterly crazy not to guarantee full redundancy on, say, a user database or business documents. However, it's quite a different matter to guarantee full 100% redundancy for, say, a few hundred TBs of user photos and videos. When you are offering a free service, it's difficult to make a business case for an incredibly expensive full-redundancy setup just to rule out an unlikely event which would maybe annoy a tiny subset of your non-paying users.

For example, I am not privy to Google's internal workings, but I very much doubt they have guaranteed full redundancy for every single video that has ever been uploaded to YouTube. Admittedly, they don't use RAID, they use a custom FS, but the principle is the same. The cost of absolutely guaranteeing so much (mostly low-value) data would be incredible, and I can't believe they would do it.

I've studied the systems of high-load social networks like Mixi and LiveJournal, and unless something has changed, they do not do it. I can't imagine Wikipedia has full redundancy on its images, or RapidShare on its user files, or Flickr, etc etc. Hell, there was an "incident" earlier this year when darling-of-the-blogosphere VC-funded Joyent, ironically using ZFS, were forced to admit they did not have redundant storage for data uploaded into not one but two of their *paid* online storage products. Something went wrong, the service was down for a week while they sorted it out, and they then decided to pull the product from the market rather than move to full redundancy since it would be too expensive. And that's when the customers were paying them!

http://www.joyeur.com/2008/01/22/bingodisk-and-strongspace-what-happened

So, it's not uncommon at all. I would actually be pretty surprised if any large percentage of the huge amount of bulk data uploaded to free services around the web was stored with the "enterprise grade" 100% redundancy you're talking about.

Databases and business documents, though, hell yes : )

Re:DL180/185

[the_B0fh]

Nobody is saying it doesn't work. What you are building, and what Sun built are two entirely different classes of equipment. If you don't understand that, and still believe your Yugo can outperform a Lexus, more power to you.

Re:DL180/185

[Wesley Felter]

For example, I am not privy to Google's internal workings, but I very much doubt they have guaranteed full redundancy for every single video that has ever been uploaded to YouTube. Admittedly, they don't use RAID, they use a custom FS, but the principle is the same.

You're right... Google actually uses triple redundancy, and that's cheaper than "enterprise-grade" anything.

Re:DL180/185

[paugq]

I'm not saying my current setup (DL320s, SA P400, RAID-6, Linux, ext3) can outperform, or even be on-par, with the RAID-less storage appliances Sun introduced. I'm saying a setup with DL180/DL185, Solaris and ZFS can.

Re:DL180/185

[paugq]

They do ship a BBU for the P400. Part number is 383280-B21

Re:DL180/185

[level4]

That's a pretty extraordinary claim. I think a citation is needed.

I find it very difficult to believe that Google immediately and permanently makes a triple-redundant (not just local redundancy, separate-system redundancy) copy of every single byte ever uploaded to Youtube.

Their filesystem is highly locally redundant in itself, superficially comparable to RAID-6 or better. But you're asking me to believe that they then have another, and then another, full copy of that entire installation?

I don't think so, but I'd be interested to be proven wrong.

Re:DL180/185

[Wesley Felter]

I don't know what you mean by local redundancy or separate-system redundancy, so I can't answer it. The Google File System paper says "By default, we store three replicas". Each replica is on a different server, although they are probably all in the same datacenter. This is a higher level of redundancy than RAID 10.

Re:DL180/185

[level4]

Local redundancy = some protection built into a single system to protect against failing drives. Like RAID 5 or 6, or GFS storing copies of data on various servers in the cluster.

Seperate system redundancy = having a full failover redundant machine/cluster in case the first one falls over.

Not exactly standard terms I know but I suspected our difference was hinging on the definition of "redundancy" so wanted to be more specific.

The Google File System paper says "By default, we store three replicas". Each replica is on a different server, although they are probably all in the same datacenter.

Yeah, that's pretty much what I thought? I thought you were claiming they had three DCs mirroring each other or something crazy.

This is a higher level of redundancy than RAID 10.

I see the problem. We are talking about different things.

RAID10 is only locally redundant, ie, inside the server/cluster. If the RAID controller screws up or the server blows up, you lose. That is not what I mean by redundancy. I would call that "fault tolerant", not redundant.

What I mean by redundancy is having a hot replica next to the first server to failover to. Or, even better, in a different location, on different power.

Google's system is not directly comparable because although it's local redundancy they mean when they say "three replicas", it's still divided up between different hardware, and it's hardware failures of varying kinds that we are trying to guard against here. So they do have hardware redundancy, which is excellent and way better than RAID10. However, if the cluster fails - not that I can imagine that happening, knowing Google - there is no higher level redundancy to switch to. Not that they need it; works well for them.

Anyway hope I straightened that out and am making sense ...

oh ok...

[phaetonic]

Fortune 500 companies typically standardize hardware, so people who say they can buy this from here, that from there, one more thing from eBay are rediculous.

Also, to those who say small businesses can't afford this, its really an option. Some places like open source hodgepodges of hardware and some do not because their small business generates enough money that investing in enterprise class hardware with gold 4 hour response from a solid company with a history of UNIX experience and integration with Solaris.

Also, said Fortune 500 companies get massive discounts, as what you're seeing is retail price.

Re:The machine is filled with SSD, not hard drives

[Anonymous Coward]

Mr AC, did you bother to read the entire story plus the information that Sun publishes?

These machines are filled with spinning disks. The base models have no SSD at all, while the higher-end models contain varying amounts of SSD to use as a high-speed cache.

The minimum price for a mere 18GB of SSD? $71,995.00.

(You can fall off your chair now.)

Go back and rethink your arguments once you do the math.

HDS/IBM already doing this

[Bardwick]

Want to see something cool, check out IBM XIV. I will grant you that unless your in Israel, there is no customer visits in your future... The "E" guy from EMC went off on his own and built it (IBM bought him). He's been credited for EMC cache algorithms which pretty much put the DMX in front of everyone else. Obligitory droids comment. This is not new anymore than Microsoft inovating.

Layered speed storage

[TurtleBlue]

From the preso I saw - it sounds like the goal is to have varying speed storage backing and some rather sophisticated "caching strategies" in front of it - i.e.

- Some amount of RAM for frequently active/accessed files
- Some amount of SSD drives for Level 2 data access, faster, but not an entire array (too prohibitively expensive)
- A bunch of spindle drives behind the whole thing for "slow" data.

It still follows the principal of more spindles for faster seek times, but a layered approach makes sense and I'm surprised I haven't seen more companies go towards it.

The techs added in (dtrace, zfs, etc) are there for better tracking of hot files, snapshots online, etc. All wrapped in a GUI front-end for the single-Admin company.

As far as the market - I guess it's who or how you define a "small" business. I'd say probably one large enough to run Oracle for some record keeping would fit small in their playbook, as this seems clearly aimed at a market that needs faster access storage. "Small" at that level seems to become semantics - someone with 50 employees but tons of warehousing comes to mind (think Dunder Mifflin if that helps)

Hate to be the one to tell you all this...

[Giant Electronic Bra]

But SUN is FAR from being the inventor of charging people $50k for something they could just as well get for free...

Name ANY big IT vendor, they all do it. My father can tell some amazing stories on that subject. Not a new phenomenon either.

Now, if you are the GOVERNMENT, they'll give you the special bonus public sector price, $150k!!!

Sounds like apples to oranges comparison

[Giant Electronic Bra]

to me. Coming from high performance transaction processing land where an operation means 'the data is ON the platter' you can't do that more often than the platter rotates to the point where the head is over the sector where the write operation starts. Basic math, 15k RPM spindle = roughly 300 times/sec. Multiply by however many spindles you got, that's what you're max throughput is.

This is one reason why IN THEORY at least an SSD would be so great, that latency is much less. So basically I'm thinking they just aren't talking about what you're talking about, and maybe that makes sense, if you're running a trading operation say, you just DO NOT CARE what is buffered someplace, if it isn't physically on the drive, it doesn't exist.

Re:Sounds like apples to oranges comparison

[markhahn]

saying 15krpm = 300 iops is indeed traditional, but reflects the conventional filesystem layout. if you have an empty track, you can log at whatever your disk bandwidth is - that is, precisely _not_ wait for a particular spot to come around again.

Re:Sounds like apples to oranges comparison

[Giant Electronic Bra]

Hmmm, well, my knowledge of the low level details of disk drive geometry is certainly WAY dated. In the 'good old days' your drive controller still had to wait for the 'start of track' to come around, at which point you could write. So in effect every IOP had a built in latency on average equal to 1/2 a disk rotation. That really had nothing to do with file system layout, it was just a hard requirement dictated by the drive controller technology.

SUN designs and sells products

[Anonymous Coward]

They try and keep their shareholders happy.

Linux, on the other hand, is going to take over the world.

The desktop will come after that.

this is not competing with the diy market

[asv108]

So far the comments on this thread consist of "I could hack together some system for x% of the Sun price."

The goal of this product is to compete with Netapp. If you've ever experienced Netapp licensing/pricing, this Sun solution is a bargain. People seem to be forgetting that this is a storage appliance.

You're missing the point

[Anonymous Coward]

This is meant to be 100x faster than the storage you're talking about:
First: This uses Hybrid Storage Pool:
The Hybrid Storage Pool combines DRAM, SSDs, and HDDs in the same system, dramatically reducing bottlenecks and providing breakthrough speed.
Second: The system's hybrid architecture gives you the speed and performance you need to shatter the I/O bottlenecks with no administrator intervention. In fact, Hybrid Storage Pools with SSDs can improve I/O performance by 100x compared to mechanical disk drives.

You're missing the point

[Anonymous Coward]

Also, the capacity of these storage servers are much more from the simple storage that you can build:
Sun Storage 7110: 2TB
Sun Storage 7210: 44TB
Sun Storage 7410: 576TB
See this video
Are you going to build your own 576TB disk now?!

Missing option

[Anonymous Coward]

D) Freakin' awesome.

small business

[Anonymous Coward]

Small Businesses are businesses that make under $25M/year by definition.

the company I work at (multinational semiconductor business) has just been degraded to "small business" by you.

You insensitive clod.

(AC, for obvious reasons)

Storage startup idea

[this great guy]

You know your "x% of the Sun price" comment makes me think that no storage vendor offers anything between the $120/TB and $750/TB range. Disks sell for $120/TB raw; DIY solutions (counting the PC hardware) can be doable around $200-250/TB; and the less expensive offerings from storage vendors always sell above the $750/TB mark. Something tells me a storage startup targetting the $300-500/TB range would be very, very successful...

too bad, seems you need a 7410 for clustering

[Lazy Jones]

so for "small" SAN/NAS usage with replication/clustering to tolerate 1 completely dead appliance you need deep pockets ... or go back to 2 Linux boxes with drbd. Correct me if I'm wrong.

Re:too bad, seems you need a 7410 for clustering

[LarsG]

or go back to 2 Linux boxes with drbd.

Oooh, neat! Could you comment on how well this works in real life? That is, how is performance, how well does it scale and how resilient is it?

Re:too bad, seems you need a 7410 for clustering

[Lazy Jones]

Oooh, neat! Could you comment on how well this works in real life? That is, how is performance, how well does it scale and how resilient is it?

I wish I could. I have not yet built such a setup and thus am shopping around for something comparable (which is why I read this article). In theory, you can do many neat things with drbd esp. together with Xen, have a look at this thread for example (3rd post).

Basically you can build fully redundant iSCSI SANs and have Xen VMs use these while being able to migrate "live" between the Xen machines. So you could lose any of your boxes and still have all your VMs running again in a very short time. All components are cheap PC boxes, so all your worries about SPOFs or expensive legacy hardware that is hard to replace (e.g. that oh so reliable SAN box) are gone. Once you figure it out. ;-)

"Unconventional?"

[pedantic bore]

It's not "unconventional" to mimic everything NetApp does five or ten years later.

Stop Messing With My Back Button

[neiljt]

Grr, when I visit sites like eweek.com, I really wish I had an "anti-bookmark" system so that if I ever attempt to go back there a dialog will pop up which says "These idiots screwed with the functionality of your back button last time you visited -- are you *sure* you really want to go there again?", and which defaults to "Heck, No".

Wankers.

Yup, the article title was misleading/stupid

[davecb]

ZFS does what I consider raid 10, 5 and 6.

--dave

It's aimed towards a very specific market

[Ideally Nowhere]

There's merit to criticizing Sun's prices, but this box is aimed towards smaller shops that with high throughput and large volumes of critical data. There are (for now, anyway) many small shops that offer specialized services to financial companies. And even with all of 10-30 employees, they easily wade through several gigs of data per day. Not only that, but in order to get contracts they need to demonstrate that they have top-to-bottom support on hand; something Sun does well (albeit at an exhorbitant price). Otherwise, they will face stiff lawsuits if their downtime costs their customers, for less-than-due diligence. This is why comparing to Google doesn't quite cut it and why monolithic companies like IBM still offer consulting services (at obnoxious prices). No doubt, Sun really needs some serious introspection if they want to remain a player but there's still a need.

Performance Issues with Windows/MS SQL

[Anonymous Coward]

We recently tried using an x4540 with ZFS arrays to host databases controlled by a windows server running SQL Server 2005. The connection was done through the COMSTAR package in Solaris Express over fibre channel, but in the end the performance was attrocious (only 25% as fast as our standard 320 SCSI JBODS running RAID5). It turns out that ZFS doesn't handle the random read/write activity of most database servers efficiently, and as a result both sequential and random queries tend to grind along extremely slow.

There are blogs and wiki notes on how to configure ZFS for databases, but the fact is ZFS needs more development and testing before it will be able to compete with hardware RAID for database storage. Keep this in mind if you are thinking of getting one of these new shiny Sun boxes for hosting db's.

Fakeraid

[sethstorm]

Because it uses the Zettabyte File System, the Amber Road has eliminated the use of RAID arrays, RAID controllers and volume management software -- meaning that it's very fast and easy to use."

A new spin on fakeraid.

RAID-Z vs. RAID-5

[Brett Diamond]

It is important to note that Sun is selling a RAID-Z solution, which is far superior to a simple RAID-5. It offers data integrity, a uniform simple interface, and extra features (snapshots, cloning, etc.) that are not part of the RAID-5 specification. Plus, it doesn't have the RAID-5 write hole.

RAID-Z (as well as it's other flavors, e.g., RAID-Z2) is not just a way to arrange disks to be more reliable and/or provide better throughput, but is an advanced file system. This means that RAID-Z offers features like compression, privileges, quotas, etc. that are above and beyond the RAID-5 specification. This is a good thing or a bad thing, depending on your requirements (probably a good thing, 'though).

One of the advantages of having a ZFS-based RAID in this type of configuration is the ZFS file system is transaction-based and so performs better in a network configuration, where local caching of data can corrupt file systems without like features (see http://blogs.sun.com/roch/entry/nfs_and_zfs_a_fine).

RAID-Z will require more write I/O and CPU than a hardware-based RAID solution, but it is possible that Sun has create such hardware in their (quite expensive) solution. At the same time, if you are creating a dedicated network storage device rather than sharing the hardware with other activities, you'll never notice the extra overhead. And if you don't want to buy Sun's hardware, they have given the RAID software away, so you can build your own at no (additional) cost.

Definition of 'small business'

[Kadin2048]

I think it should be pointed out that the definition isn't that simple. Here's a table (PDF) of how the U.S. Government defines a "small business." It varies by sector and sub-sector (a 'small' peanut farm is defined differently from a 'small' aerospace parts manufacturer, since the latter is significantly more capital intensive than the former).

Most IT-related activities are at the higher end of the spectrum (see p.30 in the PDF) which tops out at $25 million/year ("services") or 150 employees ("value added reseller"), but there are some odd special cases in there. "Technical consulting," for instance, is $7M, and "Engineering services" is only $4.5M, but "Custom Computer Programming Services" is $25M. Makes you wonder who that was gerrymandered for...

Re:Definition of 'small business'

[jcnnghm]

You're correct, I was oversimplifying to make a point. The point was that most people significantly underestimate what a small business actually is. For example, an Aircraft Services company with 1,450 employees could realistically need multiple SANs, and is indeed a small company.

Most IT-related activities are at the higher end of the spectrum (see p.30 in the PDF) which tops out at $25 million/year ("services") or 150 employees ("value added reseller"), but there are some odd special cases in there. "Technical consulting," for instance, is $7M, and "Engineering services" is only $4.5M, but "Custom Computer Programming Services" is $25M. Makes you wonder who that was gerrymandered for...

The small business size classification is generally based on the average size of companies in a field. I have no problem believing that the average Custom Computer Programming company is larger than the average Tech Consulting company, which is larger than the average Engineering Services company. When Obama was talking about raising the taxes of Small Businesses making over $250k/year, I doubt many people understood how many businesses that actually affects.

Define small business?

[Anonymous Coward]

Because I am sure what Sun defines as small business is nowhere near what YOU consider small business.

Sun defines small business between 500-1000 employees. If you are whiteboxing storage for 1000 employees start manufacturing your shit.

Nexenta + Pogo vs. Sun 7xxx?

[R Sync'r]

NexentaStor has been in the market for a while and seems to be doing well. They run on virtually any x86 and PogoLinux is offering an integrated solution with support. And yes prices are much, much less for much more capacity than Sun. Sun beats them on SSDs though. But some of the software Nexenta has done looks quite cool and we're using it behind ESX including the free Nexenta plug in to discover VMs and do ZFS replications. Have not messed around with the CDP yet or much else in NexentaStor.