Sun Adding Flash Storage to Most of Its Servers

BobB-nw writes "Sun will release a 32GB flash storage drive this year and make flash storage an option for nearly every server the vendor produces, Sun officials are announcing Wednesday. Like EMC, Sun is predicting big things for flash. While flash storage is far more expensive than disk on a per-gigabyte basis, Sun argues that flash is cheaper for high-performance applications that rely on fast I/O Operations Per Second speeds."

Comments

We are going to have two layers of storage

[javilon]

I would put the operating systems, binaries and configuration files on the SSD.

But most of what makes up the volume on current computers (log files, backups, video/audio) can be committed to a regular hard drive.

Re:We are going to have two layers of storage

[3HackBug77]

That definately makes sense, It is a very expensive idea to make EVERYTHING SSD, it doesn't really make sense, because except for on the local level it wouldn't really make a huge difference either, due to network bandwidth limits.

Re:We are going to have two layers of storage

[kcbanner]

I'm not sure how much they have improved flash, but I know CF cards can only handle like 200k writes to the same sector before that sector dies. In my soekris router I have logging turned down, but I haven't actually seen any real evidence of cards dying from too many writes. Anyone have any real evidence of this?

Re:We are going to have two layers of storage

[Anonymous Coward]

Why not just run those apps off of a ram drive? the only benefit i see in doing this would be faster boot/start time, which is kind of pointless when you have servers that stays up for months.

Re:We are going to have two layers of storage

[blhack]

"this is no longer an issue, n00b"

x500.

Re:We are going to have two layers of storage

[CastrTroy]

Why would you bother putting the programs and operating system on SSD for a server? Once the files are loaded into memory, you'll never need to access them again. SSD only helps with OS and Programs when you are booting up, or opening new programs. This almost never happens on most servers.

Re:We are going to have two layers of storage

[Anonymous Coward]

when you have servers that stays up for months I'm a Windows admin, you insensitive clod!

Re:We are going to have two layers of storage

[MyNymWasTaken]

You are correct. I inadvertently mod'ed you "redundant", rather than "insightful", so this reply is to knock that off.

Strangely enough, I had to log in using IE to post this comment. While using Seamonkey, it would state "If you continue to post this comment, all moderations done to this discussion will be undone! Are you sure you want to post?" and not me any option to say "Yes."

Re:We are going to have two layers of storage

[Anonymous Coward]

This story should be tagged: !longeranissuen00b

Re:We are going to have two layers of storage

[Archangel Michael]

We already have more than two layers.

Registers, on chip cache, off chip cache, Ram, Flash, HD, LongTerm Low Volatility (tape, CD/DVD etc)

Re:We are going to have two layers of storage

[compro01]

I was sure that figure was upwards of a million cycles per sector in modern flash chips.

Also, throw in wear-leveling and spare sectors. a million writes to a file system sector doesn't mean a million writes to a particular physical sector (could be 1000 writes each to a 1000 different sectors) and when a sector does wear out, it simply gets put out of service and is replaced with a spare one. this same principle is used in mechanical hard drives. if a sector is problematic to read from/write to, it gets marked as bad and the file system sector is remapped to somewhere else.

SSDs could quite likely last longer than mechanical hard drives in this regard.

Re:We are going to have two layers of storage

[boner]

RAM drive uses DRAM, Enterprise class DRAM ~ $100/GB and uses ca 8W/GB. Enterprise Flash, ~ $30-80/GB and uses 0.01W/GB

In addition, assume that 90% of ram-drive accesses go to 10% of the storage, you can see that effectively you are burning a lot of energy with zero gain. Multiply by up-time.

Flash has the potential of greatly improving performance/watt for most servers.

Re:We are going to have two layers of storage

[CastrTroy]

Since when is CD/DVD low volatility?

Re:We are going to have two layers of storage

[Z34107]

Random access, relational, transactional, adjective-al databases.

Random access time is (evidently) much better on flash. That's how Vista ReadyBoost works - there's a performance boost (a tiny one) if you let it put the non-sequential parts of the swap file onto a flash key.

I imagine that you could increase performance for some types of databases by running on a solid-state drive.

Re:We are going to have two layers of storage

[Anonymous Coward]

It most certainly is an issue. We bought an SSD from freshly formatted to its knees within a few weeks by logging to it (multiple gigs rolling per day).

Re:We are going to have two layers of storage

[boner]

Ummm, most programs are not completely loaded into memory and inactive pages do get swapped out in favor of active pages. While the most active regions of a program are in memory most of the time, having the whole program in memory is not the general case.

Also, DRAM burns ~8W/GB (more if FB-DIMMS), Flash burns only 0.01W/GB. Thus swapping inactive pages to Flash allows you to use your DRAM more effectively, improving your performance/W.

From a different perspective: you have a datacenter and you are energy constrained. Most applications use 10% of the DRAM 90% of the time. It may be an attractive proposition to give the applications less DRAM (at a slight performance loss) and let them swap to Flash (with a significant reduction in power). Multiply by 10000 servers, even a 20W reduction per server becomes significant.

Re:We are going to have two layers of storage

[BlendieOfIndie]

It sounds like the SSDs are internal drives for the server. A database would never be stored on an internal hard drive. Almost any commercial database is connected to a disk farm through SAN fabric.

SSDs really shine for OLTP databases. Lots of random IO occurs on these databases (as opposed to data warehouses that use lots of sequential IO).

Normal hard drives are horrible for random IO because of mechanical limitations. Think about trying to switch tracks on a record player thousands of times per second; this is whats happening inside a hard drive (under a random IO load). Its amazing mechanical HDDs work as well as they do.

Re:We are going to have two layers of storage

[dgatwood]

I was thinking about this at Fry's the other day when trying to decide whether I could trust the replacement Seagate laptop drive similar to the one that crashed on me Sunday, and I concluded that the place I most want to see flash deployed is in laptops. Eventually, HDDs should be replaced with SSDs for obvious reliability reasons, particularly in laptops. However, in the short term, even just a few gigs of flash could dramatically improve hard drive reliability and battery life for a fairly negligible increase in the per-unit cost of the machines.

Basically, my idea is a lot like the Robson cache idea, but with a less absurd caching policy. Instead of uselessly making tasks like booting faster (I basically only boot after an OS update, and a stale boot cache won't help that any), the cache policy should be to try to make the hard drive spin less frequently and to provide protection of the most important data from drive failures. This means three things:

1. A handful of frequently used applications should be cached. The user should be able to choose apps to be cached, and any changes to the app should automatically write through the cache to the disk so that the apps are always identical in cache and on disk.
2. The most important user data should be stored there. The user should have control over which files get automatically backed up whenever they are modified. Basically a Time Machine Lite so you can have access to several previous versions of selected critical files even while on the go. The OS could also provide an emergency boot tool on the install CD to copy files out of the cache to another disk in case of a hard drive crash.
3. The remainder of the disk space should be used for a sparse disk image as a write cache for the hard drive, with automatic hot files caching and (to the maximum extent practical) caching of any catalog tree data that gets kicked out of the kernel's in-memory cache.

That last part is the best part. As data gets written to the hard drive, if the disk is not already spinning, the data would be written to the flash. The drive would spin up and get flushed to disk on shutdown to ensure that if you yank the drive out and put it into another machine, you don't get stale data. It would also be flushed whenever the disk has to spin up for some other activity (e.g. reading a block that isn't in the cache). The cache should also probably be flushed periodically (say once an hour) to minimize data loss in the event of a motherboard failure. If the computer crashes, the data would be flushed on the next boot. (Of course this means that unless the computer had boot-firmware-level support for reading data through such a cache, the OS would presumably need to flush the cache and disable write caching while updating or reinstalling the OS to avoid the risk of an unbootable system and/or data loss.)

As a result of such a design, the hard drive would rarely spin up except for reads, and any data frequently read would presumably come out of the in-kernel disk cache, so basically the hard drive should stay spun down until the user explicitly opened a file or launched a new application. This would eliminate the nearly constant spin-ups of the system drive resulting from relatively unimportant activity like registry/preference file writes, log data writes, etc. By being non-volatile, it would do so in a safe way.

This is similar to what some vendors already do, I know, but integrating it with the OS's buffer cache to make the caching more intelligent and giving the user the ability to request backups of certain data seem like useful enhancements.

Thoughts? Besides wondering what kind of person thinks through this while staring at a wall of hard drives at Fry's? :-)

Re:We are going to have two layers of storage

[Anonymous Coward]

There are no spare sectors in a 32GiB SSD.

Re:We are going to have two layers of storage

[menace3society]

I disagree that these disks should be used as a write cache. Frequent, incremental modifications to files is exactly what you DON'T want to use flash/SSD for, since it will wear out larger disk "blocks" faster than regular hard-disk writing. If you're not going to take advantage of HDD technology's superior write lifetime, you might as well not have one at all.

Re:We are going to have two layers of storage

[menace3society]

No marketing or sales executive will ever countenance the adding of "spare" sectors to a disk. If there are 100 billion physical sectors, then by God it's going to say so on the sales info.

In order to do wear leveling you have to have additional metadata, which will take up additional space on disk. With SSD you pay a premium per byte over a magnetic disk. Do you really want a file system that's not going to make the best use of the space you just bought for an arm and a leg?

Re:We are going to have two layers of storage

[billcopc]

Why not just add more system ram for the kernel to use as cache ? I realize flash is a bit cheaper than DRAM now, but still it seems like a very roundabout way to improve performance when existing OS architecture can do the same job and is well documented.

Adding another layer of caching via flash just means we'll have more things to go wrong.

Re:We are going to have two layers of storage

[Anonymous Coward]

As capacity goes up, the feature size on flash gets smaller. This means less energy per bit and a thinner dielectric.

So, as density of flash goes up, write cycle lifetime potentially goes down.

HDDs have the same issue of bits being less "durable" as capacity goes up. However, the media never wears out for HDD. Furthermore, it is already accepted that there will be many bit errors and these are simply corrected with error correction codes and mapping out bad sectors.

As far as reliability goes, everybody talks about it but nobody actually buys on the basis of reliability. At the end of the day it all comes down to dollars per gigabyte for most applications.

Power usage on the other hand is becoming more and more important. That may actually be a strong selling point.

As I've said elsewhere, the first step is going to be OS support for treating Flash differently than HDD. This will allow for hybrid storage solutions. At that point we will see the medium between HDDs and Flash. Right now, it is all or nothing going Flash. In that environment it is going to be hard for Flash to get going.

Re:We are going to have two layers of storage

[CastrTroy]

But the original post was talking about putting the actual programs and OS themselves on the SSD, not the swap file. Obviously it makes sense to put the swap file, and any other frequently and random accessed, files on the SSD.

Re:We are going to have two layers of storage

[juiceboxfan]

I was sure that figure was upwards of a million cycles per sector in modern flash chips. I keep seeing this figure of a million cycles per sector but have yet to see it on a datasheet.

To paraphrase the above response;

"No marketing or sales executive will ever countenance the downward estimation of erase/write cycles per sector. If the chip can survive one million cycles per sector, then by God it's going to say so on the sales info."
Last time I checked no one was listing the lifetime of their chip at more than 100k cycles per sector.

What manufacturer is quoting anything near one million erase/write cycles per sector?

Re:We are going to have two layers of storage

[cheater512]

Hard drives are hardly unreliable.
I've never had one die on me yet.

They've worked fabulously until now and there is no reason why they will die overnight.
Price, raw speed (as opposed to seek) and size are the main factors.

Your caching ideas sound awfully like what Linux does out of the box with RAM.
And you want to *reduce* writes to SSDs, not increase them.

Re:We are going to have two layers of storage

[erikdalen]

I've had about 10 die on me. Probably around a third have died before they've been replaced because they were too small. That is during the last 20 years though.

So I wouldn't exactly call them reliable. I don't store anything important on just a single HD any more. But I don't think I'd do that with a SSD either.

Re:We are going to have two layers of storage

[dgatwood]

Five years ago, I would have agreed. These days, some of the better flash parts are rated as high as a million write cycles. If we're talking about 4 GB of flash, a million write cycles on every block would take a decade of continuous writes at 10 megabytes per second. Real-world workflows obviously won't hit the cache nearly that hard unless your OS has a completely worthless RAM-based write caching algorithm.... Odds are, the computer will wear out and be replaced long before the flash fails. That said, in the event of a flash write failure, you can always spin up the drive and do things the old-fashioned way. And, of course, assuming you put this on a card inside the machine, if it does fail, you wouldn't have to replace the whole motherboard to fix the problem.

That said, to reduce thrashing of the write cache, it might be a good idea to add a cap of a meg or two and spin up the hard drive asynchronously once the write cache size exceed that limit. Continue writing to the flash to avoid causing user delays while the HD spins up (huge perceived user performance win there, too) and flush once the drive is up to speed.

You could also do smart caching of ephemeral data (e.g. anything in /tmp, /var/tmp, etc.). Instead of flushing changes those files to disk on close, wait to flush them until there's no room for them in the RAM buffer cache, and then flush them to the flash. After all, those directories get wiped on reboot anyway, so if the computer crashes, there's no advantage to having flushed anything in those directories to disk....

BTW, in the last week, I've lost two hard drives, both less than a year old. I'm not too impressed with the write lifetimes of Winchester disk mechanisms. :-)

Re:We are going to have two layers of storage

[dgatwood]

Because write caches in RAM go away when your computer crashes, the power fails, etc. Battery-backed RAM is an option, but is a lot harder to get right than a USB flash part connected to an internal USB connector on a motherboard.... In-memory write caching (without battery backup) for more than a handful of seconds (to avoid writing files that are created and immediately deleted) is a very, very bad idea. There's a reason that no OS keeps data in a write cache for more than about 30 seconds (and even that is about five times too long, IMHO).

Write caching is the only way you can avoid constantly spinning up the disk. We already have lots of read caching, so no amount of improvement to read caching is likely to improve things that dramatically over what we have already.

Even for read caching, however, there are advantages to having hot block caches that are persistent across reboots, power failures, crashes, etc. (provided that your filesystem format provides a last modified date at the volume level so you can dispose of any read caches if someone pulls the drive, modifies it with a different computer, and puts the drive back). Think of it as basically prewarming the in-memory cache, but without the performance impact....

Re:We are going to have two layers of storage

[barryp]

Umm, reliability and convenience. If you smoke a harddisk holding your data, you can still boot the OS and have access to all your tools, networking, etc.. and can easily restore your data drives.

Also, with the OS and programs, any limited number of write to the device is a non-issue.

Re:We are going to have two layers of storage

[MrResistor]

the cache policy should be to try to make the hard drive spin less frequently

Actually, this is exactly what you don't want to do. What you really want to do with an HDD is leave it spinning for as long as possible. Spin-up is when you get most of the mechanical wear, thus shortening the life of the drive. As an added bonus it uses a lot of power, too.

Re:We are going to have two layers of storage

[dgatwood]

Yes and no. You're right that spinning up and down causes more mechanical wear on the spindle motor. However, leaving drives in laptops running continuously is also bad. Hard drives don't like heat, and laptop enclosures are not designed to dissipate heat from the drive. They basically have zero airflow across the drive, so the top of the drive enclosure and the case develop this layer of heated air that further insulates the drive from dissipating heat.

Further, AFAIK, spindle motors haven't been the most common cause of problems since the move to fluid bearings. The biggest cause of damage in modern hard drives is failure of the heads (electrically or mechanically) due to damage from the parking ramp. The odds of this damage are greatly increased if the drive is heated, as heating causes uneven expansion of the head arm. The second most common cause is a head crash. The odds of a head crash are also dramatically increased at higher temperatures for the same reason---the heads can end up flying closer to the platter, so a much smaller impact can cause a scarring event....

While the drive does park the heads during spin-up/spin-down, the drive may also park the heads at any time when the drive is inactive without spinning the drive down to reduce the risk of head crashes in the event of sudden impact. Also, newer laptop drives automatically park the heads very rapidly if their motion sensors detect the machine is falling for the same reason. That extra parking contributes much more significantly to premature failures than cycling power on the spindle motor. More to the point, this means that idle periods without spinning the drive down are exactly as bad as spinning the drive down, at least where head failures are concerned. Thus, if you can instead park the heads and spin the drive down and leave it down and only spin it up periodically, you are actually reducing the two most common causes of premature drive failure.

Most modern operating systems already limit the number of spin-downs by leaving the drive running for a minimum of a minute at a time. Using flash for write caching would take care of the other piece of the puzzle by eliminating the unnecessary spin-ups and allowing the system drive to spin down where it otherwise might never be able to do so (if the sync daemon spins it up every 30 seconds and it only spins down after a minute of inactivity... you do the math). Don't look at this cache idea in isolation. Look it as just another part of the caching and power management that is already in place. In general, this won't increase the number of spin-ups and spin-downs except in the degenerate case where the number of spin-downs is zero because of continuous trickle activity. It will usually decrease the spin-up/spin-down count by removing one of the most common (and unnecessary) causes of spin-ups.

Note: I am not a storage engineer, so take this with a grain of salt.

Re:We are going to have two layers of storage

[MrResistor]

Good points, most of my drive knowledge relates to big RAID installations, so I wasn't thinking about laptops when I responded. My own preference would be to replace the HDD in the laptop entirely, and that's getting more and more reasonable.

Yes, IDE flash drives are pretty expensive, but you can get a 32GB CF card and a CF-IDE adapter for around $150 last time I checked. Supposedly the tech that allows for 32GB CF also makes 64GB possible, which is the sweet spot for me on a laptop, but I don't seem to be seeing them anywhere.

Re:We are going to have two layers of storage

[dgatwood]

Last I checked, flash write performance (at least for CF and USB stuff) still left something to be desired. A laptop hard drive still exceeds the speed of flash when writing by a factor of about 2-3 or thereabouts. The read performance is almost even for 5400 RPM laptop drives, and within a factor of 1.5 of even 7200 RPM drives, so that's more livable. You probably wouldn't want to capture audio or video to a flash drive or do other tasks that involve continuous high speed writes, at least for now.

Compared with the time to spin up a drive, of course, small flash reads and writes are faster by maybe a factor of a thousand.... From a performance perspective, until flash gets a bit faster, it still probably makes sense to mix the two, IMHO. Medium term, though, I totally agree that in laptops, HDs should ideally go away entirely.

At current prices, flash parts in consumer form factors are still about 7x-12x more expensive (about $3-5 per gig compared with about $0.42 per gig for a laptop hard drive). You're right, though, that the cost of moving to flash drives is getting a lot more palatable. Give it five years and people will be asking what kind of idiot would put a spinning drive in a laptop. :-)

Re:We are going to have two layers of storage

[billcopc]

Fantastic! Thank you, sir!

Write cycles. again.

[Amiga Lover]

Cue up 20 comments going "But what about the limited write cycles, these things will fail in a month" and 500 comments replying "this is no longer an issue n00b"

Re:Write cycles. again.

[morgan_greywolf]

You forgot the 1000 comments prognosticating about SSDs replacing HDDs permanently "any day now" with the added bravado of saying "I knew this would happen! See, I told you!" with 3000 comments replying 'Yeah, but price/performance!", all of which will be replied to with "but price/performance doesn't matter, n00b. Price makes no difference to anyone."

Then, in a fit of wisdom, a few posters, all of whom will be modded down as flamebait, will say "There's room for both and price/performance does matter, at least for now."

Re:Write cycles. again.

[maxume]

I'm just glad there is enough interest in paying for the performance to keep the development moving at a decent clip, flash really does look like it will have a big advantage for laptop users that are not obsessed with storing weeks worth of video.

Re:Write cycles. again.

[jonaskoelker]

"Guys like us avoid monopolies. We like to compete." -- Bill Gates (parent's sig)

This is no longer an issue n00b.

Re:Write cycles. again.

[Anonymous Coward]

Cue up 20 comments going "But what about the limited write cycles, these things will fail in a month" and 500 comments replying "this is no longer an issue n00b" queue up 20 comments going "But what about the dictionary and the differences in proper use of the words cue and queue" and the 500 comments replying "people are too ignorant these days to care and soon enough they'll just put it in the dictionary, along with words like alright, alot and irregardless, n00b."

Seriously, dude, pedantry is not your forte.

Re:Write cycles. again.

[Anonymous Coward]

> Seriously, dude, pedantry is not your forte.

Heh, nah, it's not yours. From Oxford:

To cue up a videotape, an audiotape, or a compact disc is to have it ready for playing at a particular pointâ"e.g.: âoeHis brother cued up the tape, the rousing theme song from âRocky.â(TM) ( Hartford Courant; Sept. 17, 1996.)

As in, cue up the same old comments played in the same order, just like a broken record.

Do get it right!

Re:Write cycles. again.

[Anonymous Coward]

haha pwnt.

Re:Write cycles. again.

[uid8472]

And one person (i.e., me) to mention drum storage from back in the old days (i.e., before I was born), just like every other time it's come up and I've seen it and been bothered to comment, and probably with a link to multicians.org.

Re:Write cycles. again.

[Nullav]

Sorry, but the correct term is 'pwnted'...n00b.

Re:Write cycles. again.

[Detritus]

Someone once told me that IBM used head-per-track disks for VM paging on some of their early mainframes. That was a performance hot spot that could addressed by spending money on a very fast storage device.

Re:Write cycles. again.

[Anonymous Coward]

To cue up a videotape, an audiotape, or a compact disc is to have it ready for playing at a particular pointâ"e.g.: âoeHis brother cued up the tape, the rousing theme song from âRocky.â(TM) ( Hartford Courant; Sept. 17, 1996.) That'd be just dandy if we were talking about stage, theater or, even audiovisual media. But then again, we're not, are we?

I always use Merriam-Webster because it's just a better dictonary. (Sure it's purely opinion but, what the hell, it supports my point of view.)

queue
transitive verb
: to arrange or form in a queue
intransitive verb
: to line up or wait in a queue: often used with up As in, queue up the assholes who don't know how to use a dictionary.

Don't worry, some day you'll be right. As I've always said, collective ignorance will eventually get it added to the dictionary as acceptable usage. I can see it starting already.

Re:Write cycles. again.

[Anonymous Coward]

You're sounding like a broken record trying desperately hard to convince yourself you're right.

But wait, that metaphor was already lost on you.

Get it right :)

You might have liked to look a bit further into the dictionary you chose.

Main Entry: cue
Function: verb
Inflected Form(s):
cued; cu*ing or cue*ing
Date: circa 1784
transitive verb
1 : queue
2 : to strike with a cue
intransitive verb
1 : queue
2 : to use a cue

Your dictionary of choice's definition of 'cue' = 'queue' means you agree with me, presuming you still use Merriam-Webster.

Further, you agree with me as far back as 1784. Eventually it'll be added to the dictionary, right?

Good

[brock bitumen]

They are trying to push new technology on their high paying customers because they can get a premium since it's a scarce resource, this will drive up production, and down the costs, and soon we'll all be toting massive flash disks all the day

I, for one, welcome our new flash disk overlords

Re:Lifespan?

[Anonymous Coward]

this is no longer an issue n00b

Re:Lifespan?

[jo42]

When the old one fails, they sell you a new one. This is The Business Plan (c)(tm).

Re:Lifespan?

[Chabil Ha']

Oh no, the flash wear myth! Try this on.

Re:Lifespan?

[Yetihehe]

Just two posts down from http://hardware.slashdot.org/comments.pl?sid=573849&cid=23655427, congratulations!!

Samsung 256GB Flash Drive

[jackspenn]

Funny how they are getting behind 32 GB flash drive when Samsung is supposed to have 256GB flash drive out by end of year for around $1000. That would be bleeding edge, that would be news. This is just a story about SUN doing something that others have already done in for sometime now.

Re:Samsung 256GB Flash Drive

[QuantumRiff]

This is just a story about SUN doing something that others have already done in for sometime now

Really? What other top 5 computer manufacturer has been putting flash drives in SERVERS? I've seen a few laptops, but I haven't seen any used in servers or storage systems. (EMC and a few others have announced plans to do it, but haven't released anything AFAK)

Also, their "thumper" server has 48 drives in it. Would you want to pay around $1000 per drive to fill that up?

Re:Samsung 256GB Flash Drive

[bubulubugoth]

IBM has them as option for blades and racks servers...

Re:Samsung 256GB Flash Drive

[Archangel Michael]

Also, their "thumper" server has 48 drives in it. Would you want to pay around $1000 per drive to fill that up? Yes. If performance dictated it was necessary.

Just because you don't want to, doesn't mean everyone else doesn't want to also.

Re:Samsung 256GB Flash Drive

[everphilski]

$48k? Chump change. I remember back when the company I worked for at the time paid over six figures for a pimped out server back in the late 90's...

Re:Samsung 256GB Flash Drive

[Anonymous Coward]

Most server manufacturers are reluctant to include the drives in servers where disk writes are common because of possible corruptions due to sector wear.

This problem hasn't been solved by the drive manufs, although their marketing depts have convinced many!

Re:Samsung 256GB Flash Drive

[Calinous]

Samsung will have Multi Level Cells, which are slower (and cheaper). The Single Level cells are faster (up to twice as fast I think), but more expensive.
      You can go either way with it, but I think faster (and smaller) drives are more attractive than bigger and slower.
      You need to compete against the sequential speed of a 15,000 rpm SCSI drive too (SSD will beat them dead on access speed, but not all workloads are small random reads)

Re:Samsung 256GB Flash Drive

[Anonymous Coward]

So does Sun for their 4 socket x86 blades.

Re:Samsung 256GB Flash Drive

[mistahkurtz]

see above, titled "IBM"

Re:Lifespan?

[Anonymous Coward]

Two minutes after the cue, you guys are slow.

big deal

[larry bagina]

Most computers come with flash preloaded. I don't know why you'd be browsing the web or watching videos/web comics/ads/etc on a server computer. Maybe they're trying to dumb diwn ti compete with Windows Server 2008.

Re:big deal

[GuyverDH]

Please.... Please.... Please.... Tell me you were joking....

I can usually read into the comment if someone is joking or not... but this one... I dunno... Could go either way....

Re:big deal

[oodaloop]

Uh, wrong kind of flash? If that was a joke, it was a bit too subtle for me, sorry.

Re:big deal

[urcreepyneighbor]

Most computers come with flash preloaded. I heard they were going to include Java, too. Like, whatever, big deal.

Re:big deal

[mentaldrano]

Just mod him funny - then we will all know it's a joke, without having to exercise our critical thinking skills. I know I don't come here to do that, I rely on the moderators. Here, I've even got mod points, I'll do it for you.... Oops!

Re:big deal

[mgblst]

Yeah, that joke was to subtle, like a brick to the head.

+1 point for dissing Microsoft.
+1 point for being honest.

Re:Lifespan?

[brunokummel]

I wonder how many posts of those 500 comments would get modded as funny before being modded redundant.. =)

Two layers, but not those ones

[clawsoon]

We are going to have two layers, but they'll be deeper in the filesystem than that.

High frequency, low volume operations - metadata journalling, certain database transactions - will go to flash, and low frequency, high volume operations - file transfers, bulk data moves - will go to regular hard drives. SSDs aren't yet all that much faster for bulk data moving, so it makes the most economic sense to put them where they're most needed: Where the IOPs are.

Back in the day, a single high-performance SCSI drive would sometimes play the same role for a big, cheap, slow array. Then, as now, you'd pay the premium price for the smallest amount of high-IOPs storage that you could get away with.

Good news for us

[R3N3G4D3]

The fact that SSDs are becoming main stream in the corporate world means that their prices should start dropping soon, which is definitely a good thing for the rest of us.

Re:Good news for us

[morgan_greywolf]

Uh, there's a large difference between 'Sun is adding Flash storage to most of its servers' and 'SSDs are becoming mainstream in the corporate world.' Just because one vendor sells machines with SSDs doesn't mean that they will be bought or even used in the way Sun intends.

You'll know they're becoming mainstream when, two weeks from now, you start getting mails from contract house recruiters perusing your resume on Monster looking for "SSD Storage Engineers" with "10 or more years of experience on Sun equipment".

This will even further ZFS

[E-Lad]

Current versions of ZFS have the feature where the ZIL (ZFS Intent Log) can be separated out of a pool's data devices and onto it's own disk. Generally, you'd want that disk to be as fast as possible, and these SSDs will be the winner in that respect. Can't wait!

Re:This will even further ZFS

[allanw]

Current versions of ZFS have the feature where the ZIL (ZFS Intent Log) can be separated out of a pool's data devices and onto it's own disk. Generally, you'd want that disk to be as fast as possible, and these SSDs will be the winner in that respect. Can't wait! As far as I know, contiguous writing of large chunks of data is slower for flash drives than plain HDD's. I'm guessing the ZIL is some kind of transactional journal log, where all disk writes go before they hit the main storage section of the filesystem? I don't think you'd get much of a speed bonus. SSDs are only really good for random access reads like OLTP databases.

Re:This will even further ZFS

[Anonymous Coward]

The benchmarks say something like a 200x performance by putting the ZIL onto the an alternate high performance logging device.

I have been actively researching a vendor who will supply this type of device. Currently we're testing with Gigabyte i-Ram cards, connected in through a separate SATA interface. (Note: Gigabyte are battery backed SDRAM .. but I won't have lost power for 12 hours so it's a non-issue for me)

Fusion-IO is a vendor who is making a board for Linux - but as near as I can tell the cards aren't available yet, and when they are - they won't work with Solaris anyway!

The product which Neil Perrin did his testing with (umem/micromemory) with their 5425CN card doesn't work with current builds of Solaris. Umem is also a pain to work with .. they don't even want to sell the cards (I managed to get some off eBay)

I hope Sun lets me buy these cards separately for my HP proliant servers. Of course if they didn't, this is one thing that might make me consider switching to Sun Hardware! (Hey HP/Dell - are you reading this??)

Re:This will even further ZFS

[Anonymous Coward]

The benchmarks say something like a 200x performance by putting the ZIL onto the an alternate high performance logging device.

What percentage of the gain is from "alternate" and how much is from "high performance"? Traditional hard disks do well at sequential operations.

I'm surprised that it is big enough to talk about.

[fuzzyfuzzyfungus]

Given that you can get flash disks that hang off pretty much any common bus used for mass storage(IDE, SATA, SAS, USB, SPI, etc.) "Adding a flash storage option" is pretty much an engineering nonevent, and a very minor logistical task.

If Sun expects to sell a decent number of flash disks, or is looking at making changes to their systems based on the expectation that flash disks will be used, then it is interesting news; but otherwise it just isn't all that dramatic. While flash and HDDs are very different in technical terms, the present incarnations of both technologies are virtually identical from a system integration perspective. This sort of announcement just doesn't mean much at all without some idea of expected volume.

Re:I'm surprised that it is big enough to talk abo

[gbjbaanb]

"Adding a flash storage option" is pretty much an engineering nonevent but as a marketing event its a magnificent and almost unbelievable paradigm-shift approach to a massive problem that's been crying out for a reliable storage-based performance solution for years.

IOPS

[Craig Ringer]

People (read: vendors) now frequently refer to flash storage as superior when IOPs are the main issue.

From what I've been able to discern this is actually true only in read-mostly applications and applications where writes are already in neat multiples of the flash erase block size.

If you're doing random small writes your performance is likely to be miserable, because you'll need to erase blocks of flash much larger than the data actually being changed, then rewrite the block with the changed data.

Some apps, like databases, might not care about this if you're able to get their page size to match or exceed that of the underlying storage medium. Whether or not this is possible depends on the database.

For some other uses a log-oriented file system might help, but those have their own issues.

In general, though, flash storage currently only seems to be exciting for random read-mostly applications, which get a revolting performance boost so long as the blocks being written are small enough and scattered enough. For larger contiguous reads hard disks still leave flash in the dust because of their vastly superior raw throughput.

Vendors, however, make a much larger margin on flash disk sales.

This article (PDF) may be of interest:
Understanding Flash SSD performance
(google text version).

Re:IOPS

[sam_vilain]

When I last investigated this, the actual write cycle time was also much higher than for a 10k or 15k RPM disk. I think this is the bigger issue, not the fact that you have to read the whole page back first (something which is very fast anyway)

MOD DOWN

[r_jensen11]

Then, in a fit of wisdom, a few posters, all of whom will be modded down as flamebait, will say "There's room for both and price/performance does matter, at least for now." FLAMEBAIT! Mod down this mutual funding* troll!

*Search for "mutual funding money" AND "Samuel L. Jackson" - quite possibly the best dub to "clean up" some "dirty language"

MBTF

[marafa]

whats the mbtf on this kind of stuff? as far i know my 256mb flash drive has a lifetime of 100000 writes. has that improved?

Re:I'm surprised that it is big enough to talk abo

[fuzzyfuzzyfungus]

Good point. And I strongly suspect that it enables today's Dynamic CIOs to realize unprecedented First-Mover Synergies in the modern Data-Centric Enterprise Solution Space.

Wha??!?!?

[Anonymous Coward]

Connected to a PCI-x16 133mhz interface we're talking about at a non-blocking device with up to a 4GB/s connection to the bus -- compared to a SATA or SAS interface which is at best 3gb/s .. if you can even find a device interface that fast.
(notice the GB GIGABYTES vs. gb GIGABITS) .. also the seek times of ~50 microseconds really turns me on.

Yeah.. I know I could buy a 4gb/s FC RAMSAN unit -- anybody got $50k laying around .. oh wait, I need redundancy so that's $100k. (and btw it's *STILL* slower than one of these cards)

oh yeah, and they don't burn a rack unit either, and their power consumption (and heat) is ridiculously lower.

Not that any of us need high performance storage devices for things like databases, or ZFS journal logs. .. I've got almost 90 amps of redundant power coming into each of my cabinet right now.. so the power savings alone is attractive to me.

Re:Wha??!?!?

[Anonymous Coward]

Huh? PCI-X @133MHz is max. 133MB/s*4, ok, that's 4Gbit/s, about 8 times slower than 4GB/s, assuming ideal conditions... There's no '16' in it, as well. Now PCI-express x16 can transfer 4GB/s, with 32 lanes, 8GB/s full duplex. 2 Gbit/s per lane, that's 32/64 Gbit/s, for the reference. And it works at 100 MHz, but that's completely meaningless anyways these days.

Re:I'm surprised that it is big enough to talk abo

[boner]

Re: "Adding a flash storage option" is pretty much an engineering nonevent, and a very minor logistical task.

You have no idea what you are talking about. Sun customers demand that the product Sun sells them have known reliability properties and that Sun guarantees their products properly interact with each other. It takes a significant amount of resources to do this validation. At the same time SSDs and HDDs react very differently to load and can have all sorts of side effects if the OS/application is not prepared to deal with them.

Re:Lifespan?

[ShieldW0lf]

There is no flash wear myth. If it was a myth, they never would have gone to all that trouble. The whole point behind Static Wear Leveling is to mitigate a very significant and real weakness in the storage medium.

The fact that flash is only really well suited for infrequent writes and frequent non-contiguous reads doesn't bode well for its utility in OLTP applications.

Re:I'm surprised that it is big enough to talk abo

[BlendieOfIndie]

Sure, this is definitely not a big engineering feat. However, neither of Suns competitors, IBM or HP, offer SSDs at the moment. This article is indicating that SSDs are gaining momentum (if not attention) in the high-end server market. Suppliers need to offer a product before it can be sold.

Power consumption and heat dissipation

[UpooPoo]

I work in a company that has a few thousand servers running in a few regional data centers. We are looking into SSDs not because of their superior IOPs (this is a mitigating factor vs HDD performance) but because of their low power consumption and low heat dissipation. When you scale your operations reach a scale where you are using an entire data center, heating and power become more and more of a cost issue. Right now we are trying to build some hard data on actual sabings, but there's lots of spin out there that gives you an idea of what potential savings could be. Here are a few interesting links, google around for more information, there's plenty to be had:

http://www.stec-inc.com/green/storage_casestudy.php
http://www.stec-inc.com/green/green_ssdsavings.php (You have to request the whitepaper to see this one.)

All I want to know is...

[elysiana]

How do I get voyeur privileges? Is it pay-per-view? What's the ratio of male flashers to female flashers?

These are the important questions.

Re:I'm surprised that it is big enough to talk abo

[Jor-Al]

However, neither of Suns competitors, IBM or HP, offer SSDs at the moment. Year about a year too late making that comment. IBM having SSDs in their Blades: http://www.itbusinessedge.com/blogs/dcc/?p=175

RAID 4, anyone?

[mentaldrano]

In the time between now and when SSD becomes cheaper than magnetic storage, might we see a resurgence of RAID 4? RAID 4 stripes data across several disks, but stores parity information all on one disk, rather than distributing the parity bits like RAID 5.

This has benefits for workloads that issue many small randomly located reads and writes: if the requested data size is smaller than the block size, a single disk can service the request. The other disks can independently service other requests, leading to much higher random access bandwidth (though it doesn't help latency).

  One of the side effects of this is that the parity disk must be much faster than the data disks, since it must service all requests, to provide the parity info. Here SSD shines, with its quick random access times, but poor sequential performance. Interesting, no?

Re:RAID 4, anyone?

[Anonymous Coward]

Interesting, but it seems like it would result in wasted space.

SSDs come in powers of two, but hard drives tend to come in more decimal oriented numbers like 320, 400, 500, 750. The parity drive must be at least as large as the striped drives, so it will probably be a little larger.
 
  For example, 512 GB SSD holding parity information for some 500GB drives. I suppose that extra 12g could be partitioned off and used for swap or something, though.

Cheaper than RAM?

[AmiMoJo]

At the moment high performance SSDs are still more expensive than RAM. Since a 64 bit processor can address vast amounts of RAM, wouldn't it be even better and cheaper just to have 200GB of RAM rather than 200GB of SSD?

Okay, you would still need a HDD for backing store, but in many server applications involving databases (high performance dynamic web servers for example) a normal RAID can cope with the writes - it's the random reads accessing the DB that cause the bottleneck. Having 200GB of database in RAM with HDDs for backing store would surely be higher performance than SSD.

For things where writes matter like financial transactions, would you want to rely on SSD anyway? Presumably banks have lots of redundancy and multiple storage/backup devices anyway, meaning each transaction is limited by the speed of the slowest storage device.

Re:Cheaper than RAM?

[Doctor Faustus]

Since a 64 bit processor can address vast amounts of RAM, wouldn't it be even better and cheaper just to have 200GB of RAM rather than 200GB of SSD?
I don't know that you'd really need a lot of flash memory for this, maybe only a few meg allowing for spacing out the writes to avoid wear, but flash could allow you to do write caching when you normally wouldn't trust it, because it won't go away if you lose power.

Sun is Afraid of THIS!

[StCredZero]

A single ioFusion card has the concurrent data serving ability of a 1U server cabinet full of media servers. They do this by having 160 channels on a drive controller that also incorporates flash memory. Since each channel is a few orders of magnitude faster than a mechanical hard drive, one card can handle a flurry of concurrent random access requests as fast as 1000 conventional hard drives.

The perfect thing for serving media, where you don't need a few GB per customer, you need the same few GB served out to 1000's or millions of users concurrently. So while $/GB stored stinks, $/GB streamed is fantastic.

Re:I'm surprised that it is big enough to talk abo

[Anonymous Coward]

My grammar checker just had a heart attack.

Re:What a coincidence!

[Anonymous Coward]

first reply to first reply to first post!

what drives are for.

[flaming-opus]

You're confusing two very different sorts of storage. There is bulk data storage. This is a fileserver for home directories, video archives, piles of email, that sort of stuff. This is the market where the 1TB sas drive thrives. Then there's the database backing store. Almost every customer I've sold to wants a huge number of very fast, very small drives for database backing store. The extra capacity is meaningless, as they have to use so many spindles to get a decent IOPS performance. In this area, selling drives hasn't been about capacity for 10 years. IOPS, in particular read IOPS is your throttle point for these. Now that flash drives are beginning to get traction for high-end laptops, and we have affordable, SDD drives, with industry standard interfaces, there's no reason NOT to use them.

Also, fibre channel drives already cost $1000, so paying this much is nothing new for enterprise customers. An enterprise server with LESS than $50,000 of storage would be the oddball case.

IBM

[mistahkurtz]

has been doing this on their blades for about a year.

Why does everyone bother talking about...

[Anonymous Coward]

unreliable flash. Hey, 99% of cases when flash fails, WRITING fails, NOT READING like hard disks! Big deal, just write to a different sector. It's not like you lose any data, you just lose one sector of capacity! Sectors being 64-256kB, that's hardly significant. You should all go apesh*t about how unreliable hard disks really are... Once a hard disk fails, chances are it's both reading and writing capability that's gone... on WHOLE unit.

Hard drives just ain't reliable

[sjbe]

Hard drives are hardly unreliable.
I've never had one die on me yet. You're either incredibly lucky, haven't owned many hard drives or lying. I'll presume you're honest (despite your handle) and the other two are equally likely in my opinion. Personally I've had at least 15 hard drives crap out on me over the years, not those of counting friends, family and coworkers which sends the number well into triple digits. And those are just those I've seen first hand. Add in the ones I know about at companies I've worked in and the number is in the thousands easily. Hard drives are amazing things but they should NEVER be thought of as reliable. In my experience they are the least reliable thing in the computer thanks to all the moving parts. Frankly it's amazing they work as well as they do.

They've worked fabulously until now and there is no reason why they will die overnight. Sure there is. Power surges is the most obvious reason - I've lost several to those. Sometimes hard drives give warnings such as poor performance or weird noises but more often they just break all of a sudden without any warning. Even software to detect drive failures when available is unreliable in detecting imminent failures.

And you want to *reduce* writes to SSDs, not increase them. You want to reduce writes to hard drives too. What's your point?

Re:Hard drives just ain't reliable

[cheater512]

I'd admit that I've been lucky but I have had my fair share of hard drives.

I also admit that I use RAID 5 on my home server (my main data store) but not on my other computers.

How the hell do you get power surges in a computer?
You must be using some pretty dodgy power supplies.
They should die, absorbing the surge rather than letting it through.

Writes to a hard drive dont matter in the slightest.
They are the same as reads in terms of wear and tear.
If you write to SSDs a lot then your looking at it having a very short life.

Re:Hard drives just ain't reliable

[sjbe]

I also admit that I use RAID 5 on my home server (my main data store) but not on my other computers. Actually I use RAID 0 on one of my servers. Still lost two hard drives on it.

How the hell do you get power surges in a computer? Power supplies do not filter out all surges and by design they can't help with power dips. Flip you box on/off 20 times in under 5 seconds and you'll likely have some dead equipment. Have a lightning strike in your vicinity and you likely will have some friend equipment regardless of your power supply make. Recently I had a loose neutral wire on my main in my house which made voltages swing by + or - 40 volts. Not good for the equipment.

And yes I use battery backups on every computer I own plus have filtering gear on every ethernet and phone cable too. I don't trust power quality or reliability at all.

You must be using some pretty dodgy power supplies. I usually put PC Power and Cooling supplies - generally regarded as among the best - in most of my machines. Still lose hard drives and occasionally other gear from time to time.

They should die, absorbing the surge rather than letting it through. Perhaps in theory but the real world doesn't work like that. Frankly most power supplies are poorly made, under-engineered pieces of crap anyway.

Writes to a hard drive dont matter in the slightest.
They are the same as reads in terms of wear and tear. Exactly. You get twice as much wear whereas a SSD gets zero wear on a read operation. I'm joking but only a little. In either case with either device you are shortening the lifetime of the device.

If you write to SSDs a lot then your looking at it having a very short life. This has been disputed ad-nauseum here and elsewhere. I'd prefer to see some actual real world data from folks who've used numerous SSD drives for extended periods. I have seen no compelling evidence that SSD drives are any less reliable or long lasting than platter drives. Not that I would really trust either one.

Re:Hard drives just ain't reliable

[sjbe]

Actually I use RAID 0 on one of my servers. Still lost two hard drives on it. Err I meant RAID 1.

Re:Hard drives just ain't reliable

[sjbe]

How the hell do you get power surges in a computer? Forgot to mention ethernet cables and phone cables are susceptible to power surges too - more so than the main power cable in my experience. I've seen more than a few fried ethernet cards and motherboards from an insufficiently protected dsl modem.

Re:Hard drives just ain't reliable

[cheater512]

Erm. You have a UPS on your computer and it still has dodgy power?
You have something seriously wrong in that case.

Even bare bones UPSes wont allow the voltage fluctuate.
If mine detects anything astray it will 'rebuild' the power completely and give my computer *exactly* 240v.

Re:Hard drives just ain't reliable

[sjbe]

Erm. You have a UPS on your computer and it still has dodgy power?
You have something seriously wrong in that case. No, that's why I don't have hard drives dying from surges anymore. I learned my lesson long ago after I fried one too many bits of electronics. I've lived in too many areas with dodgy power to trust what comes out of the wall anymore. Now my main problems are static and shock from the odd bit of dropped equipment.

Though I did once have a UPS and the attached computers fry due to a nearby lightning strike. Get 1.21gigawats across a UPS and it doesn't matter what you've got protecting it. :-)

Even bare bones UPSes wont allow the voltage fluctuate. That's not true for all UPSes especially some of the cheaper ones. Some will allow fluctuations but only within "safe" ranges. Offline/Standby units just monitor but don't trigger until certain defined conditions are met. You need power conditioning equipment to keep the voltage pegged at specific voltage/amperage levels. (Yes I have some) Usually it's more expensive and not always necessary.

Expensive machines

[sjbe]

$48k? Chump change. I remember back when the company I worked for at the time paid over six figures for a pimped out server back in the late 90's... Server hell - I had an SGI Octane on my desk for a while that cost almost $50K. Two years later it had roughly the same performance as a $5K PC. Plus we had an SGI Onyx AND and Origin2000 that cost a cool quarter million each, plus maintenance.