Slashdot Mirror
Costly SSDs Worth It, Users Say
Lucas123 writes "When you're paying $30,000 for a PCIe flash card, it had better demonstrate an ROI. While users are still struggling with why solid state storage cost so much, when they target the technology at the right applications, the results can be staggering. For example, when Dan Marbes, a systems engineer at Associated Bank, deployed just three SSDs for his B.I. applications, the flash storage outperformed 60 15,000rpm Fibre Channel disk drives in small-block reads. But when Marbes used the SSDs for large-block random reads and any writes, 'the 60 15K spindles crushed the SSDs,' he said,"
Small (and cheap) 32GB SSD for my desktop...
Big powerful 12TB file server using traditional disks for the bulk of my data.
Performance for the stuff where the SSD makes a difference (program files), cheap storage for the stuff where it doesn't (just about everything else).
And if that 32GB drive dies (unproven technology.. MTBF is still a guess) .. I'll buy another cheap (probably cheaper at that point) one and restore from my daily backup.
SSD's outperform disc in applications where they perform better. Story at 11.
So you're saying copypasta can be copied even faster thanks to SSDs?
There's no -1 for "I don't get it."
I know that bcache exists but why isn't there a mainline kernel SSD cache available yet? I could use it for my whitebox SANs where there are dozens of terabytes but only the need for a few gigs to be written on a given day.
Am I supposed to just run out and buy SSDs for the whole load?
So sometimes it's faster, and sometimes it's slower, and always it's more expensive per GB. That makes this a pretty useless article.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
... to produce hybrid technology that would make use of both. I'm not an engineer, but I'm sure some level of hybridization could be produced that would permit a system of prioritization/fragmentation that would make the most efficient use of the hybrid drive. I can even imagine someone developing software that would analyze system performance and make recommendations as to which fragment of the drive (solid/disk) an application ought be run from.
No doubt, a naysayer lacking creativity will tell me its not possible... Please refrain because I will only tell you to stfu and try inventing instead of reproducing.
For example, when Dan Marbes, a systems engineer at Associated Bank, deployed just three SSDs for his B.I. applications, the flash storage outperformed 60 15,000rpm Fibre Channel disk drives in small-block reads. But when Marbes used the SSDs for large-block random reads and any writes, 'the 60 15K spindles crushed the SSDs,' he said,"
So when you need lots of small, random reads, 3x SSDs beat 60x HDDs. Most of the time is spent seeking the file on the HDDs, your ~4.6 ms random seek time is an order of magnitude or more slower than the flash-based drives. No surprise here.
When you are just transferring large files, most of the time is spent actually transferring data. A modern SSD might manage 300-400 MB/s read, but 20x as many HDDs are still going to beat the crap out of them.
The only mildly surprising part is that part about the HDDs winning for all writes, but I guess that really depends on how the test is set up - unless you are actually writing to random parts of the HDD, it is basically a straight-up write operation, so only throughput matters - and again, 60x HDDs are going to beat 3x SSDs (though it is important to note that SSDs are significantly slower at writing than reading in general, although still much faster than an HDD on an individual basis).
While this summary should never have been posted here (please taco, give us good stuf, not 'I know how to move the mouse pointer so I am a l33t hax0r-stuf) but the full article is interesting. It gives a couple of hands-on experiences from users that are quite interesting. It seems the SSD can gain you speed in more situations than previously thought/marketed. Although for my uses I'm not going to spend more than 2 dollar/euro per GB.
i've had a bad experience with SSD drives having returned 2 due to deal breaking problems for me .... on caused BSODs the other could handle suspend mode without locking up.
I have now 2x samsung F3s in raid 0 (plus back up on NAS)
personally i have no desire to have to do the sort of file management small sized SSDs currently demand and since reading the reviews they all pretty much suck.... however SRT is a compelling option... all the convenience of big mechanical drives plus a speed boost of SSD ... unless the price per GB of SSD can be brought down alot then i think SRT may well be the way to go.
actually I am happy to see you, however that is in fact a banana in my pocket.
You can cache your data using SSDs but still have a RAID. Adaptec, LSI, Intel, they all have cards that do it.
To me the cost per GB of an SSD really makes it tough to justify in a lot of cases especially if you want to store large programs like games where you'll need at least a 100GB SSD. However one area place I have started to use low capacity (8 or 16GB) SSDs are in low noise and/or low power environments. If you team them with an ITX Atom board and the right power supply you can build a small computer with no moving parts whatsoever. And the computer will have a very low power usage for applications like HTPCs or network appliances (like firewalls) where the machine might always be powered on.
Confidential Data not safe on SSD.
http://hardware.slashdot.org/story/11/02/17/1911217/Confidential-Data-Not-Safe-On-Solid-State-Disks
So the solution is to totally encrypt your SSD, which slows down it's performance as the CPU has to decrypt/encrypt everything on the fly. Good luck recovering data off the drive in case it won't boot.
If you don't encrypt, then a device like this or software can be used to probe your SSD
http://www.thenewspaper.com/news/34/3458.asp
I've had a boot RAID 0 of two 10,000 RPM drives for years before SSD's ever came out, mostly never used it's extreme speed, except when cloning or copying huge folders of GB's in size. Which was rare. Then the other drive also had to be just as fast and expensive.
So for most SSD is a waste, the larger, less expensive and more private HDD's are still the best value.
SSD's also have limited writes, so they don't lend themselves to a lot of data transfer and changes.
Get a 64Bit machine and bone up on the RAM, perhaps a 7,200 RPM or faster drive, that's the best solution for most users. Storage speed alone isn't a cure all for a fast computer as reads and writes don't occur all the time.
The Emperor insists his new clothes look AWESOME!
Jesus was all right but his disciples were thick and ordinary. -John Lennon
Why are they still so expensive?
If you think about it, the outcome of this test is 100% in favor of the SSD.
Think about it:
The tester was willing to test only 3 SSD's versus *60* 15K drives. So the tester thought that 20 times fewer drives was a fair test for the comparison. What is the tester actually saying here? I think I have a feeling I know. :-)
Anyway, 15K drives are not long for the market. Soon, all that will be left are economy class, 10K, and SSD's.
Only a little more maturity, and the enterprise flood gates will open. When that happens, the hapless victim will be the short stroked IOPS environment, where total IOPS was always the requirement, and that requirement was for more IOPS than capacity. I.e., if a 15K drive offers 400 IOPS, and you need 400,000 sustained, but don't have to store very much at all, your only current choices are buying a lot of 15K drives. Or a bazillion less SSD's.
The switchover point is only a heartbeat away.
Bye 15K drive. I'll miss you.
C//
Get one. It's worth it.
Just as an info dump for anyone who's not familiar with why SSDs perform so much better: SSDs have far better seek performance.
A normal HDD takes about 10ms to seek (3ms at the very high end, 15ms at the low end- 10ms is a good rule of thumb), which means you've got a princely 100 seeks per second per spindle (i.e. HDD). SSDs don't have seek limitations. Looking up a contiguous block of data vs not looking up a contiguous block of data makes no difference to an SSD.
It turns out that 100 seeks isn't a lot in serving infrastructure or, in some cases, on a desktop. When you go to read a file off disk multiple seeks are involved- you need to look up the inode (or equivalent), find the file and a large file will probably be in many different chunks require separate seeks to access them.
Even on a desktop you'll frequently be seek bound not throughput limited. Lets say you are starting up a largish java application (Eclipse might be a good example). It references a huge number of library (.jar) files which are certainly large enough to require many seeks to access. And those libraries are often linked in to system libraries which also have their own dependencies and may have additional dependencies all of which require further seeks. Plus with Eclipse it will look up the time stamps on files in the project... and so on.
During boot of a system is another time when HDD are usually seek bound- lots of different applications/services/daemons are starting at the same time, loading lots of libraries causing lots and lots of seeks.
On server infrastructure a highly utilized database will probably be seek bound not throughput limited.
The article is kind of stating the blindly obvious- if you are seek bound SSDs are better. And 60 drives gives ~6000 seeks. A typical modernish desktop HDD can get in the order of 100MB/s data transfer (average sustained), more expensive HDD can get quite a lot more. If we take 3.0Gb/s as a ceiling (i.e. the SATA 3.0 max transfer rate) then at 6000 seeks/second you are getting 3000MB/6000seeks=0.5MB per seek. So the result makes perfect sense if you looking up data that is either entirely non-contigous or smaller than 500kB- an SSD will beat you every time on seeks (since it has no seek time).
The limitations on SSDs are: they have throughput limitations, just like HDD and more importantly their write performance is usually significantly worse than a HDD (writing on an SSD often involves reading and re-writing large chunks of data, even for very small writes). You can easily construct tests where HDD perform better than SSDs (particularly something like a 60 spindle array of HDD where an awful lot of writes can be cached in the on disk's ram buffer, which is common on hire performance drives- often battery backed so they can "guarantee" the write has been committed without having to wait for a write to the magnetic media).
Of course SSDs other obvious application are where you want robustness and silence, i.e. laptops. Oddly enough their power performance isn't that much better than a normal HDD (although that might have changed since I last read about it).
Say the cost is worth the wait, whats your point mr non article?
. . . a Storage Review experiment from over a year ago:
http://www.storagereview.com/western_digital_velociraptors_raid_ssd_alternative
They put WD Raptors in RAID 0 to form a high performance (yet still affordable) platter drive setup, and then faced them off against Western Digital's new (at the time, first) SSD. Makes sense, right? Except that WD's first SSD was a complete joke, an underperforming, laughably expensive POS that I forgot about a couple days after Anand's review. When I first read about it I couldn't help but think that WD was deliberately setting it up to fail. It was at the bottom of every benchmark yet priced higher than any other (MLC) SSD. They even put a jmicron controller in it for fuck's sake (not the infamous original one, but still . . .)! Storage Review's calling it a "mid-range" SSD is very generous at best.
Even so, this supposedly screaming platter drive setup could only occasionally hang with the bottom of the barrell of SSDs, and mostly lagged behind it. And as I said, this was over a year ago. It goes without saying that they didn't worry much about heat, noise, reliability (of RAID 0), or power consumption.
Anand doesn't even list platter drives in his benchmark results anymore because they'd skew the charts so badly.
As a previous poster said, a winning strategy is to get a SSD boot drive just big enough for your OS and programs, and use platter drives for everything else. And since the SSD takes care of your performance needs, you can get the cheapest, slowest, coolest, quietest platter drives. There are some cases where both high performance and high capacity are needed at once (like video editing) but they're not the norm.
There is no such thing as poor performance on an SSD, unless you allocate it poorly. The fact that most companies aren't paying attention to fantastic ways to get reasonably priced SSDs into their equipment just proves that hype is awesome and smart still sucks. Luckily for me, smart is still making money (though not nearly as much as hype).
When the foot seeks the place of the head, the line is crossed. Know your place. Keep your place. Be a shoe.
I moved a small 4TB database from 24x 256G 15k SAS drives to 24x 240G OCZ Vertex 3 SATA3 drives. I ran a few queries on the old and the new. same data, same parameters, same amount of data pulled. Both were hooked up via PCIe 8x slots.
the SSD crushed the SAS. Not just a mere 2x or 3x crushing. A _FIFTEEN TIMES FASTER_ crushing. This was pulling about a million rows out. 12 seconds (SSD) vs 189 seconds (spindles)
Cost difference? under $50 per drive more expensive for SSD. I think our actual rate was around $10 per drive more. However, the system as a whole (array+drives+computer) was $12k less. No contest... for our particular application, SSD hands down makes it actually work.
we'll be moving the larger database (same data, same function) to SSD as soon as we can.
-- Who is the bigger fool? The fool or the fool who follows him? --
I run 3 Vertex 2 120 GB SSD in raid 0. I set the array to only use 230 GB so each drive has extra OP area. I have a dedicated 500 GB HDD that I do weekly backups to and two 2 TB HDD for storage.
Been running this way for months and I will NEVER go back to spinners for a boot drive.
Seriously folks, it really is the filesystem (or similar high level layers that are cognizant of data) that is the core issue here.
If you don't have a filesystem that natively handles tiering, then you need to offload to something that does. Most enterprises are stuck with eating whatever NetApp serves them to make badass SAN's using NetApp's WAFL filesystem. Some cases, a point solution like offloading some caching and writes for specific applications to something fast makes sense, hence some of those PCIe cards that are doing native stuff (and not translating to SATA/SAS drives mounted on the card). The quick and sleazy solution for rigs that can use it are low end MLC SSD. Pay more for SLC SSD. If you feel rich, get some DRAM based stuff (DDRdrive, HyperDrive5, ZeusRAM). Hell, Marvel was pimping their Dragonfly hardware caching accelerator for VMware SAN storage (caching on the host server rather than caching within the SAN itself).
But if you want to keep getting better, you need to look higher. ZFS and BTFS look interesting. ZFS is arguably more mature, and you can get it via a number of routes (Oracle Solaris, Illumos, Nexenta, FreeBSD, hell even that FUSE plugin for linux or that native kernel driver LANL was going to release someday).
...when your selection criteria is cost-per-IOPS. For heavy, random IOs (like transactional databases) SSD presents a cheaper solution.
Anyway, that's my two-sentence rewrite of TFA.
People go to great lengths to minimize read/writes; putting their games and apps on a HDD, changing system/user folder locations to a HDD, etc. So basically is it worth it to merely load the operating system a little bit faster? Why buy an SSD if you aren't going to use it to load your games and apps faster?
I like the whine. Sort of like it was lighting up the tires on my '69 Chevelle. WHO NEEDS CHEERLEADERS NOW, HUH!!!?
I run RF simulations at work and loaded up the RAM to cache terrain data.
64GB system memory was $1500 a few months ago. I think it is below $1000 now.
Skip the SSD and load up on ram.
Once it's cached, leave it there.
I'd like to see a thunderbolt RAM drive .. that'd be something; in my youth I'd have given it a go. Put some backup batteries in there, a mirror HD, and voila - block reads to go, in bulk. Sweet little capacitors. Do my bidding!
..don't panic
I would like to see the performance gain in eve. This could reduce lag significantly.
Who is this that even the wind and the waves obey Him? Surely this computer must submit also!
http://www.ocztechnology.com/aboutocz/press/2011/448 $500 and is "face meltingly fast".
Computing and Programming Since 1975 The Best Kept Secret in Technical Support Master of the Bare Metal Clean Install
Looking up a contiguous block of data vs not looking up a contiguous block of data makes no difference to an SSD.
Then is it safe to say that we should never again worry about running defrag utilities after moving to SSD?
That that is is that that that that is not is not.
What is Soviet Russia? It used to be Soviet Union, and Russia is the major non-Commie successor state to that.
On "why does NetApp sell their PCIe NAND flash card $30k?", here is your answer, Chris Rima: http://blog.zorinaq.com/?e=37
In a 3 words: because NetApp can.
It's not the components or engineering behind the card that cost $30k. NetApp prices it so high because the card boosts the performance of their filers by about the same amount as a ~$50k shelf of SAS disks (click that link and go read NetApp's own marketing documentation). They have got to have price points that make sense to customers.
(I know a fraction of you will think "No way!". Well, arbitrary price markups on enterprise gear do exist. This NetApp Flash Cache is effectively priced $150/GB. How do you think that certain competitors can even sell _enterprise_ flash at well below $10/GB? We are not talking 25 or 50% less, but a whole order of magnitude less expensive!)
... solid state devices have a seek time near 0 compared to spinning drives!
news at eleven ...
Are you stupid? Russia was the largest constituent of the Soviet Union, so 'Soviet Russia' simply means Russia that belonged to the Soviet Union, as opposed to the Russia that existed before the formation or after the dissolution of the Soviet Union.
Duh?
Has slashdot reset to a different perception of technology?
The youngsters know this, the elders know this. I ask the, who on slashdot does not?
Hivemind harvest in progress..
Early this year, Dell gave our office an M1220 array filled with SSDs (15 of them). To benchmark it, we put it up against an M1220 that we already owned filled with 10K SAS HDDs. Each was plugged into an "identical" (as far as any two individual computers can be identical) machine, we loaded the servers with a fresh Linux install, and booted with a 1G RAM ceiling to avoid RAM caching issues. Same filesystem type and size on both storage arrays. We spent a weekend just writing /dev/zero to the SSD array so that we wouldn't run into any of the "first write goes real fast" phenomenon that sometimes happens.
And then ran iozone3 against both, up to 4GB file sizes (once again, to take RAM caching out of play).
Where the SSD did better, the difference wasn't enough to justify either the initial cost difference or the maintenance cost. Where the SSD did worse, in some cases it did MUCH worse.
We called Dell with the results. We sent them our testing methodology. They first told us that we had to put a specific firmware version on the RAID controller and make a couple of changes to the default controller BIOS settings, to trigger some custom code on the SSD on-board controllers. We tried. No difference.
They've got a lab where they do nothing except benchmark hardware 40 hours a week. We invited them to find a configuration -- no matter how contrived and impractical for real-world usage -- that would demonstrate measurable performance advantages for the SSDs.
Six months later, they called us back and admitted that they couldn't find a way to make the SSDs look better.
We'll wait for the second generation.
We can believe in you for 3 minutes, but beyond that, even the King of All Cosmos can't be expected to wait.
Hi,
I'm just curious what database server are you using in your system. Can you elaborate a bit more, what OS, what usage patterns (data warehouse/web-mostly read/transactions/etc) are dominant in your system?
--Coder
As IT in a small healthcare business, we got to the point about 2 years ago that we had more data than we could use with our budget. I began moving our databases over to RAID 10 arrays of consumer intel 160gb drives. Speeds went up by a factor of 20, cost was LESS than any sas SAN I could find, and I haven't had a single drive drop out yet. The trick isn't to use the most expensive part, it's to have redundancy and backups.
18 months ago, I moved my VM hosts to new machines (2P amd HP's) with the same strategy, using the built in battery backed up cache controllers running arrays of internal SSD's. Until we get to be a MUCH bigger company, this will suit us just fine for MANY years to come.
My personal mission is to eliminate any spinning media that's not used for archival or backup.
Reading books is passive too, hardly interactive or group activity.
So you also never ever watch any documentaries? no historical docos?
Did you read about 911 on 912?
And porn on a tv is interactive and hardly passive ;-)
Liberty freedom are no1, not dicks in suits.