November 5th, 2009

A 4 SSD array: Apricorn pt 2

Posted by Robin Harris @ 9:28 pm

Categories: Disk drives, RAID, Solid State Disk

Tags: Card, Apricorn, Intel Corp., Performance Management, Human Resources, Workforce Management, Robin Harris

In part I of this review I measured the bandwidth of a hard drive version of the Apricorn hard card. That’s a PCI express card with 4 2.5″ hard drives mounted on it.

But a good solid state drive (SSD) is faster than a disk - so how fast is the Apricorn card with 4 Intel X25-M drives? I won’t keep you in suspense: darn fast.

Fast enough to handle full 12-bit RGB 4:4:4 at film’s 24 frames-per-second rate. I striped the 4 SSDs with the Mac’s Disk Utility software RAID 0 with the default 32KB block size, which is about the fastest. I used the Blackmagic Designs Disk Speed Test - which is close to X-bench results.

Fast enough for uncompressed 24fps video.

The X25-M uses the slower multi-level cell (MLC) flash, while the X25-E uses faster - and more expensive - single-level cell NAND flash. While both have impressive 260 MB/sec read speeds, M version writes - 70 MB/sec - are slower than an empty 2 TB Seagate hard drive at 117 MB/sec. Of course, the hard drive will slow down as it fills up and the flash drive shouldn’t.

Apricorn tested the card with 4 X25-E drives and reported over 675 MB/sec writes speeds. I didn’t have that config to test, but their other numbers have been similar to my test results.

For Macs Apricorn offers a new version of the driver that they’ve tested with Snow Leopard. I didn’t try it as I’m in the middle of a couple of jobs but I believe them.

The Storage Bits take
I’m currently testing a couple of external arrays, and while they have their advantages, internal system storage is less cluttered with fewer connectors to cause problems as well as much faster read performance - and writes, with the X25-E.

The nearest competitor is Fusion-io’s ioXtreme card. At 80 GB for $895 it offers comparable read performance and faster write performance - at least on Windows and Linux machines - no Mac driver is available.

The Apricorn supports 4 80 GB X25-M drives to achieve similar performance. At roughly $250 each, a fully configured Apricorn costs $1200 for 320 GB. You do the math. For uncompressed video the extra capacity will be helpful.

Comments welcome, of course. I’ll ship the gear Apricorn loaned me back this week. It was fun while it lasted! I’ve also done work for Fusion-io, but have they sent me review copies of anything? No.

September 1st, 2009

Build a RAID 6 array for $100/TB

Posted by Robin Harris @ 10:36 am

Categories: Disk drives, Infrastructure, RAID, Software

Tags: RAID, Storage, Hardware, Robin Harris

Imagine storage that didn’t cost much more than bare drives. High density storage with RAID 6 - double-fault - protection, reasonable bandwidth and web-friendly HTTPS access.

And really, really cheap.

Not your enterprise’s RAID array
Raw disk cost is only 5-10% of an enterprise RAID system’s cost. The rest goes for corporate jets, sales commissions, tradeshows, sheetmetal, 2 Intel x86 mobos, obscene profits and some pale and blinking engineers in a windowless lab who make it work.

But what if you don’t want 4-color brochures or the barely-clad booth babes. What if you just want cheap economical and reliable storage?

You aren’t running the global financial system - what’s left of it anyway - or a 500 person call center. But you want enough redundancy so it will stay up until morning.

Meet the Storage Pod
You aren’t the only one. Backblaze, a new online backup provider, designed the Storage Pod for their own use and are sharing it with everyone. They aren’t selling it - that’s where the build comes from - so they aren’t trying to get rich off you.

Here’s what it looks like:

Here’s an exploded diagram with a simplified BOM:

And then there’s the (free) software. 64-bit Debian Linux, IBM’s open source JFS file system and HTTPS access. Simply stated each file gets a URL. Put a web server in front of it and serve the world - or just your home.

The Storage Bits take
Many applications just need a big bucket that doesn’t cost $5,000/TB. This is it.

You can build it yourself, but it is probably more complex than a high-end gamer system. Download the 3D SolidWorks files and have Protocase build you 1 or 500 of the boxes.

But the density is good, the performance is reasonable, the availability is decent and the price is right. This is a DC-3, not a 747. It is all you need for the right application.

And at $100/TB you can mirror all your data 2 or 3 or 4 times if you need more availability - and still be way less than half the cost of name brand arrays. Get the details from the Backblaze blog.

Comments welcome, of course. BTW, I’m trying out their free trial Mac online backup - at least one of the founders worked at Apple - and I’ll let you know how it goes. I don’t have a business relationship with them either, in case you’re wondering.

August 28th, 2009

Apple kicks ZFS in the butt

Posted by Robin Harris @ 9:02 pm

Categories: RAID, Software

Tags: File System, Apple Inc., Sun-developed ZFS, ZFS, Robin Harris

It’s official: ZFS - a kick-butt file system - is nowhere to be seen in the latest release of Mac OS X, Snow Leopard. Even though it appeared in 10.5 Server, and was expected to become the default file system at some point, Apple has abandoned the Sun-developed ZFS, the first 21st century file system.

A bummer for anyone who stores data on their computer.

Why should I care?
Apple is hoping you don’t - and they’re probably right. None of the mainstream press have mentioned dropped feature, even though it is right up there with parallel processing support as a winner for users.

ZFS combines a file system and a volume manager, along with some cool architectural features, to create an easily managed and highly reliable file system. Advanced features that just work.

Some cool features.

• Manage storage, not disks. You can put all your disks in a pool and specify the redundancy level. ZFS takes care of the rest.
• No more silent data corruption.Wonky things can happen to your data to and from a disk. ZFS checksums every file before it is written and stores the checksum on the parent. When the file is read, the checksum tells the filesystem if that is the block it wrote.
• Easy snapshots. Ever wish you could roll back to a known good state? Snapshots make that easy and ZFS makes snapshots easy.
• High performance software RAID built-in. Worried about protecting your data. ZFS provides strong RAID capabilities without adding hardware.
• Transparent compression on the fly. Save capacity by compressing old and/or large files automagically.

What happened?
2 years ago it looked like ZFS was locked in to Snow Leopard. The Apple team was working with the Sun ZFS team. It was enabled as a read-only file system on 10.5 server. Apple even freakin’ announced ZFS on Snow Leopard. The advantages - to storage geeks - were obvious.

Plus the opportunity to put daylight between OS X and Windows 7. Microsoft’s ambitions for something called WinFS crashed to earth 3 years ago (see Bring me the head of WinFS.

But Apple started walking back ZFS about 9 months ago. Newer builds of Snow Leopard had less and less ZFS content until today’s official release - which has none.

Maybe some insight will emerge from secretive Apple, but don’t count on it. Removing ZFS from the server edition, where it makes even more obvious sense, suggests it is gone for good.

What did it in? Maybe it was a schedule problem - file systems require a lot of testing - and rewriting all the other bits took precedence. NIH - Not Invented Here - syndrome is another possibility. Or perhaps the uncertainty of Sun’s future led Apple to pull back.

Or maybe they just decided customers wouldn’t know enough to care, so why bother? Whatever the reason it is a major step backwards for the PC industry.

The Storage Bits take
File systems are essential but unsexy plumbing. Whether it’s a missing or corrupted file or a system slowing to a crawl because the directory is bloated, there is no error message that says “Your FS is screwed up.”

And as noted in How Microsoft puts your data at risk - which indicted Apple’s HFS+ as well -

. . . more than half of all data loss is caused by system and hardware problems. A high quality file system that took better care of our data could eliminate many of those failures.

The industry knows how to fix the problems. The question is when. With a resurgent Mac pushing ZFS maybe Redmond will see the light sooner, rather than later, and dramatically increase the reliability of all our systems.

With Apple’s retreat from ZFS everyone who uses a personal computer is the loser. Maybe the Microsoft team working to improve NTFS will now take the lead in file system quality and feature.

Comments welcome, of course. Update: I got some more theories over the weekend on why the ZFS deal fell through. Check them out on StorageMojo. The short answer: licensing; GPL vs CDDL. End update.

May 19th, 2009

RAIDfail: Don't use RAID 5 on small arrays

Posted by Robin Harris @ 11:14 am

Categories: Disk drives, RAID

Tags: Data, RAID 5, RAID, Storage, Hardware, Robin Harris

Big storage companies stopped recommending RAID 5 a couple of years ago. But I still see small 4-drive arrays touting RAID 5 for home and small office use.

Big mistake. You want to save money, but you also want to keep your data. RAID 5 isn’t worth it.

What’s the problem?
The problem is that RAID 5 only protects against a single disk failure. But SATA drives are spec’d at one Unrecoverable Read Error (URE) every ~12.5 TB.

Let’s do the math.

In a small 4 drive array using 2 TB disks, if you lose a disk you have 6 TB - 3 drives - of remaining capacity. That includes the parity data used to reconstruct the data lost on the failed drive.

Reading through that 6 TB you have a better than 40% chance of encountering an URE - and at that point the disk rebuild will stop since the RAID controller doesn’t have the information it needs to reconstruct your data.

Then you pull out your backup copies. You have backups, right?

How to use a small RAID array.
4-drive arrays have lots of advantages: cost; performance (with FireWire or eSATA) fast enough for HD video editing; and portability.

But if you care about your data, RAID 5 is too big a threat. And if you don’t mind risking your data - as in performance driven apps like video editing where the data copies are on tape or another disk - RAID 0 (striping) is cheaper and faster.

Most small arrays come with a RAID 1 (mirroring) option that copies your data to 2 different disks. Lose 1 and the other should have it - subject to the occasional URE.

If you want availability and better performance use RAID 1+0 - often abbreviated RAID 10 - which combines mirroring and striping to provide 2 complete copies of your data with the performance of 2 striped drives.

The Storage Bits take
The attraction of RAID 5 is that it gives you 3 drives worth of capacity on a 4 drive array - but at the cost of having to use backups if an URE is encountered. Better to use RAID 1 and get 2/3rds the capacity of RAID 5 with a much lower chance of data loss.

The biggest storage mistake consumers make is to believe that any storage device is 100% safe. It isn’t.

Maintain at least 2 copies of any data you value. If the data is vital, make that 3 copies. And if thinking about RAID levels makes your teeth ache, consider a Drobo or the new Drobo Pro.

Storage is cheap. Use lots.

Comments welcome, of course. Check out an earlier post Why RAID 5 stops working in 2009 for more details on the RAID 5 problem.

April 9th, 2009

Better than RAID: the new DroboPro

Posted by Robin Harris @ 8:48 pm

Categories: Disk drives, RAID

Tags: Drobo, RAID, Storage, Hardware, Robin Harris

RAID is a bad idea for home users since its management complexity and failure modes create more problems than it solves. But protection from 1 - or better yet 2 - drive failures is a very GOOD idea.

Data protection for the rest of us
Drobo is a product that protects against drive failure like RAID does without the headaches of traditional RAID arrays. Key features:

• Use drives of any capacity. Drobo will figure out how it can safely store and provide that.
• Add new drives as needed. As a corollary you can add or replace drives with new, larger drives to increase capacity or to replace flaky old drives.
• Helps the RAID 5 problem.Drobo is less prone to the growing RAID 5 problem since it only reads data blocks rebuilding after a drive failure. Normal arrays don’t know which blocks have data so a 2nd URE kills the array.
• Faster rebuild with less overhead. Traditional arrays have to read from all drives concurrently while calculating parity to rebuild data - which means the array runs slowly for hours. Drobo’s system only copies needed data to a new drive - a faster, lower overhead operation.
• Set and forget. Drobo just works: it looks like a drive to your system; when a drive fails a red light tells you which drive to replace.

Let’s roll the (virtual) tape
In this video a 3 drive DroboPro suffers a 2 drive failure - without affecting a video playback. It’s a minute long intro to an amazing product.

The Storage Bits take
The human error rate on traditional RAID arrays is shockingly high with estimates ranging from 3% to 10%. And that’s with pro admins.

Civilians don’t have a chance with standard arrays. Small businesses will find the DroboPro has all the capacity and performance they need for local storage.

For disaster recovery use a cloud backup vendor - or better yet, 2 of them. With the low-cost solutions available today no small business should ever suffer catastrophic data loss.

Comments welcome, of course. Note that the 4 slot Drobo won’t handle 2 drive failures at once. And no, I’m not on the Drobo’s payroll - I just really like the product.

March 9th, 2009

How Amazon builds the world's most scalable storage

Posted by Robin Harris @ 10:19 am

Categories: Clusters, Infrastructure, RAID

Tags: Data Center, Amazon.com Inc., Cloud Storage Market, Humans, Data Centers, Storage, Hardware, Data Management, Robin Harris

The cloud storage market is accelerating fast - despite naysayers and alarmists - and Amazon’s S3 is leading the charge. Storing over 40 billion files for 400,000 customers Amazon is the one to beat. How do they it for pennies per GB a month? Read on.

I attended FAST ‘09, the best storage conference around, where Alyssa Henry, S3’s GM, gave a keynote. Amazon doesn’t talk much about how their technology works, so even the little Alyssa added was welcome.

Aggressive goals
A multi-billion dollar business running one of the world’s largest websites, Amazon engineers understand the problem. Their goals reflect both technical and market requirements:

• Durable
• 99.99 availability
• Support an unlimited number of web scale apps
• Use scale as an advantage - linear scalability
• Vendors won’t engineer for the 1% - only the the 80% - DIY
• Secure
• Fast
• Straightforward APIs
• Few concepts to learn
• AWS handles partitioning - not customers
• Cost-effective

One key: Amazon writes the software and builds massive scale on commodity boxes. Reliability at low cost achieved through engineering, experience and scale.

With many components come many failures
10,000+ node clusters mean failures happen frequently - even unlikely events happen.

• Disk drives fail
• Power and cooling failure
• Corrupted packets
• Techs pull live fiber
• Bits rot
• Natural disasters

Amazon’s deals with failure with a few basic techniques:

-Redundancy
Increases durability, availability, cost, and complexity. Example: plan for the catastrophic loss of entire data center; store data in multiple data centers.

Expensive but once paid for costly small-scale features like RAID aren’t needed.

-Retry
Just like disk drives - it’s quicker for Amazon to retry than it is for customers. Leverage redundancy - retry from different copies.

-Idempotency
This is cool. An idempotent actions result doesn’t change even if the action is repeated - so there’s no harm in doing it twice if the response is too slow.

For example, reading a customer record can be repeated without changing the result. So they don’t retry too much there’s surge protection.

-Surge protection
Rate limiting is a bad idea - build the infrastructure to handle uncertainty. Don’t burden already stressed components with retries. Don’t let a few customers bring down the system.

Surge management techniques include exponential back off (like CSMA/CD) and caching TTL (time to live) extensions.

-Eventual consistency
Amazon sacrifices some consistency for availability. And sacrifices some availability for durability.

-Routine failure
Everything fails - so every failure handling code path must work. Avoid unused/rarely used code paths since they are likely to be buggy.

Amazon routinely fails disks, servers, data centers. For data center maintenance they just turn the data center off to exercise the recovery system.

-Diversity
Monocultures are risky. For software there is version diversity: they engineer systems so different versions are compatible.

Likewise with hardware. One lot of drives from a vendor all failed. A shipment of faulty power cords. Correlated failures happen

Diversity of workloads: interleave customer workloads for load balancing.

-Integrity checking
Identify corruption inbound, outbound, at rest. Store checksums and compare at read - plus scan all the data at rest as a background task.

-Strong instrumentation
Internal, external. Real time, historical. Per host, aggregate. When things go wrong, you need history to see why.

-Get people out of the loop
Human processes fail. Humans are slow. If a human screws up an Amazon system, don’t blame the human. It’s the system.

Final thought
Storage is a lasting relationship that requires trust.

The Storage Bits take
Amazon is the world leader in scale out system engineering. Google may have led the way, but the necessity to count money and ship products set a higher bar for Amazon.

Amazon Web Services will dwarf their products business within a decade. I’d like to see them open the kimono more in the future.

Comments welcome, of course. There’s a longer version of this on StorageMojo. And there’s the Amazon CompSci paper Dynamo: Amazon’s Highly Available Key-value Store. Not S3 specific, but close.

January 12th, 2009

Best storage at CES 2009

Posted by Robin Harris @ 9:19 am

Categories: Disk drives, RAID, Software, Solid State Disk

Tags: Apple Macintosh, Storage, Backups, Hardware, Robin Harris

Data storage is a necessary part of the digital lifestyle. The good news: the hardware is getting more stylish and the software easier to use.

Backup
The best news: Windows backup has improved markedly in the last year. Storage Appliance Corporation’s Clickfree line of Windows and Mac backup is the simplest and easiest to use Windows backup I’ve ever seen.

Simply plug in the Clickfree drive or USB cable and the system automatically starts backing up user files: documents, photos, video, whatever. A single drive will back up as many as 10 Windows machines. Even simpler than BackupKey’s lower-cost 1-click system!

I tried their Mac support on the CES show floor. Other than having to reformat the drive to a Mac HFS plus file system and typing in the volume name it worked as advertised. Not as easy or elegant as the built in Mac Time machine, but if you need to back up multiple Macs it would be more cost effective.

The problem with backing up user files is that restore can be a lengthy process. Not a problem for casual users but if you rely on your system for business use a full bootable system disk backup will get you online again in minutes rather than hours.

That is easy on a Mac with either Carbon Copy Cloner or SuperDuper!. It hasn’t been so easy on a Windows machine.

That’s why I was pleased to see CMS products introduced BounceBack Pro. This $70 download creates bootable backups to external USB drives so if your system drive fails you can start up with everything just like it was when you last backed up.

Set it to backup nightly in you will never lose more than a day’s work. I hope to have a more complete review when CMS sends me a review copy.

Solid state disks
Third-generation SSD random write performance is reaching acceptable levels. I saw a SanDisk SSD doing 350-400 4k random writes per second. While the power and weight savings are insignificant and the durability still uncertain, at least the awful performance of the first two generations of consumer SSDs appears to be receding into the past.

External hard drives
Backup will never be sexy, but a cool slim external drive doesn’t hurt. Imation was showing several drives with nifty built-in feet so the drive could be used either horizontally or vertically.

Iomega was showing consumer hard drives in attractive colors and with slip on silicon sleeves for better shock resistance. They were all so showing LifeLine software from new corporate parent, the $14 billion EMC, that adds RAID, backup, active directory support, Bluetooth support, browser based management, CIFS, NFS and FTP protocol support, a print server and useful power management.

The biggest surprise is that no one appears to be following Drobo’s lead with a truly easy to use prosumer storage device. CEO Geoff Barrall said they shipped over 40,000 units in 2007, their first full year of operation.

Now that they’ve added FireWire 800 support it is the only external protected storage array I recommend to storage civilians. For most people RAID systems are more trouble than they’re worth.

The Storage Bits take
As the Digital Age permeates every day life, the problems with digital storage become more acute. While digital storage, access and protection is nowhere near as automatic and seamless as it should be, the industry is making solid progress getting appropriate tools and products in place for consumers.

Comments welcome, of course. I actually saw most of these products at Storage Visions, a consumer storage industry show just prior to CES. Recommended for consumer storage professionals and companies.

December 4th, 2008

Seagate's "random freeze" problem: worse than reported?

Posted by Robin Harris @ 4:54 pm

Categories: Disk drives, RAID

Tags: Seagate Technology LLC, RAID, Storage, Hardware, Robin Harris

Reports that Seagate is having “random freeze” problems on its new 1.5 TB drives may be more serious than the company has admitted: 1 TB Seagate drives may be affected as well.

Seagate on the case
Seagate spokesman Mike Hall wrote:

Seagate is investigating an issue where a small number of Barracuda 7200.11 (1.5TB SATA) hard drives randomly pause or hang for up to several seconds during certain write operations. This does not result in data loss nor does it impact the reliability of the drive but is an inconvenience to the user that we are working to resolve with an upgradeable firmware.

Not great but OK if true. But then I got a note from a senior engineer at a cloud storage company:

What seagate doesn’t seem to be admitting is that their 7200.11 1.0 TB drives have that same problem also.  We’ve had to remove from service and rebuild 2 expansion servers with overnighted Samsung drives in the last 48 hours . . . .

Pulling servers and replacing all the drives is more than an “inconvenience.” It is a major operational problem.

It can kill your desktop RAID
A Tech Report commenter noted that this problem can kill a desktop RAID. How?

. . . the problem may cause a RAID system to think the drive has died. The RAID system automatically removes the drive and continues to run degraded (as designed). 20 minutes later when another drive exhibits the problem the RAID system drops the second drive and dies.

Your data is in there somewhere, but good luck ever seeing it again. Yet another reason to avoid desktop RAID unless you a) know what you are doing and b) absolutely need it.

The Storage Bits take
Other than airfreighting Samsung or Western Digital drives, what can you do? First, check for firmware revisions that Seagate has admitted have problems: SD15, SD17, or SD18. If you don’t have them you should be OK.

If you do, what then? Cross your fingers and hope for the best. If they are in a critical RAID system you should make sure your backups are current and complete.

You may need them.

Comments welcome, of course.

June 19th, 2008

Apple announces ZFS on Snow Leopard

Posted by Robin Harris @ 11:24 pm

Categories: Infrastructure, RAID, Software

Tags: Disk, File System, Apple Inc., Microsoft Corp., Sun ZFS, Checksum, RAID, Storage, Hardware, Robin Harris

Finally, a modern file system on a consumer OS
As if Grand Central weren’t enough bad news for Microsoft, now they have ZFS to contend with. Building a reliable, high-performance file system takes years and Microsoft doesn’t have years to respond.

The formal announcement is for Snow Leopard server, which is how Apple introduces new file systems. HFS+ first arrived on a server version as well.

Who cares?
Anyone who stores data should.

Microsoft’s NTFS is 20 year old technology borrowed from DEC. Fine for small disks and puny CPUs. Not so great for today’s data intensive systems and applications.

Silent data corruption is common - only you don’t know it - because the corruption shows up as other problems, like missing DLLs.

ZFS: open source from Sun
ZFS is the first desktop file system with true end-to-end data integrity. Thanks to sophisticated tree-based checksums it detects and corrects silent data corruption anywhere in the data path: disks, cables, interfaces and more.

The checksums are stored with the parent block, so the file system always knows that the child block is both uncorrupted and the correct block. That’s just one of the errors that NTFS and most other commodity file systems - including the Mac’s HFS+ - are prone too.

Sun’s ZFS engineering team started working on ZFS 7 years ago as a clean-sheet design. It combines file system and volume management functionality. Instead of managing individual disks, you manage a pool of blocks. ZFS takes care of the details.

Turning up the heat on Microsoft
For all of Microsoft’s fine talk about innovation they don’t do squat unless someone else does it first. Remember IE 6? ZFS is a modern and innovative file system that solves some difficult data storage and integrity problems. Like these:

No more Disk Warrior
Data corruption on PCs and Macs is a sad and stupid fact of life. Power failures, flaky RAM, poor grounding, (slowly) failing hard drives, driver glitches, phantom writes and more conspire to rot your data.

ZFS eliminates that. All blocks are checksummed and the checksum is stored in a parent block. ZFS always knows if the block is correct and/or corrupt. Every block has a parent block (with one obvious exception that gets special treatment), so the entire data store is self-validating. You’ll never have to wonder if all your data is correct again. It is.

No RAID cards or controllers
ZFS implements very fast RAID that fixes the performance knock-off against software RAID. In ZFS all writes are the fastest kind: full stripe writes. And the RAID is running on the fastest processor in your system (your Mac), rather than some 3-5 year old microcontroller.

Just add drives to your system and you have a fast RAID system. With Serial Attach SCSI and SATA drives you’ll pay for the drives (cheap and getting cheaper), cables and enclosures.

No more volumes
Every time you add a disk to your Mac you see another disk icon on the desktop. If you want to RAID some disks you use Disk Utility (or something) to create the volume. Slow, error-prone, confusing.

ZFS eliminates the whole volume concept. Add a disk or five to your system and it joins your storage pool. More capacity. Not more management.

Backup made easy
ZFS does something called snapshot copy, which creates a copy of all your data at whatever point in time you want. Copy the snapshot up to a disk, tape or NAS box and you are backed up.

Create a snapshot on every write if you want, so if your database barfs you can go back to just before it choked.

But that’s not all!
For in-depth treatment of ZFS see here and here. Includes links to more technical info and benchmarks.

The Storage Bits take
It would be nice if Microsoft were driving innovation and reliability, but - like General Motors - they prefer to rest of their laurels. And like General Motors, they are facing a long and painful decline if they don’t get their act together.

GM says they are proud that 1 in 4 cars sold in America are GM - but the number used to be 3 out of 5. Microsoft is rightfully proud of their 90% market share. But that share can change - as it has for IE - and they have nowhere to go but down.

As users we benefit from the competition. Kudos to Apple for bringing the latest technology to consumers.

Comments welcome, of course. For more background on data corruption issues check out 50 ways to lose your data, How data gets lost and How Microsoft puts your data at risk.

April 14th, 2008

2.5" disks to become new standard in 2009

Posted by Robin Harris @ 12:12 pm

Categories: Disk drives, Infrastructure, RAID

Tags: Disk, Delta, $/GB, $/GB Delta, Atrato, Storage, Hardware, Robin Harris

When I first started selling storage, 14″ disk drives were all the rage. Then came 9″, 8″, 5.25″ and then 3.5″ drives - where we’ve been stuck for the last 15 years - even though they have gotten thinner.

All this shrinkage is courtesy of drive vendor’s incredible engineering work. You don’t know it - and engineering-dominated vendors hardly think to talk about it - but disk platter feature sizes are smaller than Intel’s latest 45 nm fabs can produce.

And Intel’s chips aren’t rotating 250 times per second, either.

A Lotus Elan vs a Bentley Type-R
Disks are mechanical devices. Making them smaller as magnetic densities increase has many benefits:

• Smaller components have less mass, so motors and actuators can be smaller. Greater shock resistance, less damping and head settling time too.
• Greater disk I/O density per cubic inch - get more work done in less space.
• Power, cooling and packaging are all smaller too. Even though data centers don’t much care about green computing, it is a nice added benefit.

Translation: the Lotus corners better than the Bentley and uses a lot fewer resources in the process.

The industry is quietly preparing
3.5″ drives will be with us for years to come, but the signs of the coming change are unmistakable:

• 3 years ago the capacity difference between volume 2.5″ and 3.5″ drives was 4-5x. 160 GB vs 30-40GB. The delta has come down to 2-3x: 1 TB vs 320 GB - and soon 500 GB.
• High-performance 2.5″ drives, both 10k and 15k, with high-end FC and SAS interfaces are now available.
• The $/GB delta is also dropping. They don’t have to be equal because the lower power, cooling, size and greater durabililty translate into real economic benefits for all but the most price-sensitive users.

The first 2.5″ arrays are appearing
Atrato, Xiotech and RAID Inc. have all announced 2.5″-based storage arrays that offer some compelling advantages over 3.5″ arrays.

• Atrato puts 160 spindles into a 3 rack unit box
• RAID Inc puts 12 15k drives into 1 rack unit pizza box
• Xiotech doubles the number of spindles - with the same power and cooling - over their 3.5″ version of their new ISE

The Storage Bits take
Smaller disks and smaller arrays are Very Good Things. We get more I/Os for a given power, cooling and volume investment. The drives get more reliable. The $/GB goes up, but volume production soon pushes it back down.

So I’m calling it: by the end of 2009 most new storage arrays will be announcing with 2.5″ drives. High-end workstations, like the next-gen Mac Pro, will be both smaller while containing more drive bays.

Low-end consumer desktops will be the last to transition, but even they will by the end of 2011. The last 3.5″ disk will roll off the assembly line in 2014.

Remember, you heard it here first.

Comments welcome, of course.