"SSD week" recap

June 18, 2010 – I don’t think it’s purely coincidental, but there was an uncharacteristic amount of rapid-fire announcements related to solid-state disk (SSD) drives this week. Three SSD startups launched (with two based in Israel). There was one SSD-related acquisition. And one vendor trumpeted the now familiar tune, “MLC SSDs are as good as SLC SSDs.”

In short, four interesting SSD announcements in the course of three days. And all four articles covering those announcements made the Most Popular Articles list on infostor.com this week. The industry analysts must be right about increased end-user interest in SSDs, although 39% of infostor.com visitors have no plans for using SSDs because they’re still too expensive (see below).

A quick recap:

At the high end of the announcements, Kaminario launched with the introduction of the K2 (meant to conjure up an image of the mountain, not the skis). See “Startup claims 1.5 million IOPS on RAID array.” An entry-level configuration starts at $200,000 for 300,000 IOPS and 1TB of DRAM, but the system can scale to millions of dollars and IOPS. Since its system is based DRAM technology, Kaminario will compete most directly with vendors such as Texas Memory Systems (which has been selling DRAM devices since disco was popular) and Violin Memory.

Speaking of which, Violin bought Gear6 this week. You might (but probably not) know Gear6 as a supplier of a souped-up Memcached distribution, but that’s not why Violin plucked them. Gear6 also has NFS caching software. If Violin can orchestrate that code into its SSD arrays it will give Violin an entry into the lucrative NAS acceleration market (see “Violin to acquire Gear6 for caching software”).

Anobit (add another bit) was another SSD startup that launched this week, with an MLC-based array that the company says provides the performance and reliability (“endurance” in SSD parlance) of more expensive SLC SSDs. You’ll see a lot of similar claims in the months ahead, as the mathematicians at SSD vendors have come up with a variety of tricks to mask the inherent drawbacks of inexpensive MLC flash NAND technology. Anobit’s secret sauce is its Memory Signal Processing technology (see “Anobit claims MLC SSDs rival SLC SSDs”).

And on the PCIe front, Virident launched its tachIOn SSDs. (Weirdly enough, when you type “tachIOn” in Word it automatically gets changed to “tachyon” which, by the way, is a hypothetical subatomic particle that moves faster than light, although most of us old-timers think of it as a Fibre Channel controller chip). Virident is going head-to-head with Fusion-io, essentially trying to beat them at their own game (see "Virident ships SLC SSDs for PCIe").

Despite the rapidly falling prices of flash memory and SSDs, many end users still say SSDs are too expensive. In an infostor.com QuickVote survey conducted earlier this year, 24% of the respondents had planned to deploy enterprise-class SSDs in the first half of this year; another 26% planned to buy them in the second half of the year; but 49% had no plans to deploy SSDs, with 10% citing insufficient reliability and 39% saying SSDs were still too expensive.

If you’re really interested in SSDs, consider attending the Flash Memory Summit, August 17 – 19 in Santa Clara.

Startup’s SSDs top 1,000,000 IOPS

March 17, 2010 – According to independent test results, Pliant Technology’s Lightning Enterprise Flash Drives (EFDs), which are based on solid-state disk (SSD) technology, topped one million IOPS in a configuration with 16 drives.

Many storage vendors have issued press releases claiming performance of more than, or at least very close to, one million IOPS. Examples include Broadcom (press release), Emulex (press release), and Microsoft and Intel (blog post and video).

The reaction from most people (including me) would be “So what?” and/or “Who the heck needs one million I/Os per second?” It’s a legitimate response, because almost nobody needs that level of horsepower, even in the context of rapidly growing virtual server environments, multi-core processors, etc.

But that was also the response when 10GbE, 8Gbps Fibre Channel, 6Gbps SAS, etc. came onto the scene. And we know how those technologies are doing. The fact is, end users’ need for speed is insatiable, and if you don’t need it today – you will tomorrow.

The testing of Pliant’s SSDs was done by OakGate Technology, which supplies testing tools for IT technology providers.

As with any benchmark tests, it’s important to (a) have a grain of salt on hand and (b) look at the test configuration. The 1.1 million IOPS results were achieved with 16 drives and an 80/20 read/write ratio with 4KB block sizes.

The testing also revealed that that configuration can produce 2.3 million IOPS with a 100% read stream and 512-byte blocks.

Two ZT Systems Linux platforms with dual quad-core Xeon processors were used in the Lightning EFD tests, each running the OakGate FireOak test suite, which was controlled by a Java-based GUI running on Windows XP. Each system included four Adaptec 1045 SAS HBAs, with each HBA connected to two 3.5-inch, 300GB Lightning LS300 EFDs from Pliant.

For complete test details, see OakGate CEO and founder Bob Weisickle’s videos on YouTube:

· Benchmark 1: Real-world Workload
· Benchmark 2: Maximum I/O Rate
· Benchmark 3: Maximum Bandwidth
· Benchmarks 1-3 (complete version)

Guest blog: SSDs on the cusp of the next big storage trend – By Jerome M. Wendt, DCIG

December 30, 2009 -- Ever since I’ve been involved with storage, the topic of solid-state disk (SSD) drives has come up in passing from time to time. But my first real hint at how disruptive SSDs were going to be came when I was attending C-Drive, Compellant's annual user conference, in the spring of 2008.

One of the speakers, a Compellent user, was lamenting how difficult it was to optimize a disk array for database performance and how Compellent’s storage system had helped him in that area by striping data across the array. One of the Compellent presenters then asked the user if he would consider using SSDs to improve performance and the user responded positively, which I think caught the presenter off-guard.

The presenter then turned to the audience of end-users and asked them how many would consider implementing SSD technology. Well over 50% of them raised their hands, which again appeared to catch the presenter by surprise. At that point I knew SSD technology was on the cusp of becoming the next big thing in storage, and it was not long after that conference that Compellent announced support for SSDs.

Heading into 2010, SSDs are gaining end-user interest, as well as vendor hype. Already they’re being hailed as the end to high-performance disk drives, and some observers predict that SSDs will eventually replace all disk drives - possibly as soon as the end of this coming decade.

Whether or not that comes to pass remains to be seen. SSD technology is still maturing, and prices have to come down to the point where most businesses can justify and afford deployment.

There are three notable ways that SSDs were positioned for use in the enterprise during 2009, making SSDs my third major trend of 2009. [To read about the other two trends, see “2009: The Beginning of the Corporate Love Affair with Cloud Storage” and “Deduplication is the Big Success Story of 2009.” ]

First, most SSDs are being initially deployed on external networked storage systems. Vendors such as Compellent, Dot Hill, EMC, HDS, Pillar Data Systems and others have added support for SSDs to their storage systems or can virtualize external SSD systems (e.g., NetApp's support for Texas Memory Systems’ RamSan 500).

Installing SSD drives, from providers such as STEC or Pliant Technologies, in existing storage systems was certainly one of the fastest and easiest ways to bring SSDs to market in 2009. However, I’m already questioning how long this trend of putting SSDs into storage systems lasts before companies figure out that they’re not getting the full range of anticipated benefits.

This scenario may take two or three years to play out, but what I eventually see emerging as the problem with putting SSDs into storage systems is the storage network itself.

Most networked storage infrastructures in use today are Ethernet, Fibre Channel, or a mix of both. Unfortunately, neither one of these technologies - even the latest versions - currently offers sufficient bandwidth to deliver on SSD's potential.

One technology that seems best positioned to break this bottleneck is InfiniBand. The problem is that no one outside of the high performance computing (HPC) space seems ready to commit to this protocol or even take a serious look at it. However, the forthcoming network bandwidth logjam is already recognized within storage circles and, in talking with CTOs of a number of storage providers, some are already testing InfiniBand with their storage systems.

Second, SSDs are showing up in servers. In a trend that runs counter to trends over the last decade, storage is moving back inside servers again. This trend is being largely (and I would say almost exclusively) driven by Fusion-io. While I have written a number of blogs about Fusion-io's ioDrives in the past, and their disruptive nature, that’s only part of the story. (Fusion-io puts SSDs on PCI-Express cards that are placed inside of a server to achieve higher performance than what high-end storage systems can achieve.)

The rest of the story has to do with the elimination of cost and complexity. Many storage networks are still neither easy nor intuitive to manage, especially in high-end storage environments.

Fusion-io eliminates the need, and associated expense, to deploy any storage network at all. While its ioDrive reintroduces the inability to easily share unused, excess storage capacity with other servers, its design is so compelling (especially when viewed in the context of cloud computing and cloud storage architectures) that the ioDrive will be a force in the coming years - maybe more so than any other SSD architecture.

Third, SSDs are moving into the storage network. This approach is currently being championed by Dataram with its XcelaSAN product. What makes this approach so different from the other SSD architectures is that it’s not inside the server or the storage system, but resides in the storage network between the servers and back-end storage.

The argument that Dataram makes for this implementation of SSD is that approximately 5% of data on a storage system is active at any one time. However, most storage systems only have a fraction of that much cache. For instance, a disk array with 20TB of active data should in theory have about 1TB of cache in order to provide the read and write performance that high-end applications need.

A quick check will reveal that many midrange storage systems are lucky to offer a fraction of that, with most supporting in the range of 3GB to 5GB of cache. Dataram argues that by putting its XcelaSAN in front of an existing midrange storage system, it can act as an alternative to cache while turning a midrange storage system into a high-end solution at a fraction of the cost.

Not a bad idea, but there are some hurdles to overcome. First, Dataram needs to convince users that placing its device inline between the server and back-end storage is a good idea.

Another hurdle the company faces is from midrange storage system providers. Depending on how Dataram positions its product, why should vendors such as Compellent, Pillar and other midrange providers agree to play second fiddle to Dataram and sit behind its solution, especially if they are offering SSD in their own systems? Further, SSD is becoming lucrative, which means that the disk-array providers would be giving up the big margins to Dataram.

Another potential obstacle Dataram faces is from traditional network storage providers such as Brocade and Cisco. These guys never take kindly to anyone intruding on what they view as their turf, so if Dataram garners any momentum at all, expect these heavyweights to step in and try to snuff it out. So while I like both Dataram's architecture and argument as to why deploying SSD in the network makes sense, convincing the channel to sell it, end-users to buy it, midrange storage system providers to endorse it, and storage networking providers to leave them alone is going to be an uphill challenge.

All this said, SSD gained solid momentum and mind share among end-users in 2009 and is poised to emerge as a major trend in 2010. However, which of these architectures becomes the predominant one remains to be seen. In the near term (2010 to 2011) I’m placing my bets on storage system solutions, but until technologies such as those from Fusion-io reach critical mass, and others from vendors such as Dataram are tested in the market, the real impact of SSD is yet to come.

To read more of Jerome Wendt’s blogs, visit the DCIG web site.

MLC for enterprise SSDs: The controller angle

August 3, 2009 -- According to conventional wisdom, enterprise-class SSDs require single-level cell (SLC) technology, as opposed to the lower-cost multi-level cell (MLC) technology – both of which fall under the NAND flash memory category. That’s because, generally speaking, SLC has performance, reliability and endurance (longevity) advantages. However, SLC media is much more expensive than MLC media. (In terms of raw media, the cost difference is about 4x.) As such, a variety of vendors are working on ways to combine the advantages of SLC with the low cost of MLC, in many cases using MLC media with advanced software and/or hardware.

In my last post (see “WhipTail: Software solves MLC SSD issues” ) I looked at WhipTail Technologies’ software-based approach to this issue. Other SSD vendors, such as STEC, are addressing the issue primarily from the controller angle (which, of course, is very much a software issue as it involves firmware and algorithms).

I recently spoke to Scott Shadley, senior product manager at STEC. He identifies the following as the key areas that will require advancements if MLC is going to succeed in enterprise-class SSDs.

--Existing ECC algorithms in SSD devices need to be improved, because ECC requirements for MLC are much more stringent than for SLC.

--Vendors need to mitigate the wear issues with MLC, because there’s about a 10x difference in the number of recycles per data sheet that SLC can withstand vs. MLC. Many SSD vendors utilize wear-leveling techniques to mitigate this problem, but “the goal is to limit, or minimize, writes in general, which wear leveling does not address,” says Shadley.

“Vendors have to come up with new ways for the controller to handle, or manipulate, incoming data so that the SSD is only writing a minimal amount of times, either by controller caching, external caching, and/or algorithms that manipulate the data to minimize the amount of data stored on the media,” he explains.

--Shadley also notes that, because MLC is slower than SLC, enterprise-class SSDs based on MLC technology will require faster processors, more flash channels, better ways of accessing the flash on those channels, and the ability to take advantage of the new NAND interfaces. “The goal is to minimize the performance and cost differences between MLC and SLC,” he says.

I expect to see some announcements in this space at next week’s Flash Memory Summit, August 11—13 at the Santa Clara Convention Center.

For more information on SSDs, see InfoStor’s SSD Topic Center.

WhipTail: Software solves MLC SSD issues

July 29, 2009 – One of the more promising developments in the solid-state disk (SSD) drive space is the potential use of low-cost multi-level cell (MLC) NAND flash memory in enterprise applications and arrays, vs. the high-cost – but more reliable and durable – single-level cell (SLC) technology, as I mentioned in my previous post (see “Intel slashes SSD prices” ). This can basically be accomplished in two ways: via software or via controller enhancements.

Relative newcomer WhipTail Technologies is an example of a vendor that’s using software techniques to overcome some of the inherent limitations of MLC flash memory; namely, write amplification issues that limit the ability of NAND to perform random writes in an effective manner (a performance issue), and wear-out issues (a reliability, or durability, or endurance problem).

To address the performance part of the equation, WhipTail uses buffering (not caching) techniques, in which writes are aggregated into a buffer that’s sized to the erase block of the NAND, according to WhipTail CTO James Candelaria, who claims that this technique enables performance close to the performance specs of the NAND media.

Specifically, the company claims performance of more than 100,000 I/Os per second (IOPS) of sustained, random I/O with 4KB block sizes and a 70/30 read/write split. Other performance specs include a latency of 0.1 milliseconds, and bandwidth of 1.7GBps (internal to the chassis).

The other major problem with MLC flash is wear-out. For example, SLC is rated at about 100,000 cycles per cell, while MLC is rated at only 10,000 cycles/cell before the cell becomes unreliable (and that may go down to 2,000 to 4,000 writes/cell with smaller die sizes).

To address the wear-out issue, WhipTail uses a technique called linearization, which essentially entails writing forward across the disk and not revisiting blocks until the entire array has been utilized. This not only decreases wear on the media, but also increases performance. Working in conjunction with linearization, a defrag process ensures that there is always a minimum amount of free space. This technique also works in conjunction with the drive’s wear-leveling algorithms.

The company’s internal tests indicate that if you rewrite an entire array once a day, the device will last seven years (or longer than most other components in the storage hierarchy).

You can get the details on these two techniques on WhipTail’s web site, as well as details on their products, but what about pricing?

Candelaria contends that WhipTail “provides tier-0 [SSD] performance at the price of tier-1 arrays.”

Well, a 1.5TB WhipTail SSD array is priced at $46,000 retail; a 3TB version at $75,600; and the new 6TB configuration, introduced this month, at $122,500.

Summit NJ-based WhipTail was spun out of TheAdmins, a reseller, early this year and has been working on its SSD technology since late 2007. Its first product went GA in February. The company sells through resellers, with eight VARs signed up so far.

For more information on SSDs, see InfoStor’s SSD Topic Center.

And if you’re really interested in solid-state technology, consider attending the Flash Memory Summit, August 11—13 at the Santa Clara Convention Center.

Intel slashes SSD prices

July 23, 2009 –There are still some issues that need to be ironed out with solid-state disk (SSD) drives (e.g., reliability and endurance), but the biggest problem -- and gating factor to adoption -- has been the outrageous price of these devices.

One way to reduce prices is to use the less expensive multi-level cell (MLC) NAND flash technology, as opposed to the more expensive, reliable and durable single-level cell (SLC) technology. But at least for enterprise-class applications, that requires improvements in either controller and/or software technology (which I’ll blog about in an upcoming post).

Another way to reduce SSD prices is to go with a different manufacturing process. That’s what Intel announced this week for its X25-M (Mainstream) line of SSDs, which are admittedly designed primarily for desktops and laptops as opposed to enterprise arrays and applications.

Intel claims a 60% price reduction due to moving from a 50-nanometer manufacturing process to a 34nm process (smaller die size), and a quick price check seems to legitimize those claims.

For example, the 80GB X25-M SSD is channel-priced at $225 in 1,000-unit quantities, a 62% reduction from the original price of $595 a year ago. And the 160GB version is priced at $440, down from $945 when it was first introduced. Both of those SSDs come in a 2.5-inch form factor, with a 1.8-inch version, the X18-M, due in August or September.

Intel claims performance of “the same or better” compared to the 50nm predecessors, citing up to 6,600 I/Os per second (IOPS) on 4KB write operations, and up to 35,000 IOPS on read operations. The company also claims a 25% reduction in latency, to 65 microseconds.

Calculated on a cost-per-GB basis, SSDs are still way more expensive than traditional spinning disk drives, but SSD price wars should come as good news for users with the need for speed.

For more info on Intel’s SSDs, click here.

For general information and news, visit InfoStor’s SSD Topic Center.

We need standards for SSDs

The Storage Networking Industry Association (SNIA) recently announced that it has formed a Solid State Storage Initiative (see SNIA launches SSD initiative). In addition to the SNIA's normal activities such as evangelizing, proselytizing and cheerleading, the SSSI will contribute to standards relating to solid-state disk (SSD) drives.

What we really need here from the SNIA are standards that users, integrators and OEMs can use to compare SSDs, as well as SSDs vs. traditional hard disk drives (HDDs). At a minimum, this standard, or standards, would address performance, providing apples-to-apples metrics to enable comparisons of vendors' performance claims.

But in the case of SSDs, the metrics would have to go way beyond that. For one, they would have to include capacity and price. This would approximate what we get from the Storage Performance Council's SPC benchmarks.

However, the SSD metrics should also encompass durability/reliability and even power consumption. I doubt that it would be possible to come up with a single metric that measured IOPS/$/GB/watts, but the industry will desperately need at least a series of metrics to enable users/integrators/OEMs to make sense out of the nonsense that currently dominates in marketing materials.

The hottest technologies at SNW

In my last blog I promised to reveal The Hottest Technology at last week's Storage Networking World (SNW) show but first, the runners up.

5. Solid-state disk (SSD) drives. There was a lot of talk about SSDs at SNW (mostly in the context of the raging SSD-vs.-HDD debate), but there weren't many product introductions at the show, although Intel did announce production shipments of its new line of enterprise-class flash drives, which will eventually spur further price erosion in this yet-to-get-hot market.

4. Cloud-based storage. Again, a lot of talk, but few vendors. The only cloud storage vendor I met with was Nirvanix, although this services category is expected to grow rapidly over the next couple quarters. For more on this subject, see "What is cloud-based storage?"

3. Storage efficiency technologies, most notably data de-duplication and thin provisioning. Data de-duplication for secondary storage is becoming widespread, but there was a lot of talk at the show about data de-dupe for primary storage from vendors such as NetApp, Storwize, Ocarina Networks, and there will be much more to come.

2. Server virtualization. This is clearly the dominant IT trend, but there are so many storage technologies focused on optimizing virtual servers that it didn't quite make the top of my list.

And the winner is . . .

Fibre Channel over Ethernet (FCoE). This is admittedly an odd choice because, in terms of end-user adoption, FCoE may be years away, but

For one, FCoE was the topic of the only major press conference at the show (hosted by QLogic, Cisco, NetApp and VMware).

Two, a number of vendors made FCoE product -- or at least product certification -- announcements at the show (albeit prototypes in most cases), including EMC, NetApp, Brocade, Emulex and QLogic.

And three, FCoE will provide controversy (as in 10Gbps iSCSI over Ethernet vs. Fibre Channel over Ethernet) for years to come.

I'll address the controversial side of FCoE in my next blog.