Today’s announcement of the E-Class storage array is an important milestone for Nimbus Data and solid-state storage in the enterprise. Until now, most solid-state storage arrays have been fairly small-scale, focused on point performance rather than enterprise-wide capacity. But the E-Class, which scales to 500 TB and sports a redundant, multi-protocol interface, is the first all-flash array to go toe to toe at the top of the market.

The State of Solid

No one would deny that solid-state storage is making a huge impact on the market. With mind-bending performance and reduced power requirements, flash memory matches up nicely with modern data center requirements. But the one missing element has always been capacity: NAND flash is more expensive than magnetic disks on a gigabyte by gigabyte basis.

Check out my 4 Horsemen series for more on the issues of disk!

The solid-state enterprise storage market started with point products targeted at specific needs within the data center. Companies like Texas Memory Systems and Violin have long supported the most challenging database applications with their external arrays, while Fusion-io, Virident, Micron and others stashed flash within the server. These companies were able to sell expensive storage into performance-hungry niches, but have found it difficult to address the capacity needs of the broader storage market.

Technology has the answer to this challenge, as demonstrated by Pure Storage, Nimbus, SolidFire, and others. Thin provisioning makes up much of the difference in cost, and deduplication or compression can even bring parity on a per-capacity cost basis. And even with these features turned on, solid-state storage arrays absolutely murder spinning disks in terms of random I/O performance.

Accumulating Nimbus

Unlike Whiptail, Pure Storage, Kaminario, SolidFire, and the rest of the startup crowd, Nimbus Data is not a new company. Founded by former TrueSAN wunderkind, Thomas Isakovich, Nimbus began as a disk storage startup before transitioning to an all-flash lineup two years ago. The company has steadily improved its product line over the years, adding NFS and SMB for a unified storage experience as well as InfiniBand for extreme performance.

Unlike most other companies in the space, Nimbus builds their own flash memory modules from raw NAND. This allows the company to avoid some of the tricky engineering required to qualify and adapt to the peculiarities of existing consumer or enterprise SSD modules. It also gives the company greater control and better flexibility to launch new capacity points when they are ready, rather than when their suppliers give the go-ahead.

Nimbus is rolling out a dual-PCB SSD module which doubles performance and capacity

The E-Class includes a new dual-PCB module which stripes data internally for better performance and capacity. This bumps each Nimbus E2000X enclosure to 20 TB, twice the capacity previously achieved, in just two rack units. And each of these enclosures draws as little as 100 W, allowing them to be stacked tall without exceeding the power capacity of typical datacenters.

E Is for Enterprise

The real innovation in the Nimbus E-Class is a brand-new “dual active” redundant controller architecture. Most previous solid-state arrays had a single controller, requiring users to mirror two entire arrays for high-availability. In contrast, most enterprise storage systems feature multiple active controllers with no single point of failure.

Read Multipath: Active/Passive, Dual Active, and Active/Active to understand what I’m on about here!

The E-Class introduces a new controller architecture for Nimbus. Each controller services all access to a LUN or volume until a failure is detected, in which case the alternate controller immediately comes online. But both controllers can have active storage at once, in what I call a “dual active” scenario. Although not truly “active/active”, the E-Class is in a different league from older single controller arrays.

Interestingly, the performance of Nimbus’ solid-state enclosures is high enough that the controllers do not need to mirror internal cache or hash tables. They simply write them out to SSD to be picked up in the event of a failure. This simplifies engineering for a dual controller system, and may lead to additional controllers added in the future.

Straightforward Pricing

Solid-state storage provides much more performance per dollar than spinning disk, but most customers still pay on a raw capacity basis. Rather than rocking the boat with alternative pricing models, Nimbus sticks to a straightforward method: $25,000 per controller, plus $100,000 per 10 TB enclosure. A minimum E-Class configuration includes two controllers and one enclosure for $150,000 with no extra cost for software licensing or features.

This seems fairly expensive for 10 TB of storage, but is actually quite competitive even with disk-based storage systems in the high-end, high-feature enterprise market. Thin provisioning increases the usable capacity of the E-Class, and the all SSD architecture means performance will not suffer. Unlike PCIe solutions, the E-Class is a shared, networked device and can be used by many servers at once.

Features and More Features

Speaking of features, Nimbus includes just about anything you could ask for in a storage array:

• In-line thin provisioning and deduplication
• Snapshots and synchronous or asynchronous replication
• 10 Gb NFS (2, 3, and 4) as well as SMB (CIFS/SMB1 and, SMB2)
• 10 Gb iSCSI
• 8 Gb Fibre Channel
• 40 Gb QDR InfiniBand

Although it has not yet been announced, Nimbus has added VMware VAAI support to the HALO operating system found in the S- and E-Class arrays. The company will support the major block storage components for vSphere 5: Block zeroing, atomic test and set, and full copy. Tom told me that Nimbus found the T10 interfaces fairly straightforward to implement but are still working on the NFS primitives. Although Nimbus does not yet have a vCenter plug-in, I expect that one is in the works.

Tom also tells me Nimbus is a “hypervisor hugger“, in that they intend to support features there rather than try to add them to the array. This is a smart choice for a smaller company, and I am glad to see Nimbus embracing the server virtualization market. I imagine an array like the E-Class would totally demolish any competing disk-based array in a virtual infrastructure deployment!

Stephen’s Stance

Nimbus has always been an interesting company, with a longer history in the storage world than most startups. Their switch to all-flash architecture was perfectly timed with the market shift, and the new E-Class comes just at the right moment. Boasting 500 TB of maximum capacity, a fully redundant “dual active” controller architecture, massive performance (even InfiniBand), and complete feature set (once VAAI is released), Nimbus may have hit on their hands.

© sfoskett for Stephen Foskett, Pack Rat, 2012. | Nimbus E-Class: The First Big, Redundant, All-Flash Enterprise Array
This post was categorized as Enterprise storage, Gestalt IT, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

"So it is down to you, and it is down to me."

Seagate and Western Digital appear to have locked up the majority of the hard disk drive (HDD) market with their respective acquisitions of Samsung and Hitachi’s business. Leaving Toshiba with just a sliver, the American companies will soon become giants, each with more than 40% of the total HDD share and a full line of products. Despite the noise made by solid-state disk (SSD) lovers, the HDD market is likely to continue to rake in profits for decades, and these two giants will battle it out for the foreseeable future.

Western Digital Looks To The Enterprise

Originally maker of integrated circuit chips, Western Digital entered the storage market in the early 1980s, producing hard disk drive controllers. It wasn’t until 1988 that Western Digital produced its first hard disk drive, after acquiring Tandon. These were decidedly low-end products, competing in the desktop PC business with the likes of Quantum and Maxtor, two companies that would later merge and sell to arch-rival Seagate.

Western Digital moved steadily upmarket after the year 2000, expanding buffer cache and platter speeds. This culminated in the Raptor line, the first 10,000 rpm serial ATA (SATA) hard disk drive, and Western Digital is still known as a purveyor of high-performance desktop hard disk drives today. The VelociRaptor, for example, is popular among gamers for its small low-latency platters and high spindle speed.

Although Western Digital sells a wide variety of hard disk drives, they’re not a familiar face in the enterprise storage market. They’ve produced a number of raid storage devices but have never been able to break in the high-end, and have similarly been left out of many OEM contracts.

All this will change shortly, as Western Digital will soon acquire Hitachi Global Storage Technologies (HGST). Formed as a merger of the hard disk drive businesses of IBM and Hitachi, HGST is a formidable competitor in many OEM areas, including enterprise storage. The combined company will control nearly half the storage market, offering products in every niche.

Toshiba looks mighty tiny next to Seagate/Samsung and Western Digital/HGST!

Seagate Expands In Asia

In contrast to Western Digital, Seagate is a familiar name in much of the storage market. Founded by a group of industry legends, including Al Shugart and Finis Conner, Seagate move rapidly from the personal computer space into the enterprise. By the late 1990s, Seagate was prime supplier for enterprise storage companies, competing with IBM and Hitachi.

Although formerly dominant, Seagate was surpassed in market share by Western Digital even before they acquired HGST. The new Western Digital would have dwarfed Seagate, whose 30% market share left them in a distant second place. It is perhaps easier to understand Western Digital’s moves that Seagate’s, but there is much logic in acquiring the hard disk drive assets of Samsung.

First, the transaction, worth 1 1/3 billion dollars, bring Seagate back within spitting distance of the new Western Digital. It also opens up the vast Asian OEM market, where Samsung has had much success, and guarantees a market for Seagate hard disk drives in Samsung products. But the relationship between these two companies goes much further: Samsung and Seagate are now related companies, just as Hitachi and Western Digital will be once the acquisition is complete. In both cases, the new companies will have a strong East-West alliance.

The NAND Angle

Although much of the attention in both transactions has revolved around a hard disk drive business, one should not overlook the solid-state implications. Samsung is the world’s largest supplier of NAND flash memory, and Seagate will gain an important relationship with the company. This may be the furthest reaching aspect of the transaction, since Seagate will be able to leverage this relationship as high-performance storage transitions to flash memory.

HGST had already been working with Intel to develop high-performance flash-based storage, and their combination with Western Digital will continue and expand this relationship. Intel, partnered with Micron as IMFT, is another leading supplier of flash memory chips, and the collaboration with HGST looked promising in the enterprise space. Therefore, both companies gain access to key flash memory technology thanks to these transactions.

Stephen’s Stance

Both Seagate and Western Digital have much to gain from these transactions. Western Digital becomes a full line giant of the industry, a credible competitor, and a successful supplier to OEMs. Seagate also retains its credibility in the market, but also gains access to Samsung, one of the strongest electronics companies in the world. Time will tell which of these companies got the better deal.

© sfoskett for Stephen Foskett, Pack Rat, 2011. | Seagate Versus Western Digital: The Hard Disk Drive Battle Lines Are Drawn
This post was categorized as Computer History, Enterprise storage, Everything. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

It won't be long before other manufacturers adopt the new SATA SSD form factor introduced in the MacBook Air

More information about the unconventional SSD used in Apple’s new MacBook Air. As I discussed in my previous coverage of this new flash form factor, it resembles a PCI Express Mini Card but is much smaller. Toshiba has now proved my speculation that the device uses SATA signals rather than the PCI Express lane used by the similar AirPort card. We also know that the lauded performance of the device is due to its chips and controller rather than skipping SATA in favor of PCIe as some had speculated.

Toshiba Blade X-gale™ HG Series SSD

Toshiba is one of the world’s largest NAND flash manufacturers, and the company uses these chips to produce integrated solid state disks (SSDs) for OEMs. Toshiba’s SSD product offerings are divided into two lines: The mainstream SG series and the high-performance HG Series. The SG line includes standard SATA, slim SATA, and mSATA cards, the latter using the Mini-PCIe form factor. These products only promise 50 MB/s of write performance, while the HG Series can top 180 MB/s writing.

The new "Blade X-gale" SSD form factor used in the MacBook Air is part of the high-performance HG Series from Toshiba

Since the MacBook Air’s SSD is part of the third-generation HG series, its performance is much better than the SSDs used in typical netbook-class computers. NAND flash drives typically suffer during writes, but Toshiba promises sequential write performance for the HG Series of 180 MB/s, matching the read performance of their SG Series. This is almost as fast as the 3 Gbps SATA interface used!

Tiny SSDs like these can only use a few high-density flash chips, so eking out this kind of performance is doubly impressive. The “blade” form factor is about the size of a large USB flash drive and includes just four NAND chips on the top of the board. Toshiba includes both read and write cache in their controller, as well as encryption hardware which is apparently disabled by default. The device supports TRIM, even though Mac OS X does not (yet).

Toshiba offers Blade X-gale SSDs in 64 GB, 128 GB, and 256 GB models. Apple apparently uses all three, offering the smaller pair in the 11 inch MacBook Air and the larger two in the 13 inch model. These products are not available at retail yet, and Japanese reseller PhotoFast has apparently withdrawn their GM2 SFV1 Air Upgrade Kit which used similar modules, so there’s no telling when MacBook Air owners will be able to upgrade.

The 256 GB module is 3.7 mm thick, 1.5 mm more than the 64 and 128 GB siblings. One assumes that this reflects its use of 8 NAND chips rather than 4, and this might lead to better performance as well. Speaking of dimensions, the card is actually 24 mm wide and 108.9 mm long. My previous guesses were quite a bit off.

Stephen’s Stance

This new SSD form factor is certainly intriguing. Although no standard name has been coined at this point (Toshiba’s name, “Blade X-gale,” is a trademark), we will be watching with great interest to see if it catches on. I’ll be calling it a “blade SSD” until I hear a better name.

The compact dimensions of the connector and module itself should be welcomed by tablet and portable device manufacturers, and the fact that it can carry PCI Express as well as SATA signals makes it very appealing. The next-generation NAND chips should easily double capacity within the next year, too!

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Toshiba Offers “Blade” SSDs (Like Apple’s MacBook Air)
This post was categorized as Apple, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

The Four Horsemen of Storage System Performance: These four ugly gentlemen stand between you and your data.

Why do some data storage solutions perform better than others? What tradeoffs are made for economy and how do they affect the system as a whole? These questions can be puzzling, but there are core truths that are difficult to avoid. Mechanical disk drives can only move a certain amount of data. RAM caching can improve performance, but only until it runs out. I/O channels can be overwhelmed with data. And above all, a system must be smart to maximize the potential of these components. These are the four horsemen of storage system performance, and they cannot be denied.

Overcoming the Limits of Spindles

Perhaps the previous discussion of spindles left you exhausted, imagining a spindly-legged centipede of a storage system, trying and failing to run on stilts. The Rule of Spindles would be the end of the story were it not for the second horseman: Cache. He stands in front of the spindles, quickly dispatching requests using solid state memory rather than spinning disks. Cache also acts as a buffer, allowing writes to queue up without forcing the requesters to wait in line.

Cache may be quick, but practical concerns limit its effectiveness. Solid state memory is available in many types, but all are far more expensive per gigabyte than magnetic hard disk media. DRAM has historically cost 400 times as much as disk capacity, and even NAND flash (the current darling of the industry) is more than 40 times as expensive. Practically speaking, this means that disk devices, from the drives themselves to large enterprise storage arrays, usually include a very small amount of cache relative to their total capacity.

When specifying a storage system, the mathematics of cache and spindles adhere to a simple rule: More is better for performance but worse for the budget. This leads to a trade-off, where a point of diminishing return tells us to stop adding both spindles and cache and accepting the storage system as it is.

A History of Cache

Cache was not always as common as it is today. When even a small amount of DRAM cost hundreds of dollars, adding a single RAM chip to a hard disk drive would have broken the bank. So many drives had no cache at all well into the mid 1990′s. Operating systems of the time used expensive system memory as a buffer for storage operations rather than expecting cache in the disk controller or drive – remember setting the Buffers command in config.sys?

This was not as bad as it seems, at least in theory. Operating systems stand a fighting chance of “knowing” what data will be requested next, and could therefore request it ahead of time. They also might get a hint about data that will never be used again and can thus flush that from the so-called buffer cache. Although MS-DOS wasn’t very good at this, modern systems have greatly advanced in this respect using a technology called demand paging.

Caching at the array was the key differentiator for early enterprise RAID systems, overcoming the punishing slowdowns caused by parity calculations when data was written. EMC adapted their DRAM-based solid-state storage systems to become a cache in front of hard disk drives and the Symmetrix was born. The Data General (now EMC) CLARiiON was notable as well, bringing a large intelligent write cache to the vast market of midrange systems that could never justify the high price of a Symmetrix. Today, all vendors, from IBM to HP to NetApp to HDS, have vast and clever caches.

The importance of cache on enterprise storage performance can not be over-stated. Mix together rotational latency, seek time, and RAID penalty and you get seriously-compromised I/O response time. But cache can eliminate this penalty entirely, provided there is capacity, by confirming the write and queueing it for later (a concept known as write-back caching). Busy shared storage systems would be simply unusable without cache.

Five Uses for Disk Buffers

Hard disk drives today normally contain a small amount of RAM to use as a buffer for I/O requests. This serves the following needs, though not all are found on all drives:

1. A read cache, allowing frequently-requested data to be read from memory rather than involving mechanical disk operations
2. An I/O-matching mechanism, allowing slower disks and faster interfaces to work together
3. A read-around (ahead or behind) pre-fetch cache, saving a few blocks around any requested read on the assumption that they will also be requested soon
4. A read-after-write cache, saving recently-written data to serve later read requests
5. A command queue, allowing write commands to be reordered, avoiding the “elevator seeking” common to early hard disk drives

Disk buffer size has expanded rapidly in recent years, with some devices including 64 MB or more or DRAM. Seagate’s Momentus XT drive even includes 4 GB of NAND flash as a massive read cache!

Write-Through and Write-Back Cache

There are two basic methods of caching data:
The earliest systems used read-only or write-through caches. All I/O requests pass through the cache, which usually saves the most recent and serves them up when a read is requested. They don’t buffer write requests at all, simply passing them through to the storage system to process. They are safe, since the storage device always has a consistent set of committed writes, but they do nothing to offset the RAID penalty.	Most modern storage systems use a write-back (also called “write-behind”) cache, which acknowledges writes before they are committed to disk. They use non-volatile RAM, battery-backed DRAM, or NAND flash to ensure that data is not lost in the event of a power outage. Though far more effective, this type of memory is also far more costly.

Just about every modern storage array uses caching, and most employ the write-back method to accelerate writes as well as reads. Some have very smart controllers that perform other tricks, but Smart is another Horseman for another day. As mentioned before, RAID systems would be nearly unusable without write-back cache allowing the disks to catch up with random writes.

Onward: I/O, and Smarts

The horseman of spindles is harsh, but he does not rule the day. There are many ways to overcome his limits and his three brothers often come into play. These are cache, which bypasses the spindle altogether; I/O, which can constrain even the fastest combination of disk and cache; and the intelligence of the whole system, which limits or accelerates all the rest. We will examine these horsemen in the future!

I’ve been meaning to write this up for a long time. Thanks for listening and commenting!

Note: Some of these links include affiliate codes that help pay for this blog. For example, buying an Amazon Kindle with this link sends a few bucks my way! But I don't write this blog to make money, and am happy to link to sites and stores that don't pay anything. I like Amazon and buy tons from them, but you're free to buy whatever and wherever you want.

© sfoskett for Stephen Foskett, Pack Rat, 2010. | The Four Horsemen of Storage System Performance: Never Enough Cache
This post was categorized as Computer History, Enterprise storage, Personal, Terabyte home, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Hybrid Hard Drives: Take Two

Thar she blows! Seagate's hybrid ssd/hard disk drive is real!

The new drive family sports 4 GB of NAND flash and 32 MB of DRAM operating in tandem as a cache. It is a 2.5″ model, and since Seagate doesn’t currently produce a drive fatter than 9.5 mm one can assume it is a two-platter model and will work in most laptops. It sports a 3 Gb/s SATA interface with native command queueing (NCQ), nicely up to date but nothing special.

There are three models in the Momentus XT line, all with the same 4 GB/32 MB cache:

Unlike the previous-generation H-HDD drives, the new Seagates have fully-integrated SSD cache featuring speedy SLC chips. The dependence on the host operating system to make caching decisions was one of the things that sunk H-HDDs in the past, but this looks to be an entirely different solution. Seagate looks to have integrated the flash as an extension of the RAM cache and is using the drive’s own logic to determine what to cache and when. This will not only be more-generally applicable (not requiring a special OS) but will likely work better, since on-drive cache management has improved greatly over the years.

Stephen’s Stance

Assuming I’m right about Seagate’s fully-integrated cache architecture, this drive ought to blow away everything else on the market. The Register includes test results showing SSD-like performance for many workloads, yet this drive is half the cost and twice the capacity. It beats the 10k VelociRaptor drive in every test and will absolutely smoke any “normal” 7200 or 5400 rpm laptop drive. Feel free to exclaim “wow!” at this point.

How excited am I? How about this: Although I upgraded it just last month with a 640 GB Toshiba hard disk drive, I want a Momentus XT in my MacBook Pro. I’d rather have one of these than a straight SSD, considering the mix of performance, capacity, and price. Maybe I can move the Toshiba into an MCE Technologies OptiBay?

I will be watching this release with great interest. Word is that Seagate will officially unveil the drive in a webcast on Wednesday, May 26. I look forward to a flood of performance tests from my favorite consumer sites, and expect it will interest the enterprise guys, too (this means you, Howard!) Could hybrid drives finally be getting real?

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Smoking-Fast Laptops: Seagate Momentus XT Hybrid SSD Disk Drive Confirmed!
This post was categorized as Apple, Enterprise storage, Personal, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

The tech industry has been buzzing about solid state drives (SSDs) again lately, but many questions remain. Even after many major vendors (Apple, EMC, and Dell to name a few) have introduced NAND flash-based disk into their core products, it is unclear whether non-disk storage will fly or flop. I’m betting it will find a nice niche, but that traditional spinning disks are here for a good long time.

Apple’s Flashing Success

When Apple switched from hard disks to flash in their mainstream product line, the world was abuzz with the novelty: Would flash displace hard drives? Sure, the company still offered disk-based storage for those needing vast capacity, but most people found that 8 GB or so of storage was plenty for daily use. Of course, instead of the MacBook Air, I’m talking about the iPod family, which contains just a single disk-based model.

Like the Air, the iPod demonstrates that what matters in the “take it with you” market is portability in the form of low weight, perceived durability, and compact dimensions. And NAND flash excels when it comes to packaging. The flash-based iPod is an excellent semaphore for this market segment in other ways, too. Audio files are fairly small, so music users don’t need all that much storage, relatively speaking. They will gladly ignore the cost per GB, too, at such small capacity points: iPod Nano buyers pay ten times more per GB than iPod Classic buyers.

In the case of the iPod, the compact size and joggable durability afforded by the flash iPods is worth the money to most buyers, not that flash player has sufficient capacity to meet their needs. The MacBook Air teaches a slightly different lesson: Although reviewers are quick to point out that the speed and battery life difference between the hard disk and NAND flash versions of the mini notebook are negligible, early buyers were happy to pay $1000 extra to skip the disk. In this case, they paid for quick access time, light weight, and durability that exist as much in their perception as in real-world benchmarks.

EMC’s Heavyweight Champion

In the exact opposite corner of the data storage world lurks EMC’s top-line Symmetrix DMX storage array. When the company announced the availability of NAND flash drives as their top-tier choice for storage, it turned the heads of the whole enterprise storage industry. Although the technology implementation is substantially different from Apple’s iPod, EMC’s move suggests that another group of customers exists who are similarly unimpressed by a low cost per GB: Enterprise application managers.

Many have suggested that enterprise flash is not yet competitive in terms of price, capacity, reliability, or even performance. And they have publicly disagreed with EMC CEO, Joe Tucci, who claimed effective parity after 2010 at last year’s EMC World event. After all, today’s enterprise flash drives are far more than ten times more expensive than their spinning brothers, and disk capacity continues to march higher by the month.

But the comparison is not about the cost of apples or oranges. In the enterprise storage space, flash drives sot at the top of the pyramid, with just a few units added into the traditional tiered storage mix as a “tier zero” of maximum performance. It is not as simple as pulling out a set of 146 GB FC drives and replacing them with a similar number of flash units. Instead, a few key applications or data sets are migrated up to the pinnacle, with the rest of the stack remaining the same.

There is huge promise when this tiered model is combined with storage virtualization, especially the automated variety. If the tiny percentage of storage that truly needs top-tier performance could be moved to a few solid state disks, the whole stack will benefit from reduced device contention. If automation could make the decision on a block-by-block basis, the effectiveness would be much greater.

I Still Remember

There is another kind of solid state disk in play, too. For over two decades, company after company has pushed the idea of packaging high-performance DRAM as a disk substitute for enterprise storage, just as EMC has now done with NAND. These RAM-based disks offer even higher performance and prices than their flash-based cousins, and none has taken the industry by storm.

Way back when a tiny EMC was one purveyor of solid state storage, I recall the philosophical conundrum posed by the devices: Is it better to package DRAM as storage and use it in a conventional manner or to use that same memory as a cache for actual disks? The market voted for the latter, with EMC and others introducing in-array cache to accelerate RAID to great effect. System memory expanded in parallel, with modern servers optimally caching data in three or more levels internally as well.

Where Does the Flash Go?

For most uses, this is precisely the correct configuration. The priciest and quickest “storage” is placed close to the CPU, with performance and cost dropping and capacity increasing as one moves outward.

Where does flash belong, then? Apple teaches us that NAND flash delivers the goods when it comes to the portable market, and it is likely that the use of this technology in this area will only continue to grow. And EMC shows that there is a need for higher performance in the enterprise storage world as well, though perhaps not enough for pure DRAM devices.

The message is clear: As long as the cost of disk continues to lead, NAND flash will remain a niche product. There are certainly markets for NAND-based devices, from portable computing to the enterprise, but disk just works too well to be displaced. While one can never see too far into the future of storage, it seems clear that conventional hard disks will remain the dominant media for a few more generations of technology at least.

© sfoskett for Stephen Foskett, Pack Rat, 2008. | Flash Forward or Flash Back?
This post was categorized as Apple, Computer History, Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Fram approaching in front of iceberg upernavik, CC-by-SA copyright Kim Hansen

Solid state (NAND flash) storage is all the rage right now, but there are many lingering questions regarding its true performance, reliability, and cost. But no question is more important in determining its ultimate usefulness than that of location: Where should flash storage be placed to maximize return on investment?

Storage companies have argued that flash disks can be used most effectively in external storage devices, arguing that it’s simpler to just leverage existing storage technologies. Server companies have tended to prefer to place it inside the server, asking why, if flash disks are capable of massive random I/O performance and extremely low latency, one would put them at the other end of a Fibre Channel or iSCSI connection, which introduces latency and tends to combine I/O operations?

The Case For Servers

The argument for placing flash in (or very close to) servers boils down to two key contentions:

1. Distance = latency, so moving quick flash devices away from I/O-hungry CPUs erodes their effectiveness
2. Granularity (or lack thereof) is the core problem facing storage management, so moving flash (and other types of storage) closer to the omniscient application is likely to bring greater effectiveness

The server folks are relying on a technical argument – that placing high-speed cache where it could theoretically do the most good is the right decision. And they are right, in a perfect world: A flash-aware application talking to a low-latency flash device over PCI ought to really fly!

There is some disagreement in the server-side argument as well: Is flash a “fake disk” or a new level of caching between RAM and storage? It seems that the pitch leans toward the latter, even when the SSD appears as a disk drive. This is what Fusion-IO is pitching: They skip old-school disk connections like SATA and SAS altogether, placing their storage on PCI Express and asking hardware and software vendors to integrate it as best they can. Consider Sun’s flash integration for ZFS, for example. Note, by the way, that Intel’s Turbo Memory products also offer PCIe flash, despite what you might be hearing.

Of course, we can just use a flash drive in place of an internal hard drive. Just about everyone makes something like this now, and they work pretty well in some cases. Then there are hybrid drives, which have gone nowhere so far.

But will this work? We need operating systems and applications that can make use of this local flash, and that has been a problem. Intel’s flash-on-the-motherboard idea never caught on, even as Vista included ReadyBoost, because the truth is that operating systems, file systems, or applications must be re-engineered to really make use of flash in a server. That’s happening, but slowly.

The Case for Arrays

Then we turn to the other end of the storage pipe. EMC put flash in the DMX in January, and Compellent is doing it as we speak. IBM went wild with a bunch of Fusion-IO drives and an SVC over the summer, too. All this proves that flash works in storage arrays!

Why? Simply because modern storage arrays are already engineered to make good use of disk drive capabilities. This is a “what works” strategy – even though it doesn’t sound as nice in theory, the array doesn’t need lots of re-engineering to see some benefit from flash. And post-RAID virtualized systems like that Compellent can really make hay with a few super-speed flash drives, since they can move hot blocks to flash dynamically.

Sure, there is latency between the CPU and the flash drive, but storage arrays are really computers in their own right. So they can derive the same benefit from flash that a server could, and they can share that benefit to connected servers rather than leaving it locked up.

Why Not Everywhere?

How about we end the debate. Flash works great in the server, and it works great in the array. Why not just put it anywhere it makes sense in your particular environment? Have an operating system, application, or file system that can make use of server-side flash? Go buy a Fusion-IO card! Have a virtualized enterprise storage array? Get some SSD there, too. And remember that it’s not all about NAND flash – RAM-based solid state storage from companies like Texas Memory Systems, Gear6, and Violin are even faster!

But remember one thing: This stuff is still very very expensive, so you have to really need the performance to make a case for flash.

Image by Kim Hansen, GFDL or CC-BY-SA (source)

© sfoskett for Stephen Foskett, Pack Rat, 2008. | SSD: So Close and Yet So Far
This post was categorized as Enterprise storage, Virtual Storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Although Hitachi was first to announce and ship a 2.5″ 500 GB drive, theirs used four platters, pushing thickness to 12.5 mm – too much for most laptops and external enclosures.  So Samsung’s announcement of their 9.5 mm two-platter SpinPoint M6 disk was greeted with enthusiasm, but it has taken months for actual drives to ship.  

All that looks to be changing, however, as at least one vendor is currently offering SpinPoints for sale in the United States, and OEMs are rapidly raising their hands.  Earlier this week it was familiar name Verbatim, who announced a 500 GB addition to their SmartDisk line.  Today it is Mac-friendly LaCie with a dual-drive bus-powered RAID 0 unit delivering a massive terabyte in a palm-sized package.

All this mobile storage goodness will continue pushing cost down and capacity up in the hot mobile market.  Where 120 GB was exciting last year, today we are seeing 200, 250, and even 320 GB drives in affordable notebooks and portable enclosures.  I missed snapping up a 320 GB Verbatim unit last week at Best Buy for just $139, but I expect to see a lot more at this bargain price over the next month or so.  That company claims their 500 GB drive will start under $300, and I expect it will drop 1/3 off that price on sale right off the bat.

Where does this leave supposed bargain NAND flash drives?  They’re not looking as attractive, with even the cheapest consumer units priced over $400 for a usable 60 GB.  Until vendors start hitting massive volumes, NAND will continue to command ten times the price per GB of old fashioned “spinning rust”.  With similar power requirements, expect NAND to remain a niche product for another few years at least.

© sfoskett for Stephen Foskett, Pack Rat, 2008. | Big Little Disks Are On The Way
This post was categorized as Enterprise storage, Terabyte home. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.