I’ve long wished for a new enterprise storage superpower. Competition is good for everyone, and the enterprise storage space has always been highly competitive. Traditional SAN storage powers (EMC, HDS, HP, and IBM) have been under continual attack from tech-heavy upstarts like EqualLogic, LeftHand, Compellent, Xiotech, and 3Par. The smaller (revenue-wise) NAS market has been more serial, with NetApp knocking off Auspex, then challenged by EMC. Yet innovators have been thick there as well, from Exanet to Ibrix, Isilon to Onstor.

Through it all, one thing has been clear: The major companies, though perhaps lagging in technology, were usually able to withstand the attack of the upstarts through sheer strength of salesforce. Storage is a strategic investment, and selection of a storage platform is much more far-reaching than many IT product decisions. The inertia of an installed storage environment makes it a real challenge to switch vendors, giving the established players massive leverage.

It became clear to me and many others that the best way for upstart companies (and, by extension, technologies) was to be part of an established vendor’s sales process. OEM relationships were a big part of this (witness the success of BlueArc and even NetApp and HDS) but acquisition was a much stronger proposition. If customers were warmer to OEM products than independent sales, they are much hotter when it comes to acquired technology. HP, Dell, IBM, and EMC have all demonstrated the power that comes when an established company buys a startup and puts the power of their sales force behind these new products.

This explains Dell’s fantastic success with EqualLogic. They took a product that was emerging as dominant in its niche (midrange iSCSI SAN) and blasted it into the market, while at the same time optimizing manufacturing and deployment. EMC did the same with Clariion and DataDomain, and HP is showing strong signs of health with LeftHand and Ibrix. Then there is IBM, who took XIV out of Israel and made it a source of irritation to the rest of the industry.

Many industry watchers have long wondered what would happen if the smaller guys got together, forming a new superpower of their own. Would 3Par, BlueArc, and Sepaton be a real challenger? What about Xiotech or Compellent and Isilon or FalconStor? Is mixing and matching some smaller companies a recipe for success? The answer was often a counter-question: What if someone like Dell, who knows how to manufacture and sell, picked them up instead? This seemed much more like a sure-thing, since the established management and financials stave off potential integration issues.

It appears that this is the future. Established players will pick up smaller companies, fortifying their offerings and accelerating sales in a way the little guys weren’t capable of. Dell’s billion-dollar acquisition of 3Par reportedly headed off a similar offer from HP, and will likely spark another acquisition. I imagine the management teams at Compellent and Xiotech just got a lot busier…

Clearly, Dell and HP are playing this game. IBM and EMC are in it, too. But what about Cisco and Oracle? Could they be planning storage acquisitions of their own, to the detriment of partners like EMC and Hitachi? What about the strong contingent from Japan, NEC and Hitachi? And who gets picked up next? We shall see!

© sfoskett for Stephen Foskett, Pack Rat, 2010. | Dell + EqualLogic, Exanet, Ocarina, 3Par = What?
This post was categorized as Computer History, Enterprise storage, Gestalt IT. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Frankenweenie saves young Victor in Tim Burton's macabre short film

When is a solution integrated and when is it a Frankenstein-like mashup of tangled tech? Apparently, that line is crossed when it’s your competitor’s offering…

In my time in the storage industry, I’ve seen enough franken-storage come and go to make me skeptical whenever a new “integrated” solution is announced. But a lot of this stuff works just fine, so I also know that integrated solutions aren’t always bad!

The latest industry blog flame war centers around NetApp’s recently-announced solid state storage solution, which pairs a V-Series NAS head and a Texas Memory Systems RamSan-500 flash storage system. Perhaps NetApp’s Val Bercovici did get a bit over-excited in his post on the topic, but he wasn’t just talking about the RamSan: He was laying out how NetApp’s WAFL technology can work in an SSD world, and using some recent performance test numbers on that solution as well as their PAM cache cards as an illustration of this.

The next thing you know, we have EMC’s Storagezilla and IBM’s Barry Whyte calling the company out for what they (and others. like Storagebod) see as an underwhelming product offering. That’s all well and good, and I’ll let the reader decide if NetApp’s moves warranted a press release, but now things have gotten uglier…

EMC’s Chuck Hollis called the whole RamSan idea to account, saying it was “Frankenstorage“, causing NetApp’s Alex MacDonald to engage in a little “I know you are but what am I” in reference to EMC’s CLARiiON/Celerra “unified storage” solutions.

It’s time to bring some sanity to this whole integrated solution concept. Every product in the storage world is an amalgamation of OEM parts to one extent or another, and there are always integration issues. Certainly many of EMC’s offerings could be the subject of name-calling: They use STEC SSD drives in the DMX, they use Quantum deduplication engines in their CDLs, and their Celerra NS platform does include a complete Fibre Channel SAN behind the curtain. But they’re not alone, and not even wrong in doing this: Every vendor relies on OEMs, and as a wise man said, “working with an OEM gives you the flexibility to pick best of breed technologies” and that’s exactly what customers want. Any objective person would welcome qualification and integration of TMS’ RamSan with a solid platform like the NetApp V-Series – it’s a certifiable win for the customer. Just like they would be happy to see EMC leveraging great technology from Quantum and STEC.

Chuck goes on to point out some downsides to these OEM combinations, and they’re certainly fair criticisms:

• When you’re buying this from this guy and that from that guy, it’s bound to cost more because everyone needs their cut.
• Since all attempts at unified heterogeneous device management have failed, a combo is certainly harder to manage than a single device.
• With multiple vendors in the mix, fingerpointing is common once support is needed.

But these criticisms can be mitigated by the vendors themselves. They can give up some margin in order to gain market share. They can create unified management interfaces for the combinations they sell and support. And they can really support what they sell, refusing to give in to the temptation to say “not my problem” when the going gets rough. And companies deal with these problems all the time! Frankenstorage doesn’t have to be so scary…

This post can also be found on Gestalt IT: The Difference Between “Integration” and “Frankenstein”

© sfoskett for Stephen Foskett, Pack Rat, 2009. | The Difference Between “Integration” and “Frankenstein”
This post was categorized as Enterprise storage, Gestalt IT. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.

Let’s take a look at the two company’s product lines and market positions as we determine the impact of this deal.

Update: Fujitsu is denying the deal, even though the market loves it, but it still makes sense for WDC to pick up either Fujitsu or Hitachi’s disk drive business to better compete with Seagate.

Update 2: Rumors in December are that the deal is back on.

Final (?) Update: The deal is off as of January 2009.

Introducing Fujitsu

Although they’ve never enjoyed the high profile of (post-Maxtor) Seagate,  Western Digital, and (post-IBM) Hitachi, sixth-ranked disk manufacturer, Fujitsu, has enjoyed success in two key OEM markets: Laptops and servers. The company’s disks are found in many portable computers (including my Apple MacBook Pro), sharing this market with number-three maker, Hitachi. Although Fujitsu reportedly does not manufacture their own drive wafers, they apparently handle all other aspects of design, construction, and assembly.

The company offers a variety of portable 2.5″ drives:

• The BH series is a thin 9.5 mm unit in capacities up to 320 GB
• The BJ is a high-performance 7200 rpm, 9.5 mm mechanism for high-end machines
• The new BT is fatter, at 12.5 mm, but reaches 500 GB
• Finally, the CJ is an upgrade to the BJ, with built-in encryption in addition to 7200 rpm and 9.5 mm thinness

In the enterprise space, Fujitsu offers both 2.5″ and 3.5″ models:

• The 3.5″ MBA3 is available in both Fibre Channel and SCSI variants at 10k and 15k rpm speeds and sizes to 300 GB
• The MBB2 is a 2.5″ drive for blade servers, with low power consumption, 10k rpm speed, and a SCSI interface
• The MBC2 bumps the 2.5″ drive up to 15k rpm

The company also offers a variety of 2.5″ drives for embedded use, with enhanced vibration and shock tolerance and 24×7 operation.

So let’s sum it up:  Fujitsu is the number-six maker (in terms of sales), is strong in the OEM laptop and server markets, and has a respectably diverse set of drive offerings, but mostly focuses on 2.5″ mechanisms.

Western Digital

Now let’s turn to the reported buyer, Western Digital (NYSE: WDC). WDC is well known in the hot (and hotly competitive) retail space, selling buckets of external and portable USB hard disk drives to consumers. The company has been a fixture of the PC market for decades, originally as a manufacturer of a variety of components but focusing on hard disk drives in the early 1990s. The company has grown organically, only acquiring IP and R&D through acquisition rather than buying up its competitors like rivals Maxtor and Seagate. Reports say that Western Digital also buys their platters, but handles the rest of the engineering and construction activities in-house.

Western Digital’s Caviar line was the performance toast of the PC community in the 1990s, and they have recently had a resurgence with their Raptor and VelociRaptor drive units. The latter, a 10k rpm 2.5″ mechanism designed for desktops, was especially lauded in the press and blogs. Although their RE2 and RE3 enterprise SATA drive lines sport respectable specifications, Western Digital has never been a common sight in the data center.

On the mobile side, WDC has two lines: The 7200 rpm Scorpio Black and 5400 rpm Scorpio Blue. The latter is especially common wrapped in the company’s well-regarded My Passport portable USB enclosure, and is also used by OEMs like Verbatim. I transplanted one of these 320 GB Scorpio Blue drives into my MacBook Pro, in fact.

WDC’s bread and butter has long been its desktop drives, however. The company currently manufactures three lines of Caviar drives (Black, Blue, and Green) for different market segments, and their My Book external desktop drives are hot sellers in the consumer sector. The company boasts “green” drives with lower power requirements, as well as embedded units for DVRs like TiVo.

A Nice Match

So what would Fujitsu bring to Western Digital? Greater OEM leverage and increased manufacturing capability, chiefly, along with more R&D depth. Both companies have similar manufacturing capabilities, and Fujitsu’s strong OEM laptop and server lines would mesh nicely with WD’s existing desktop and consumer strength. Although the combined company would still trail Seagate in market share (30% vs. 35%), it would be a force to reckon with in the market, and could force further consolidation among the smaller players. Hitachi’s OEM drive manufacturing business would certainly face pressure, and they would have been a nice alternate target for Western Digital. All in all, I give this rumor two thumbs up!

© sfoskett for Stephen Foskett, Pack Rat, 2008. | Western Digital + Fujitsu = More Competition for Seagate and Hitachi
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.

Lots of innovation is currently on the horizon in the enterprise drive space, notably the application of desktop and mobile technologies to the space. Right now, you can buy a 15,000 RPM 2.5″ dual-SAS enterprise mechanism from two different companies! Or maybe you want a 1 TB bulk drive with SATA? These are a far cry from the bread and butter 10- and 15k 3.5″ SCSI and FC drives we’ve long been accustomed to. Click through for the full story…

Segment Differentiators

The market for enterprise drives is quite different from the desktop world. Buyers are less price- and power-sensitive and much more concerned about raw performance.

Vendor claims of enhanced reliability have been questioned by many, but it cannot be doubted that these drives are often engineered differently. Seagate representatives told me that they focus on stronger casings to reduce flexing under stress of faster rotational speeds and increased heat. They also often use different bearing, airflow, and filter designs. And enterprise drives use lower-capacity platters. But objective studies have not revealed great reliability improvements.

Drive interfaces are quite different than other market segments. Although SATA and Serial -Attached SCSI (SAS) is becoming more common, the bulk of enterprise drive shipments use Ultra SCSI or Fibre Channel. State of the art is 3 Gb dual-SAS and 4 Gb FC, with parallel SCSI and especially 2 Gb FC becoming less common. We will soon begin seeing 6 Gb/s SAS, and multi-channel drives promise to multiply interface performance.

Spindle speed has long been the defining characteristic of enterprise drives. Back when 5400 (and even slower) ATA drives were common in desktops, enterprise arrays relied on 7200 and 10,000 RPM SCSI and Fibre Channel drives. These days, desktops have ratcheted up to 7200, and enterprise units have climbed to 15,000 RPM, with 10,000 RPM becoming the new standard. These speeds (7200, 10k, and 15k) are available across the board on all enterprise drives.

Emerging Trends

Two trends are particularly notable in the enterprise space: “Enterprise” brothers of high-performance desktop SATA drives, and 2.5″ form factor units.

All four current enterprise drive vendors offer high-capacity SATA drives in the enterprise market. Ranging in size to 1 TB, these drives are increasingly being deployed for tiered storage, disk-based backup, archiving, and other less performance-sensitive tasks. Paired with RAID-6, it is likely that the reliability of enterprise storage arrays using SATA drives will be satisfactory. Certainly their massive capacity will bring the per-GB price point down!

SATA in the enterprise has been controversial, but much of this has to do with the meager specifications of the SATA drives examined, rather than any limitation of the protocol itself. It must be pointed out that SATA and SAS share the same physical connectors and interface chips, and the introduction of command queueing (albeit a different implementation) across the board in the latest SATA drives means the interface could be appropriate for just about any use if paired to a fast-enough drive mechanism. I wonder if their opinions would change if they tested Western Digital’s 10,000 RPM Raptor instead of a 7200 RPM Hitachi or Seagate drive.

Undoubtedly, multi-channel and 6 Gb SAS will outperform SATA in the long run. And the implementation of command queueing, drive naming, and bus expansion in SAS remains superior. But SATA is plenty fast for many of today’s needs, especially when tiered storage is considered.

The other big trend is the introduction of 2.5″ enterprise drives. With ultra-fast 15,000 RPM 2.5″ enterprise drives now available from multiple vendors, this segment simply cannot be ignored. In fact, the fastest enterprise drive available today (by a slight margin) is a 2.5″ Seagate Savvio! Combine these amazing performance claims with the fact that more physical spindles can be packed into the same space and we have a winning combination for enterprise arrays! However, these ultra-dense 2.5″ arrays will also be ultra heavy, and paradoxically ultra hungry for power and cooling, since many more drives will be used, even though each drive is more efficient. This will lead to the same weird situation we now see with blade servers – where so much density is achieved that empty rack space must be preserved to keep weight, power, and cooling demands in check!

Specific Drive Offerings

There are a multitude of enterprise drives on the market, but many users have little say in what they get. Each vendor selects their own drives when it comes to enterprise storage arrays! But by examining the array of offerings, we can learn something about the market. More information is available in this Tom’s Hardware comparison table, which is updated regularly, as well as this Tom’s Hardware article.

Fujitsu

Fujitsu’s drive names can be perplexing, with dozens of different drives currently offered. Generally, though, you can figure them out with a bit of detective work. They use the second two letters in their naming scheme for drive families or generations – MAW is older than MAX, then comes MAY, MBA, and MBC. The number is the drive’s size – 3 for 3.5″ and 2 for 2.5″. The final two letters is the drive’s interface – NC or NP for parallel SCSI, RC for SAS, and FD for Fibre Channel.

Fujitsu focuses solely on the mobile and enterprise markets, and is joining most other vendors in the 2.5″ enterprise race. Let’s start with their 3.5″ units, though. Replacing the old 10,000 RPM MAW line is the MAX3, available in SCSI (NC/NP) or dual SAS (RC) and offering 36, 73, or 147 GB of capacity and 15,000 RPM. The MBA3 line is offered in 2 or 4 Gb FC (FD), SCSI (NC/NP), or dial SAS (RC) and boasts 73, 147, or 300 GB and 15,000 RPM.

The company has jumped into the 2.5″ form factor as well, which is no surprise given its line of mobile drives. The MAY2 RC was first, with dual SAS interfaces, 10,000 RPM speed, and 36 or 73 GB of capacity. This was upgraded to 73 or 147 GB with the similar MBB2 RC. Both boast 16 MB of cache. The new MBC2 RC, announced in May and available in 36 or 73 GB, is the stunner, though, with 15,000 RPM.

Hitachi

Hitachi’s naming focuses on drive speed – the 7K, 10K, and 15K names are easy enough to decode. Next is an indication of platter size and generation, which is much less specific. Larger numbers are generally newer and seem to refer to the flagship of that line. So the “300″ line tops out at 300 GB, the “147″ is older and smaller, and the “1000″ is the big up-to-1 TB SATA unit.

Hitachi’s enterprise credentials rest on traditional 3.5″ drives with Ultra SCSI and Fibre Channel interfaces. The 10K300 is the company’s entry level, with 10,000 RPM and capacities of 73, 147, and 300 GB. Although that drive boasts only SCSI and 2 Gb FC, the faster 15K147 adds 3 Gb SAS and 4 Gb FC, along with 15,000 RPM speeds.

Just introduced is Hitachi’s top-dog drive, the 15K300. Combining 15,000 RPM speed and 300 GB capacity in a 3.5″ enterprise drive is impressive, and 2 Gb FC is no longer offered.

Hitachi’s new entry-level enterprise model is the A7K1000. Based on the 7K1000 desktop SATA drives, this unit has 3 to 5 platters for 500 GB, 750 GB, or 1 TB capacity. The similarities might lead some to question this 7200 RPM SATA drive’s enterprise credentials, however.

Hitachi also just introduced a 2.5″ enterprise drive, the C10K147. With sizes of 73 and 147 GB, speeds of 10k RPM, and a SAS interface, there is little to differentiate this drive apart from its small size.

Seagate

Seagate’s entry-level line is the Barracuda ES. Now in its second generation (the ES.2), this line is an uprated Barracuda SATA drive with optional dual-port SAS for enterprise applications. Running at 7200 RPM like its desktop brother, the ES line reaches 1 TB.

Seagate’s mainstream enterprise offering is the Cheetah line. The seventh-generation 10K.7 is the 10,000 RPM traditionalist, in 73, 146, and 300 GB capacities and Ultra SCSI and 2 Gb FC interfaces. The 15K.4 offers 36, 73, and 146 GB capacities, 15,000 RPM performance, and adds 3 Gb SAS as an option.

Seagate’s performance leader is the perpendicular-recording 15K.5. The company boasts 100 MB/s sustained throughput from this modern 73, 146, or 300 GB 15,000 RPM drive. 4 Gb FC, 3 Gb SAS, and Ultra SCSI are all offered. A variant of the 15K.5 is the Cheetah NS. Available in 300 GB and 400 GB sizes and 4 Gb FC or 3 Gb SAS, Seagate claims power optimization benefits in addition to best-in-class capacity.

Seagate entered the 2.5″ enterprise drive market back in 2004 with the Savvio line. The 36 or 73 GB 10K.1 was upgraded in 2006 to 73 or 146 GB in the 10K.1 line. But the real winner is the 15,000 RPM Savvio 15K announced in January. This 36 or 73 GB drive is easily the highest-performance 2.5″ on the market, boasting impressive 104 MB/s throughput numbers.

Western Digital

Western Digital trails the pack, offering no 15,000 RPM, FC, or SCSI drives at all. They do boast the world’s only 10,000 RPM SATA drive, the Raptor, as well as a decent-sized 750 GB SATA unit in their RE2 line. I’m looking forward to their future offerings, especially if Green Power is included (as expected).

Conclusion

The arena of enterprise hard drives is perhaps the most interesting in the entire market. Combine the twin impacts of large, slow 3.5″ drives and tiny, fast 2.5″ drives and we will soon see the market split along tiered storage lines. Although simply re-driving a storage array might not make it greener, tiered storage will reduce the average cost and increase the specific performance of enterprise arrays.

Some specific things that excite me in this space:

• The amazing 15k RPM 2.5″ drives from Seagate and Fujitsu give just the right amount of capacity and performance in a tiny form factor. These are perfect for servers and storage arrays alike. I expect Hitachi to come along with a similar drive soon, and I wouldn’t be surprised if laptop-king Toshiba joins in, too.
• Western Digital’s entry into the enterprise space won’t work without OEM wins, but their 10k RPM Raptor drives should prove enticing. I’d expect to start seeing these in tier-2 servers soon, but don’t look for them in enterprise storage arrays, at least not in this generation of hardware.
• The exit of Ultra SCSI is all but assured. Farewell, 68-pin connectors!
• SAS promises to finally offer the front-end connectivity demanded by today’s large, fast drives.

Tomorrow we’ll be looking at laptop and mobile drives. See you then!

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Specialized Server/Enterprise Hard Drives
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.

H3C already has a long list of products, most based on in-house hardware and OEM software. On the storage side, the company makes an iSCSI storage array platform dubbed “Neocean”. This storage platform, selling strongly in China, is alleged to leverage technology licensed from FalconStor (on the low-end IX1000), Intransa (on the bigger IX5000), as well as iVivity and Xyratex. OEM storage developer Ciprico today announced that it will be working with H3C on the next generation. H3C also sells a WAFS accelerator leveraging Expand Networks software. All of these should be coming to the United States next year.

Huawei itself is also getting back into the storage market in the form of a joint venture with Symantec, creatively called Huawei-Symantec. This company is set to be coming out with a line of network devices with Veritas-based software built in. We’re hearing about virus scanning and content indexing appliances, as well as NAS and SAN arrays which will include storage foundation software from Symantec right out of the box.

Who knows what’s next from Huawei? I’d guess expanded services, more resellers in the West, and more OEM deals to create bigger systems. In a few years, they might give Hitachi and EMC trouble in the enterprise market, especially when big server vendors like Sun, SGI, Dell, and HP start rethinking their OEM strategies…

© sfoskett for Stephen Foskett, Pack Rat, 2007. | Storage from behind the great wall
This post was categorized as Enterprise storage. Each of my categories has its own feed if you'd like to filter out or focus on posts like this.