Tuesday, December 21, 2010
Pure-play software vendor Symantec held on to its #2 ranking with a 16.5% share on revenue of $518 million, followed by IBM (13.4%) with revenue of $421 million and NetApp (8.4%) with $263 million.
The race is a bit tighter for fifth and sixth place, with CA (3.3%) pulling in $104 million and edging out #6 HP (3.2%, $99 million).
As was the case in the disk array market in the third quarter, the real winner appears to be NetApp, which experienced a growth rate of 19.8% in Q3 2010 vs. Q3 2009. EMC had the second highest growth rate at 13.9%. HP was the only vendor to have negative growth (-9.4%).
The overall storage software market racked up $3.1 billion in revenue, for a growth rate of 8.7% over the same quarter a year ago and a 6.3% boost over the previous quarter, according to Laura DuBois, IDC’s program vice president, storage software.
In terms of storage market segment growth, the top three were storage infrastructure (+37.3% year-over-year), archiving (+12%) and data protection and recovery (+10.7%).
DuBois notes that the big boost in storage infrastructure can be attributed largely to increased spending on automated storage tiering. Other segments of the storage software market include replication, storage management, device management, and file systems.
For more info, see IDC’s press release: “Storage Software Market Continues on Its Growth Trajectory in the Third Quarter”
Related blog post: “Disk arrays: NetApp, HP duke it out for #3 spot”
Friday, December 17, 2010
“It’s the first quad-port, fully converged 10GbE controller, and it can perform at full line rate on all four ports, or 40Gbps bidirectionally,” claims Robert Lusinsky, director of product marketing in Broadcom’s Ethernet Controller Group.
One caveat: Although Broadcom is currently sampling the chip to OEMs, and claims to have some design wins with Tier-1 vendors, production shipments of the chip aren’t expected until the third quarter of 2011.
Broadcom cites Intel, Emulex and QLogic as its primary competitors in the 10GbE converged networking space. Lusinsky says that Broadcom’s key differentiators vs. some of its competitors are port count, chip size and performance.
To date, the highest performance claims in the converged adapter market have been around one million IOPS.
As always, however, performance claims should be taken with a grain – or large block – of salt, and they depend on a wide variety of factors, including CPU utilization.
For server vendors needing to cram more and more chips on their motherboards, Broadcom points to the size of the BCM5784X0 chip: four ports on a 23mmx23mm (0.82 square inches) device. That compares, for example, to Intel’s dual-port 25mmx25mm 82599 processor.
Broadcom’s converged controllers are available in three configurations: the dual-port BCM57800 and BCM57810, and the four-port BCM57840.
Other features of Broadcom’s 40-nanometer converged controller include support for PCIe 3.0, Energy Efficient Ethernet, and virtualization technologies such as SR-IOV, NIC partitioning and Virtual Embedded Bridge (VEB). The chips provide full hardware offload for iSCSI, FCoE, TCP and RDMA.
For more info, see the company’s press release.
Monday, December 13, 2010
As I mentioned in my earlier post on this saga, the weird thing about it is that Dell’s offer is still significantly below what Compellent had been trading at. Compellent’s stock at one time hit $34 a share. So this is what the financial people call a “take-under.”
In fact, as I write this Compellent’s stock is trading at $27.92, which is still above Dell’s bid price. That would suggest that investors expect the price to go higher, although almost nobody expects a bidding war to ensue at this stage of the game.
Although most of the financial community seems to think that this is a done deal, it’s important to note that Compellent’s shareholders have yet to approve the deal (although both the Dell and Compellent boards have approved the terms of the deal). And it’s likely that lawsuits against Compellent’s board will be filed.
Since this story may not be over yet, it may be premature to speculate on how Dell will position the Compellent line vs. the Dell/EqualLogic line, but I presume that Dell would position Compellent’s arrays “above” the EqualLogic iSCSI arrays and primarily in the Fibre Channel SAN space.
Whatever the positioning between Compellent and EqualLogic, it’s certainly doubtful that the Dell-EMC reseller relation will last the two years left on the agreement (although Dell officials this morning said that they will continue to offer the EMC options).
According to a note by Stifel Nicolaus analyst Aaron C. Rakers: “For its October 2010 quarter, Dell generated storage revenue of $543 million (3.5% of total; ~3.8% if we include Compellent’s most recent results) revenue, which was up only 7% yr/yr (vs. industry growth rate at +19% yr/yr) and down 13% sequentially. Dell had reported that its EqualLogic (iSCSI SAN) revenue grew 66% yr/yr, which leaves us to estimate ~$165 million in revenue (~30% of Dell’s total storage revenue). IDC recently estimated that Dell had a ~9.1% revenue share in the total external disk storage market, which includes the Dell/EMC relationship; Dell having a ~34% estimated revenue share in the fast growing iSCSI SAN market with EqualLogic.”
Related blog post:
Dell closing in on Compellent acquisition (last Thursday)
Friday, December 10, 2010
“Tier 1 storage vendors will move past point solutions for deduplication next year,” says Tom Cook, CEO at Permabit. “They’re working toward end-to-end deduplication, across SAN, NAS, unified [block and file], nearline and backup.”
“When that happens, once their customers ingest data and get it into a deduplicated state they’ll never have to re-hydrate that data throughout its lifecycle. The data will stay deduplicated through processes such as replication and backup. That’s a huge savings in workflow, footprint and bandwidth,” say Cook.
“Today, the big vendors use a variety of point solutions, but they’d like to use a single data optimization product across all their platforms, whether it’s block or file, primary or secondary. End-to-end deduplication will creep into the market in 2011 and 2012,” Cook predicts. (Permabit sells deduplication software – dubbed Albireo – to OEMs.)
Personally, I don’t think that single-solution, end-to-end deduplication will happen that quickly, in part because of the huge investments that the Tier 1 vendors have made in their “point solutions,” but we’ll see.
Dennis Rolland, director of advanced technology at Sepaton, has some predictions that are similar to Cook’s, as well as some differing opinions regarding trends in the data deduplication market.
“Dedupe will be required in more places going forward, including primary storage in addition to nearline storage, and end users will have to cut down on how many dedupe solutions they have because of the complexity in managing many disparate solutions,” says Rolland, “but we’ll probably still have distinct solutions for primary and nearline storage deduplication.”
Rolland thinks that the emphasis on deduplication benefits such as capacity, footprint and cost savings is shifting. “Dedupe enables low-bandwidth replication, which in turn enables companies to economically deploy DR [disaster recovery] sites,” he says.
Rolland also links two technologies that will no doubt make my list of The Hottest Storage Technologies for 2011 (assuming I get around to making such a list): data deduplication and cloud storage.
“Dedupe is an enabler for cloud storage,” says Rolland. “It makes it practical to deploy cloud storage because you’re sending, say, 10x less data over the WAN. That has significant implications for deploying cloud-based DR.”
(Sepaton bundles data deduplication software with its virtual tape libraries, or VTLs.)
Meanwhile, Quantum released the results of an end-user survey this week that suggests U.S. companies could save $6 billion annually in file restore costs by adopting deduplication.
According to the survey of 300 IT professionals, respondents spend an average of 131 hours annually on file restore activities, with 65% restoring files at least once a week. Based on the average wage for IT professionals in the US ($31.55 per hour according to PayScale.com), that equates to $9.5 billion. However, Quantum’s survey also found that those companies that are most efficient at file restoration predominantly use deduplication and can complete restores in approximately one-third the average time of all respondents. So, according to Quantum’s press release: “If the broader US market was to achieve similar data restore efficiencies, the potential annual savings for US businesses would be approximately $6 billion.”
This survey seems a bit misleading to me because it’s not really focused on the advantages of data deduplication per se in a file restore context but, rather, the advantages of disk-based backup/recovery.
Steve Whitner, Quantum’s product marketing manager for DXi, explains: “If you back up to regular [non-deduplicated] disk and you have a need for DR, you have to get that data to another site and you can’t keep data on conventional disk for very long – maybe a few days or a week. So the real issue is not the speed of restore; it’s the fact that companies can now store a month or two of deduplicated backup data on disk.”
You be the judge. Here’s Quantum’s press release and here are some supporting slides from the survey results.
One thing is clear: In 2011, the focus will shift from deduplication for nearline/secondary storage to deduplication for primary storage. Witness two of this year’s biggest storage acquisitions: Dell buying Ocarina Networks and IBM acquiring Storwize. (Storwize’s technology is now in the IBM Real-time Compression business unit.)
Related blog posts:
What is progressive deduplication?
Data deduplication: Permabit finds success with OEM model
Thursday, December 9, 2010
It looks like I will be proved wrong. Although many financial analysts predicted that, after losing the battle with HP for 3PAR, Dell would set its sights on Compellent, I predicted that Dell would turn away from the over-priced storage stocks and make a few acquisitions in non-storage IT sectors.
As I write this, Compellent is trading at $29.37 per share. That would suggest that investors think that (a) a bidding war will ensue and/or (b) negotiations will drive the price up.
I doubt that a bidding war will ensue, mainly because none of the cash-rich storage giants really need Compellent’s technology (although at least one financial analyst predicted that NetApp might jump into the ring). Making a bidding war even less likely: Dell has more than $13 billion in cash, and even though it backed out of the 3PAR bidding there’s no way Compellent will go for much more than $1 billion vs. the $2.4 billion that 3PAR commanded.
Compellent’s stock has almost doubled over the past couple months due to acquisition rumors. And back in August (prior to the start of the 3PAR bidding contest) the stock was trading at about $12 a share.
If this deal goes through, it will be interesting to see how Dell positions Compellent’s products relative to Dell’s (enormously successful) EqualLogic line. There’s certainly a good deal of overlap there. But what will be more interesting is what happens to the Dell-EMC relationship, or lack thereof.
Both Dell and Compellent officials said that there was no assurance that the acquisition deal will be finalized, and they don’t plan to comment further until the deal is either consummated or goes south.
Tuesday, December 7, 2010
That’s how Enterprise Storage Forum contributor Jeffrey Layton begins his exhaustive two-part series on SSD performance degradation.
For years (decades, actually), the focus on SSDs was on the exorbitant prices of the devices. Then the attention shifted to reliability, or endurance, issues. But SSD and controller manufacturers made great strides in those areas over the past couple years.
Now, the focus may be turning to performance degradation over time.
Layton’s articles are the best I’ve read on this subject. However, a warning: The articles are very long, very technical, and very detailed. But if you (a) are using, or considering using, SSDs (b) have sufficient technical credentials and (c) have a lot of time, I strongly recommend reading the articles.
Part 1 examines the issues that cause performance degradation in SSDs, and looks at some of the solutions (or “workarounds,” because all SSD solutions seem to involve trade-offs) that vendors have implemented.
Part 2 looks at technologies/issues such as write amplification, over-provisioning and the TRIM command, and then delves into some very in-depth testing of Intel’s X25-E SSD in ‘before’ and ‘after’ stress test scenarios. Check it out on Enterprise Storage Forum:
Fixing SSD Performance Degradation, Part 1
Fixing SSD Performance Degradation, Part 2
Friday, December 3, 2010
Total capacity shipped grew 65.2%.
In the external array market, EMC held on to its #1 spot by a wide margin, with $1.35 billion in Q3 revenue and a 26.1% market share. IBM was a distant second with $667 million in revenue and a 12.9% market share.
But the real race is for the #3 position, where NetApp and HP are in a virtual dead heat. (Even dead heats are virtual these days.) NetApp had an 11.6% market share in Q3, followed closely by HP with an 11.1% slice. IDC considers it to be a statistical tie when less than a one percent revenue difference separates two vendors.
Dell finished fifth, with a 9.1% market share on revenue of $471 million.
All of the top five vendors had healthy, double-digit revenue growth (ranging from 11.3% for HP to 28.3% for EMC), but it was NetApp that busted the charts with a whopping 54.9% growth rate.
Looking at the leader board trends over the past few quarters, it would seem safe to say that NetApp has blown past HP and is closing in on Big Blue, except for HP’s 3PAR acquisition. With HP’s marketing muscle behind the 3PAR product line, revenue could crank up pretty quickly. For now, however, 3PAR had a market share of only 0.83% in the third quarter. (Isilon’s slice was 0.75%.)
Other highlights from the IDC report: The NAS market was the fastest-growing segment of the overall storage systems market, posting 49.8% growth in Q3 2010 vs. Q3 2009. EMC led the NAS market with a 46.6% share, followed by NetApp with a 28.9% share.
The iSCSI segment of the overall market also did well, posting 41.4% revenue growth, with Dell/EqualLogic in the top spot (33.8% share) followed by EMC and HP in a tie for second place.
For more details, read the IDC press release: “External Disk Storage Systems Market Records Fourth-Highest Quarterly Revenue in Third Quarter”
Thursday, December 2, 2010
In the third quarter, tape media manufacturers pulled in almost $203 million in revenue, which is expected to actually increase slightly in the next quarter to $205 million, according to the Santa Clara Consulting Group (SCCG). And that’s just from tape cartridges; those figures don’t include revenue from tape drives and libraries.
All segments (formats) of the tape market are declining rapidly, with one exception: the LTO format.
LTO tape cartridges accounted for more than 86% ($175 million) of total cartridge shipments in the third quarter, or 6.2 million units, according to SCCG. Sales of LTO-5 (the latest generation, introduced early this year) cartridges were up 100% in Q3 vs. Q2, and accounted for 5% of unit shipments and 15% of revenues. LTO-4 cartridges accounted for 48% of unit shipments and 46% of the total LTO revenues. Even the LTO-3 segment grew in the third quarter (up 3%).
One reason that users stick with tape is that it’s still way cheaper than disk. Those LTO-5 cartridges cost only four cents per GB (assuming 2:1 compression).
According to a study conducted by The Clipper Group research and consulting firm, the cost ratio of storing 1TB of data on SATA disk vs. LTO-4 tape is 23:1. Even if you compare the costs of archiving data on a VTL with data deduplication vs. tape, the cost ratio is 5:1, according to the Clipper Group study.
And for the energy-conscious: Tape is 290X less expensive than disk in terms of energy costs. (The Clipper Group study was conducted two years ago, but with cost reductions in both disk and tape those ratios are probably still accurate.)
Sure, accessing data from tape is painfully slow, but successive generations of LTO have typically doubled the transfer rate. (Exception: LTO-5 only provided a nominal increase in transfer rate vs. LTO-4 – 280MBps vs. 240MBps – but LTO-6 is expected to almost double the transfer rate to 525MBps. That assumes 2.5:1 compression, vs. today’s average compression ratio of 2:1, which will come from the use of larger compression history buffers, according to Bruce Master, an LTO Program representative and senior program manager of tape storage systems at IBM.
And LTO-6 will boost native cartridge capacity to 3.2TB, or 8TB in compressed mode.
The LTO Program vendors (HP, IBM and Quantum) claim that more than 3.3 million LTO tape drives, and more than 150 million LTO cartridges, have been shipped.
In addition to the usual capacity and speed improvements, LTO-5 includes some nifty features such as media partitioning (which enables users to create two partitions on the cartridge, one for content and one for an index) and the Linear Tape File System (which is a free download that leverages dual partitioning and provides file system access at the operating system level).
The LTO Program celebrated its 10th anniversary last month.
Monday, November 29, 2010
According to a blog post on The Wall Street Journal’s wsj.com, the VCs behind Isilon made out quite well (see “EMC-Isilon Deal Is Another Data-Storage Win for VCs”).
According to that article: Atlas Ventures and Madrona Group, which provided Isilon with its Series A funding in 2001, as well as Sequoia Capital, which led the company’s Series B funding, all held significant stakes in Isilon.
Atlas said it will reap $473 million from the EMC-Isilon acquisition, or 20X its initial investment.
Madrona will get more than 15X the $15 million it invested, or $225 million+, according to a Madrona representative.
Sequoia Capital (which apparently did not sell any of its holdings prior to the acquisition announcement), will score $394.4 million.
Isilon was founded almost 10 years ago by Sujal Patel. The company went public in late 2006 at $13 a share. Going public was rocky in the beginning, with shares falling about 50% in the first year. Isilon’s stock price kept sliding through 2008 and 2009, when investors such as Atlas and Madrona added to their holdings.
Under the terms of the agreement announced a couple weeks ago, EMC will pay $33.85 per Isilon share.
Read the full blog post at wsj.com: “EMC-Isilon Deal Is Another Data-Storage Win for VCs”
Tuesday, November 23, 2010
However, the company posted a net profit of $66 million in 4Q10, or 14 cents a share, which is down from $73 million, or 15 cents a share, a year ago. In addition, guidance for the current quarter was, well, cautious. Company executives predicted revenue for the current quarter of $535 million to $550 million, which translates into flat to -3% growth vs. the previous year That guidance was below Street projections. Brocade’s 1Q11 guidance was variously described as “cautious,” “tepid” and “dour.”
In other words, the company’s performance in this quarter was stellar, but the outlook for 1Q11 disappointed. Mixed signals, indeed.
Brocade earned the applause of analysts – and some Buy ratings -- by (narrowly) exceeding their expectations, but investors sent the stock down late Monday, a slide that continued into Tuesday, presumably because of the lackluster projections for the current quarter.
Brocade (NASDAQ: BRCD) shares closed at $5.13 today, down 10%.
Personally, I don’t think Brocade’s guidance on this quarter looks very bad at all, given an expected slowdown or stall in federal spending and Brocade’s challenging competitive position on the Ethernet side of the business against Cisco, HP and others.
Speaking of which, Brocade’s Ethernet-related revenue numbers suggest that the company is picking up steam against those formidable rivals.
4Q10 revenue from Brocade’s Ethernet line was $142.6 million (representing 26% of total revenue.) That’s an increase of 17% vs. the previous quarter. And the company gained about 300 new customers for its Ethernet switch line in the fourth quarter, bringing the total to 1,600 since the acquisition of Foundry.
Brocade’s SAN storage business remained at about 57% of total revenue (vs. 58% in the last quarter), or $316 million. Services accounted for the remainder of the total revenue, or about $92 million.
Gross margin was 62.3%, compared to 54.1% in the previous quarter.
Brocade, HP debut 8Gbps SAN starter bundles
Brocade tackles network convergence, VM mobility
Brocade increases density, throughput of DCX Backbone
Wednesday, November 17, 2010
The company reported financial results today for its fiscal second quarter, highlighted by quarterly revenues of $1.207 billion. That’s up 33% over the same quarter a year ago. Similarly, NetApp’s revenue for the first six months of this quarter is up 34% vs. the previous year.
Net income was $165 million, or $0.42 per share, which compares to net income of $96 million, or $0.27 a share, a year ago. That’s a 72% increase year-over-year.
And gross margin was a whopping 66.9%, while operating margin was 19.8%.
The brightest spot was the company’s hardware revenues (which NetApp refers to as “product” revenue).
“NetApp produced 49% year over year growth in product revenue and our highest non-GAAP operating margin in over a decade,” said Tom Georgens, NetApp’s president and CEO. Product revenue in the second quarter was $780 million. And none of that came from the products NetApp announced a couple weeks ago as part of a wide-ranging refresh (see “NetApp overhauls product line, from arrays to OS”).
Other product lines were not so impressive. For example, software revenue was $177.9 million, up 5% year over year, and services revenue was $249.5 million, up 16% year over year.
What may raise some eyebrows is that NetApp is sitting on net cash of more than $3 billion, which will surely spark speculation on possible acquisitions that NetApp might make given the current climate of red hot M&A activity in the storage space.
Company officials predict third quarter revenue in the $1.24 billion to $1.29 billion range, which would translate into sequential growth in the 3% to 7% range.
NetApp [NASDAQ: NTAP] shares closed at $49.25 today, down 6.5%. The company’s 52-week trading range is $28.92 to $57.96.
Trading on the company’s stock was halted this afternoon due to a leak of the earnings report prior to NetApp’s official conference. Shares plunged, apparently because investors were not pleased with the Q3 earnings estimates.
Related blog post: “NetApp hits a home run in Q1”
Related article: "NetApp overhauls product line, from arrays to OS"
Also today, scale-out NAS specialist (and NetApp competitor) BlueArc reported record third quarter revenues and a 100% quarter-over-quarter growth in channel revenues. More details were not available in time for this post. (BlueArc is a private company.)
Hitachi Global Storage Technologies (GST) claims to be the first hard disk drive (HDD) vendor to ship both SAS and Fibre Channel “enterprise-class” SSDs, albeit in only limited quantities to OEMs, with production shipments ramping over the first half of next year.
See “Hitachi GST enters SSD market.”
LSI announced a PCIe accelerator card based on SSD technology this week, with shipments slated for the end of the month. With 4KB block sizes, LSI claims performance of 240,000 IOPS on sequential reads and 200,000 IOPS on random reads. At an MSRP of $11,500, LSI’s 300GB WarpDrive SPL-300 card isn’t cheap but the company has some cool technology that integrators can bundle, including MegaRAID CacheCade and FastPath software.
See “LSI ships SSD-based accelerator card.”
Violin Memory introduced cache appliances based on flash and DRAM memory that significantly boost the performance of NFS-based NAS systems. The secret sauce in the NFS caching software comes from Violin’s acquisition of Gear6 earlier this year.
See “Violin’s cache appliances accelerate NFS NAS.”
Also yesterday, Anobit announced that it had received $32 million in a funding round led by Intel Capital, Intel’s equity investment unit. That brought the total funding behind Anobit to more than $70 million. Anobit specializes in NAND flash memory and Memory Signal Processing technology.
It’s not surprising that so many vendors are trying to get a piece of the SSD market. Market researcher IDC predicts that the total SSD market will grow at a 58% CAGR in unit shipments and a 44% CAGR in revenues over the next few years.
However, the fastest growing segment of the overall market is enterprise SSDs, where unit shipment are expected to surge at a CAGR of 74%, and revenues are expected to grow at a 54% clip, through 2014.
IDC expects the total SSD market to exceed $7 billion in 2014.
Related article: “Vendor group develops standards for PCIe SSDs”
Monday, November 15, 2010
Some observers have speculated that the $2.25 billion suggests that there were other suitors involved. I don’t think so. I think EMC wanted to seal the deal without a bidding war, and $2.25 billion should do the trick.
EMC’s positioning of the deal was interesting. Predictably, the words “big data” and “cloud” came up a lot in EMC officials’ explanation of the deal, but so did the synergies between Isilon’s platforms and EMC’s Atmos platform.
According to EMC’s press release on the announcement: “EMC’s Atmos and Isilon’s solutions will offer customers a highly scalable, low-cost storage infrastructure for managing ‘Big Data.’ . . . EMC Atmos object storage provides the perfect complement to Isilon for massive globally distributed environments and object access to data for usages like Web 2.0 applications.”
EMC went on to estimate that the combined revenue from the Isilon and Atmos platforms will hit a $1 billion run rate during the second half of 2012.
EMC also emphasized synergies between Isilon’s clustered scale-out NAS platforms and systems/software from Greenplum, which EMC acquired earlier this year.
In short, all of these acquisitions are complementary, not internally competitive. Nice positioning, and in fact it’s true. It’s rare that a vendor can make acquisitions of these sizes without having to shake up its existing product lineup (which is what HP will have to do as it folds 3PAR’s systems into HP’s venerable disk array lineup).
Not surprisingly, the EMC-Isilon announcement sparked more acquisition rumors, but now vendors that play in Isilon’s ballpark are getting some attention, including BlueArc (to be acquired by long-time partner Hitachi Data Systems?) and Panasas, which is moving into more commercial markets (see “Panasas Pushes Scale-Out Storage Performance Envelope” on Enterprise Storage Forum).
If $2.25 billion seems like a high price to pay for a barely profitable Isilon, consider the fact that IDC predicts that the market for scale-out NAS will grow on average 36% per year, reaching $6 billion by 2014.
EMC snaps up Isilon for $2.25 billion (Enterprise Storage Forum)
Isilon revenue up 77% (InfoStor blog post)
NetApp overhauls product line, from arrays to OS (InfoStor news story)
Thursday, November 11, 2010
Fueled by a combination of acquisition rumors and a stellar quarterly financial report (see “Isilon revenue up 77%”), Isilon’s shares hit an all-time high of $29.48 late last month. The 52-week trading range: a shocking $5.32 to $29.48.
More recently, shares of Isilon (NASDAQ: ISLN) seem to have stabilized, closing at about $26 today.
EMC was reportedly in deep and exclusive talks with Isilon regarding a buyout, with the New York Post reporting that deal was in the $2 billion neighborhood (where 3PAR lived). More recently, various reports said that the EMC-Isilon talks were either dead (because of the pricey neighborhood) or stalled. I’d bet on the latter.
Realistically, I don’t think that (the barely profitable) Isilon can get $2 billion+ from anybody. So I’d still say that EMC is the most likely acquirer.
However, assuming that EMC is out of the bidding, financial analysts cite the usual suspects as possible bidders: Dell, NetApp, HP, IBM and Oracle – usually in that order. However, a bidding war a la HP vs. Dell for 3PAR will be highly unlikely this time around, in part (as I’ve said before) Isilon’s technology is great but it’s a product line gap plugger rather than the game changer that 3PAR would have been for Dell and could be for HP.
Financial analysts such as Morningstar’s Mike Holt, on thestar.com, cite Dell as the most likely candidate to snap up Isilon. However, rather than take what would be sort of a consolation prize after losing its bid for 3PAR, I’m guessing that Dell heads in a different (non-storage) direction in its acquisitions. And besides, Dell has the Exanet technology.
Isilon’s clustered scale-out NAS technology would fit nicely into NetApp’s lineup, but NetApp has shed a lot of blood, sweat, tears and money integrating the Spinnaker technology, and acquiring Isilon would be a lot of expensive crow to eat.
With Isilon’s hiring of Qatalyst Partners (which shepherded the 3PAR bidding), it seems quite likely that Isilon will be acquired. Then again, given the company’s traction and recent earnings report, the Isilon could very well go it alone.
Related blog post: “Isilon revenue up 77%”
Thursday, November 4, 2010
First, a note on terminology: Vendors such as QLogic and Emulex use the term converged network adapter (CNA) to describe cards that support protocols such as 10GbE, Data Center Bridging (DCB), iSCSI and FCoE. Other vendors, such as Broadcom, are expected to use the term Converged Network Interface Card, or C-NIC.
Intel doesn’t use either of those terms, referring to the technology in general as “unified networking” or just referring to its Ethernet X520 Server Adapter card.
Intel’s approach to unified network cards is architecturally different from the CNA approach taken by vendors such as Emulex and QLogic. With CNAs, protocols (and protocol offload functionality) are implemented on the CNA, which architecturally is similar to a host bus adapter (HBA).
In contrast, Intel uses native initiators for FCoE and iSCSI that are implemented in the host operating system (Windows and Linux) kernel. Thus Intel’s use of the term “open FCoE.”
“We rely on native protocols in the OS kernel, so it’s free and easy to implement and use,” says Sunil Ahluwalia, senior manager of product marketing in Intel’s LAN Access Division. “CNAs are special-purpose adapters that are more complex and costly.”
Intel’s Ethernet X520 Server Adapter has an MSRP of $799 (and often sells for considerably less), which is about half the price of CNAs.
Rather than implementing hardware-based full offload of the storage-over-Ethernet protocols, Intel offloads only the data path part of the protocol.
“We have what we call ‘intelligent offload,’” says Ahluwalia. “We don’t offload the full protocol because we don’t think that’s required, and full offload adds complexity and cost to the cards because you need processing engines, memory, etc. on the card, which also adds power requirements.”
CNA vendors have historically claimed that the native initiator approach (a) lacked full support for protocols such as FCoE (which is no longer the case) and (b) had relatively poor performance.
Intel begs to differ. However, it’s important to note that all vendors in this space can point to internal or third-party testing that shows superior performance for their products. As always, benchmark results should be taken with a grain of salt.
“Intel and Microsoft demonstrated one million IOPS with native iSCSI earlier this year, but the question was: Who needs that level of performance in real-life applications?,” says Ahluwalia, “so we performed some tests [with Demartek] running applications such as Exchange and SQL and we found little difference in performance between a CNA and an initiator approach.”
For full info on the testing procedures and results, see Demartek’s “Intel 10GbE Adapter Performance Evaluation for FCoE and iSCSI.” (Demartek’s site has a lot of other interesting FCoE-related content).
“We found that the performance of the Intel X520 adapter was comparable to competitive 10Gb FCoE adapters for a broad spectrum of tests,” said Dennis Martin, Demartek’s president. “Because the adapter performance was reasonably close in most of these tests, IT professionals need to consider the cost of these adapters, especially in environments where many adapters are required.”
Intel’s Ahluwalia also claims that the native initiator approach does not consume a lot of CPU cycles (another criticism of this approach).
To take one example from the Demartek tests: In a Microsoft Exchange Jetstress benchmark with 5,000 mailboxes, Intel’s card/initiator consumed only about 2% CPU utilization.
But you really have to take a close look at the Demartek tests, as well as testing of CNAs conducted by other vendors and third parties, to get to the bottom of the performance claims in the unified, or converged, network adapter market.
Related article: QLogic announces 10GbE NICs, CNAs
Tuesday, November 2, 2010
Breaking those numbers down a bit, software revenue was $35.8 million, up 7% year-over-year and 26% sequentially. Services revenue was $39.5 million, up 19% year-over-year and 13% over the prior quarter. Income from operations was $9.2 million. That compares to $6.5 million in the same quarter last year.
CommVault executives highlighted a number of recent parternships that could fuel growth, including a distribution agreement with Hitachi Ltd. subsidiary Hitachi Computer Peripherals, as well as a partnership with Mezeo Software that will integrate CommVault’s Simpana software with the Mezeo Cloud Storage Platform.
But the big highlight in the quarter was CommVault’s launch of the long-anticipated Simpana 9 software, which included significant enhancements for data protection in virtual server environments, as well as source-based (in addition to target-based) deduplication and integrated array-based snapshots (see “CommVault unveils Simpana 9”).
Even though it was only introduced on October 5, Simpana 9 accounted for 13% of CommVault’s software license revenue in its second quarter, according to Stifel Nicolaus analyst Aaron C. Rakers.
Rakers also reported that the number of >$100,000 deals in CommVault’s Q2 were up 17% year-over-year and 45% sequentially, indicating that CommVault is well-positioned at the high end of the market vs. chief rival Symantec.
(Rakers maintained his hold rating on CommVault’s shares [NSDQ: CVLT], which were trading around $29 today.)
Related blog post: “Who will be acquired next? And the Top 10 are . . .”
Monday, November 1, 2010
Maybe it's just an anomaly driven by the whimsical interests of Internet users who are for whatever reason searching on this topic.
However, it may be a more ominous indication that SATA disk drives are wearing out and that the traditional RAID technologies used to protect them are failing. As a result, users are looking for information as to why RAID, in some circumstances, is not doing the job in their environments.
The death of RAID (or at least RAID 5) has previously been forecast by some analysts. But even now, when I look at the features of new storage arrays, the number of RAID options that they support is always prominently mentioned.
A good example is Overland Storage's recent introduction of the SnapSAN S1000. It offers at least 10 different ways that RAID can be configured (including RAID 5) on a storage array that starts at less than $10,000, so it's clear that RAID is not dead or even on its last legs.
But there is no disputing that capacities of SATA disk drives will cross the 4TB, 8TB, 16TB and 32TB thresholds over the next decade. As that occurs, it becomes questionable whether current RAID technologies will be able to protect disk drives of these sizes. If the increased interest in DCIG's 2008 blog entry (see "Data Loss on SATA-based Storage Systems -- Coming Soon to Your Company?") is any indication, the answer would apprear to be 'no.'
Am I predicting the death of RAID? Clearly not. RAID technology is as much a part of the storage landscape as tape, and odds are that innovation will continue to occur in RAID to make it a relevant technology for the foreseeable future.
Yet it was clear from speaking to a few users and storage providers at Storage Networking World (SNW) last month that new approaches to protecting data stored on larger capacity SATA disk drives are going to be needed.
One specific company that I met with at SNW was Amplidata, which is already innovating in this space to overcome two of the better known limitations of RAID, including:
•The increasing length of time required to rebuild larger capacity drives. Rebuild times for 2TB drives are already known to take four hours or longer to complete, and in some cases -- depending on how busy the storage system is -- it can take days for a rebuild.
•The need to keep all disks in a RAID group spinning so no power savings can be realized. Spin down is likely to become more important in the years to come as more data is archived to disk, with it likely becoming a function of the storage array to intelligently manage and place the archived data on these drives -- as opposed to the software -- to facilitate the spin down of drives.
Amplidata's AmpliStor distributes and stores data redundantly across a large number of disks. The algorithm that AmpliStor uses first puts the data into an object and then stores the data across multiple disks in the AmpliStor system. By storing the data as an object, Amplidata can reconstruct the original data from any of the disks on which the data within the object resides.
This technique eliminates growing concerns about the rebuild times associated with large disk drives since the original data can be retrieved and reconstructed even if one, two or even more disks fail. Also, should disk drives in the system be spun down to save energy, they do not need to be spun up to retrieve needed data since the data can be retrieved and reconstructed from other spinning disks in the system.
While it is unlikely that AmpliStor or its underlying technology will be widely adopted in the next few years, the simple fact is that increasing capacities of disk drives will eventually make technologies such as AmpliStor a prerequisite in almost any high-capacity enterprise storage system.
So in the same way that enterprise storage vendors started to adopt RAID 6 about five years ago to prevent the loss of data should two SATA drives fail, look for some variation of Amplidata's AmpliStor to find its way into enterprise storage systems over the next decade to prevent the loss of data on these ever larger disk drives. At the same time, expect RAID to find a new home on smaller storage arrays where the level of protection and speed of recovery that RAID currently provides is more than adequate.
This post originally appeared on the DCIG site (see "RAID is Certainly not Dead But its Future Looks Small").
Wednesday, October 27, 2010
Compellent announced record revenue of $42.1 million, a 31% boost over Q3 2009 revenue. GAAP net income was $3.3 million (or 10 cents a share). Gross margin was 55.9%. And the company gained 179 new customers in the quarter, bringing its total to 2, 303.
Investors jacked up Compellent's stock price by more than 32%. The shares closed at slightly over $26 in after-hours trading -- an all-time high. And that's on top of about a 10% gain on Tuesday after Reuters reported that Compellent was actively seeking suitors and had held talks with Qatalyst Partners, among others. (Qatalyst, as you may recall, shepherded the $2.4 billion buyout of 3PAR by HP, and was also involved with Data Domain when they were acquired by EMC. Isilon also recently engaged with Qatalyst, according to various reports.)
ok, so Compellent's share price run-up was obviously based on a combination of the acquisition rumors and the company's Q3 performance.
On the acquisition front, most financial analysts and speculators (as though there was a difference) still think that Dell is the most likely acquirer of Compellent. I don't think so, if only because at the price Dell would have to pay for Compellent I don't see how they would reconcile the product line with the EqualLogic line.
However, if Compellent is not bought . . . as the company's tag line says: "The future is fluid," and that may apply more to the company's stock price than to Fluid Data.
Because of my skepticism about a Dell takeover of Compellent, I maintain my #3 ranking for Compellent in my List of the Most Likely Storage Acquisitions (see "Who will be acquired next? And the Top 10 are . . .").
Related blog post: "Isilon revenue up 77%"
Tuesday, October 26, 2010
The biggest problem was that the early models didn't have onboard data management for the flash component, according to Jim Handy, director of the Objective Analysis research and consulting firm; instead, they relied on data management embedded in Microsoft's Vista. It didn't work well.
In their more recent hybrid drives, manufacturers are integrating the data management functionality. And to boost performance, they're putting in much larger caches compared to the early hybrid drives.
So far, Seagate is the only HDD manufacturer shipping hybrid drives -- the Momentus XT, which was introduced in May -- but the other drive manufacturers are expected to follow soon.
In a just-released report on the market ("Are Hybrid Drives Finally Coming of Age?"), Objective Analysis predicts that shipments will go from about one million units this year (if Seagate is successful), which represents about $120 million in revenue, to 600 million units by 2016, representing revenue of $34 billion.
Hybrid drives are typically associated with PCs, but Handy says that they can be used anywhere HDDs are used today, including enterprise-class disk arrays.
What's driving renewed demand for hybrid drives? "Hybrids provide near-SSD functionality with HDD capacity and price," says Handy.
Can anything stop the expected rapid growth of hybrid drives? "Only if people believe Steve Jobs' recent statement that hard drives are dead," says Handy.
For more information on Objective Analysis' report, see "Are Hybrid Drives Finally Coming of Age?"
(Note: The report is targeted at manufacturers, and is priced at $5,000 for a single copy and $10,000 for a site license.)
Friday, October 22, 2010
But what really caught my eye was Isilon's gross margin of 63.5% in the third quarter, which set an all-time record for the company.
Isilon's revenue performance, coupled with the acquisition rumors, is why the company recently vaulted to the #1 position in my list of potential acquisition candidates (see "Who will be acquired next? And the Top 10 are . . .").
Isilon's shares (NSDQ: ISLN) are trading at $29.21 as I write this. That's up 115% over the last three months, and up 330% over the last year. That run-up is in part due to the mania surrounding HP's $2.4 billion buyout of 3PAR, but it's also due to just plain performance.
In the third quarter, Isilon earned $0.09 per share. That compares to analysts' estimates of $0.05 per share.
The company's market value is hovering around $1.8 billion, justifying the speculation that EMC may be willing to pay around $2 billion for Isilon.
In its earnings call yesterday, company execs raised their fiscal year revenue growth estimates to almost 60%, compared to their previous estimates of about 40%.
Isilon claims more than 1,400 customers, including heavyweights such as NBC, Eastman Kodak and MySpace. Approximately 37% of Isilon's revenue comes from the media/entertainment market, 16% from life sciences, and 11% from the Web/Internet vertical market.
Isilon is known primarily for its strengths in scale-out NAS, but the company recently added support for the iSCSI SAN protocol in its OneFS operating system, enabling both file (NAS) and block (SAN) I/O under a single file system and volume (although Isilon's storage systems do not support the Fibre Channel or FCoE protocols).
We'll keep an eye on those acquisition rumors, but even without the sales and marketing clout of a larger vendor Isilon seems assured of ongoing success.
Related article: "Isilon puts multiple tiers under one file system"
Tuesday, October 19, 2010
The key addition to the 9.0 release is a technology that Arkeia now refers to as "progressive deduplication," which was previously called "sliding window with progressive matching deduplication."
You'd have to be a math wiz to understand this technology from an algorithmic perspective, but here are the basics, culled from a recent conversation with Arkeia CEO Bill Evans:
Progressive deduplication is an alternative to the older fixed-block deduplication and the newer, more common variable-block deduplication.
Arkeia's data deduplication implementation is global (vs. local), byte-level (vs. file-level), source-side (vs. target-side, although it supports both approaches as well as a mix), in-line (vs. post-processing), and content-aware. But the real differentiator is in how the software handles block sizes.
As with variable-block deduplication, the block size can be adjusted for optimal deduplication ratios, but Arkeia claims a "better" implementation that is more content-aware (or application-aware) than existing approaches. Arkeia Network Backup 9.0 software automatically adjusts block sizes based on file type in order to maximize dedupe ratios.
Arkeia acquired the data dedupe technology when it bought Kadena Systems about a year ago.
Arkeia claims two key advantages of progressive dedupe vs. traditional variable-block dedupe: It's faster (which reduces the size of backup windows) and it delivers higher deduplication ratios (which reduces storage capacity and network bandwidth requirements).
The company isn't ready to make specific performance or dedupe ratio claims, but CEO Evans reports that in internal tests using VMDK files the company achieved a 38% improvement in dedupe ratios compared to "one of the leading deduplication vendors" (which I assume to be either Data Domain or Quantum).
"We think we'll have better dedupe ratios than any other vendor," says Evans.
The proof will be in the pudding. Until we get some independent benchmark results, we'll have to take these claims with a grain of salt, but progressive deduplication appears to be an interesting technology that could take deduplication to a new level.
The Arkeia Deduplication Option will be priced at $2,000 per media, server, but will be free for companies that license Arkeia's software or appliances (physical or virtual) by December 31.
To participate in the beta program for Arkeia Network Backup 9.0 and the progressive deduplication technology, click here.
Next month, Arkeia will release a deduplication profiling tool that will enable users to measure actual deduplication ratios at various block sizes to determine the optimal block size for each file type.
Arkeia integrates backup with VMware vStorage
Arkeia acquires Kadena for data dedupe
Thursday, October 14, 2010
To understand where these folks are coming from, a little history: The company started out about six years ago as VM6 Inc., a systems integration and services provider specializing in VMware virtualization. According to COO and co-founder Eric Courville, the company started encountering a lot of remote sites that didn't have expertise in virtual servers, much less complex SANs. In addition, those sites found it difficult to justify the costs associated with VMware and SANs.
VM6 delivered its first software product -- VMex 1.0 -- about five years ago, but it was based on VMware and required a shared-storage physical SAN.
Given the requirements of remote sites and SMBs, VM6 went back to the well and introduced in September 2009 Version 2.0 of VMex, with two key changes: The software now leveraged Microsoft's Hyper-V, instead of VMware, and VMex 2.0 eliminated the need for a physical SAN.
Courville claims that VMex now provides all of the storage functionality users would expect from an iSCSI SAN, including high availability, but the software leverages DAS and eliminates the need for expensive SAN components such as switches.
VMex sits below Hyper-V and on top of Windows 2008 R2. Connections between servers provide high availability via heartbeats and failover. VMex also includes block-level replication across connected servers.
In addition to the Virtual Shared Storage (V-SAN) module, the VMex suite includes Advanced Clustering for high availability, support for Virtual Desktop Infrastructure (VDI), and virtualization management and monitoring tools. The latest version is VMex 2.1.
The company is targeting companies with less than 50 servers.
If you're looking for inexpensive, SAN-free virtualization, it might be worth checking out VM6.
Thursday, October 7, 2010
But first, let's take a step back. IBM recently acquired Storwize, which specialized in real-time data compression for primary storage. Big Blue has apparently scuttled the Storwize name as it applies to that company's products. The products now fall into the IBM Real-time Compression (note the url on that page) operation, and the products are generally referred to as IBM Real-time Compression Appliances for NAS, although IBM seems to be keeping the specific model names; e.g., the STN 6500 (up to 16 Gigabit Ethernet connections) and STN 6800 (up to eight 10GbE connections).
However, headlining today's product blitz was the IBM Storwize V7000 array which, at least for now, apparently does not have any of the technology associated with the products from the former Storwize company.
I thought I was missing something, so I checked in with Greg Schulz at The Server and StorageIO Group (formerly StorageIO), who was also confused on IBM's re-purposing of the Storwize moniker. Here's what Greg had to say:
"If IBM was trying to make a cloud storage announcement, they may have succeeded in creating a layer of fog around the renaming of the data footprint reduction (dfr) company formerly known as Storwize to Real-time Compression.
"That may be straightforward, but what's confusing, or foggy, is the use or recycling of the Storwize brand name, which was gaining ground and awareness around real-time compression for primary storage, to name an SVC-based storage virtualization system. Are they trying to say that using the V7000 is storage wise, or smart? Are they trying to differentiate from SVC, or storage virtualization, or virtual storage? Or trying to ride the growing awareness around the Storwize brand name?"
As mentioned, the confusing part is that the Storwize V7000 does not have Storwize's (the company) real-time compression technology, at least not yet.
The IBM Storwize V7000 is a mid-range storage system that incorporates elements of IBM's SAN Volume Controller (SVC), as well as Big Blue's Easy Tier technology and the XIV interface. (IBM claims that Easy Tier provides a performance improvement of up to 300% via automatic migration to solid-state disk drives, and the SVC functionality enables users to virtualize existing storage resources.) The Storwize V7000 also includes IBM technologies such as FlashCopy, Systems Director and thin provisioning.
Hardware specs of the 2U array include up to 24 2.5-inch drive bays or 12 3.5-inch drive bays, up to 24TB of capacity using 2TB SAS drives or 14TB using 600GB SAS drives, eight Fibre Channel host ports and four 1Gbps iSCSI host ports, and a RAID controller that supports up to nine storage expansion units.
In his blog post (which is a good one) on the Storwize V7000, IBM employee and SVC specialist Barry White says that the V7000 integrates:
"Something old (SVC)
Something new (the controller and enclosure)
Something borrowed (DS8000 RAID)
But it's ALL BLUE!"
For another good blog on the Storwize V7000, check out this Storage Buddhist post.
Shipments of the Storwize V7000 are slated for mid-November.
Particularly if you're an SVC fan, the V7000 may be a very cool product -- and it is Blue through and through -- but I'm still scratching my head over the Storwize branding.
"They seem to like the name Storwize so much that they've elevated it," says the Taneja Group's Arun Taneja. "The V7000 is the first in a series of products that could replace IBM's entire mid-range arrays. This is a very strategic product, but the question is: How high can it scale?"
Also at the event in NYC today, IBM introduced the high-end System Storage DS8800 array, which IBM claims is 40% faster than the DS8700. As are its high-end competitors, IBM has moved away from 3.5-inch disk drives, going only with 2.5-inch, 6Gbps SAS drives on the DS8800, which can be configured with up to 1,056 drives for a total capacity of 634TB. The array can also be configured with SSD drives, and support for Easy Tier is expected next year.
Monday, October 4, 2010
For the following list of the Top 10 storage acquisition candidates, I factored in the opinions of InfoStor.com readers, which I received after posting "The Top 10 storage acquisitions of 2010." Interestingly, the majority of the reader responses came from channel professionals -- VARs and integrators -- leading me to believe that channel pros are much more interested in mergers and acquisitions than are end users. I also factored in opinions from industry and financial analysts. And topped it off with my own misguided opinions.
(Note: The ticker symbol links below take you to that company's entry page on the InfoStor Market Index, which provides up-to-the-minute info on the company's stock as well as company- and competitor-related news.)
#1 -- Isilon
Isilon scooted to the top of this list because of two recent developments: (a) The company hired Qatalyst Partners to solicit potential acquisition offers. Qatalyst was the advisor to Data Domain when EMC acquired Data Domain, and Qatalyst also shepherded the HP-3PAR acquisition. (b) The NY Post reported on Friday that EMC may be close to acquiring Isilon for about $2 billion (see "EMC in exclusive talks to buy Isilon").
Isilon is known primarily for its strengths in scale-out NAS, but the company recently added support for the iSCSI SAN protocol in its OneFS operating system, enabling both file (NAS) and block (SAN) I/O under a single file system (although Isilon's storage systems do not support the Fibre Channel or FCoE protocols).
In addition to EMC, financial analysts have cited Dell, HP and IBM (and, less likely, Oracle or Cisco) as potential acquirers of Isilon (NSDQ: ISLN).#2 -- CommVault
I don't understand why CommVault's (NSDQ: CLVT) stock jumped so high during the HP-Dell-3PAR bidding war, but it did. I guess it was because CommVault has been an acquisition speculation darling for years.
CommVault has done a great job stealing revenue from the big four vendors in the data protection space, in part because CommVault's underlying architecture is newer and designed better for rapid enhancements, as evidenced in the recent release of its Simpana 9 software (see "CommVault unveils Simpana 9").
Conventional wisdom on Wall Street has Dell (NSDQ: DELL) as the most likely suitor under the assumption that none of the leading backup/recovery vendors (EMC, Symantec, IBM, CA) would be interested in CommVault, but I'm not so sure about that. Any of those vendors would have to eat some crow if they acquired CommVault, but they might be better positioned for the long haul.
Another possible CommVault suitor: NetApp which, by the way, is the #4 storage software vendor -- ahead of CA and HP (see "The Top 6 storage software vendors").
In addition to the fact that many financial analysts put the company at the top of their storage acquisition target lists, CommVault earns the #2 spot on this list because I think the storage M&A focus is going to shift from hardware to software. And with CommVault, it's hard to argue with low debt, high growth and high margins.
#3, #4 -- Compellent, Xiotech
Isilon, Compellent and Xiotech have all been cited often as potential Dell acquisitions post-3PAR/HP. I doubt it, because Isilon/Compellent/Xiotech are gap pluggers rather than the game changer that 3PAR would have been for Dell. Isilon, Compellent or Xiotech don't give Dell the high-end array technology that would enable Dell to go up against EMC/IBM/Hitachi, although they would be good complements to Dell's EqualLogic line. On the other hand, any one of these disk array vendors would help Dell doff its "Dude, you're getting a Dell" image.
Given their technology differentiators, and the difficulty of being a relatively small player in the contracting disk array market, it's likely that one or both of these vendors will be acquired by someone.
Xiotech, which is not publicly traded and has a nice differentiator with its Intelligent Storage Element (ISE) technology, would be the least expensive buy in this category.
The question with Compellent is whether the company has enough differentiation from what the potential acquirers already have in their portfolios.
#5 -- Permabit
Earlier this year, Permabit introduced an "OEM embeddable" version of its data deduplication software. dubbed Albierio. The company seems to be off to a good start with its OEM strategy, having already racked up reseller deals with BlueArc and Xiotech (see "Data deduplication: Permabit finds success with OEM model").
As the last standing "independent" data deduplication player, Permabit could thrive with its OEM model, but in light of the IBM-Storwize and Dell-Ocarina acquisitions in the data reduction space I think Permabit is an attractive acquisition candidate. The company's focus now is on deduplication for primary storage, but there's no reason Permabit's technology couldn't be used across all storage tiers -- and that could be very attractive for some of the larger storage vendors that have a diverse mix of data reduction solutions in their portfolios. Of course, that assumes that a "one size fits all" approach to data reduction is the way the larger vendors want to go in this space.
#6 - Brocade
IBM has been mentioned most frequently as a potential suitor for Brocade. That might have changed recently with Big Blue's announcement that it plans to acquire Blade Network Technologies (BNT), making it less likely that IBM would go for Brocade.
BNT specializes in blade and rack Ethernet switches, so it's not an overlap with Brocade's business, but the IBM-BNT acquisition will still dampen speculation that IBM will scoop up Brocade. More likely, perhaps, IBM will go after Juniper Networks. (Both Brocade and Juniper are IBM partners and, making it even more interesting, Juniper is a BNT partner.)
If we take IBM out of the Brocade acquirer lineup, that would leave Dell as the most likely acquirer (although Dell is also tight with Juniper). But that goes to the heart of the question of whether Dell will, post-3PAR, try for another disk array vendor (Isilon, Compellent, Xiotech?), software vendor (CommVault?) or surprise everyone and light out into LAN/SANland via Brocade.
Making a Dell acquisition shift toward networking more likely, Dell recently hired a former Cisco exec -- Dario Zamarian -- to run its networking business. Assuming Dell turns to networking in its acquisition spree, it's a 50/50 bet between Brocade and Juniper, according to some Wall Street wags.
In an article posted on MarketWatch (see "Brocade targeted by M&A rumor mill"), Wedbush analyst Kaushik Roy was quoted as saying, "It makes a lot more sense for Dell to buy Brocade than IBM. With Dell, it's a no-brainer. If Dell has half of a brain, they should be taking Brocade out right now."
Then again, if Dell goes after Brocade (NSDQ: BRCD), they could wind up in another crazy bidding war with, say, IBM or Oracle. And Dell's been there, done that.
In addition to IBM and Dell, analysts have cited Oracle and Hitachi as possible acquirers.
#7 -- BlueArc
Hitachi Data Systems (HDS) has been mentioned as a potential suitor for BlueArc, but it doesn't seem to be in HDS' genes to take the acquisitions route. But HDS and BlueArc do have a tight relationship and BlueArc has some attractive technology differentiators.
# 8 -- FalconStor
Only one of our readers mentioned FalconStor as an acquisition target, and it was in the context of the company potentially being acquired by HDS or NEC, but FalconStor would be relatively inexpensive and the company has great technology (although not everybody knows it because FalconStor's software is often sold "under the covers" by its OEMs and resellers).
In addition to getting storage management and data protection software (VTL, CDP, data deduplication, replication, etc.) across a variety of product lines, an acquirer could put the hurt on a lot of competitors because, although a relatively small company, FalconStor's tentacles reach across a lot of vendors via its reseller deals.
The possibility of FalconStor (NSDQ: FALC) being acquired heightened recently with the resignation of ReiJane Huai, the company's CEO. In the wake of the resignation, FalconStor tapped Jim McNiel as interim CEO and president. Some analysts think that McNiel may be more open to acquisition than was Huai.
#9 -- Symantec
The latest rumors regarding Symantec centered on Microsoft as a potential acquirer (see "Is Microsoft Looking to Buy Symantec?" on InfoStor sister site eSecurity Planet).
However, that (unlikely) move would be more for Symantec's security product line, rather than its storage software, and would be in response to Intel's acquisition of McAfee.
Given the breadth of Symantec's product line, and the fact that the company would be expensive and is not a pure-play storage vendor, I put Symantec low on this list. I don't see Symantec being bought any time soon, if only because of the company's huge market cap.
#10 -- NetApp
Rumors about NetApp (NSDQ: NTAP) being acquired have existed as long as the rumors surrounding CommVault and Brocade have, maybe longer. I think the time to buy NetApp is long gone, but due to the persistence of the rumors the company still makes our Top 10 list.
I see NetApp more as an acquirer than a target. The problem with that (for NetApp) is that, if the IT market does contract down to five or six soup-to-nuts vendors, NetApp doesn't have a chance of making that list even if it does dip into its deep piggy bank to make some acquisitions. It's that logic that keeps NetApp on our Top 10 acquisition targets list.
One other company that occasionally comes up in acquisition conversations is Quantum. This would be for the company's data deduplication technology and products. However, an acquirer would also get Quantum's tape business, and I don't think any of the likely acquiring vendors is looking to add tape to their portfolio.
And in the Least-Likely-But-Often-Mentioned category of acquisition targets: EMC. The Wall Street Journal reported last week that Oracle may be eying EMC (see "EMC Shares Rise On Oracle Buyout Rumor"). Sure, that would fill in the Tier-1 storage hole in Oracle's portfolio and, more importantly, give them VMware, but I still think the Oracle-EMC acquisition speculation is ludicrous.
Related blog post:
The Top 10 storage acquisitions of 2010
Monday, September 27, 2010
The full results of the survey won't be available until next month, but I chatted with Doug Hazelman, Veeam's senior director of product strategy, about some of the preliminary results of the survey. Here are a few snippets:
Although a virtual machine (VM) can be built and deployed in minutes, performing a full recovery of a backed-up VM still takes nearly five hours on average. That compares to an average of six hours required to recover a physical server -- not much of an improvement.
Based on the survey results, 47% of full server recoveries are being performed merely to recovery a single file or application item (as opposed to recovering just the file or application item).
Also according to the survey, 63% of IT organizations experience problems (e.g., failed media, inability to start VMs after recovery, etc.) every month when attempting to recover a server. And failed recoveries cost enterprises more than $400,000 per year on average. However -- and here's the kicker -- only 2% of all server and VM backups are tested for recoverability each year, and on average these tests are performed once every two months (which translates into as many as 60 days of bad backups).
And in a related finding, the survey respondents report that testing the recoverability of a single backup takes IT teams approximately 13 hours. Not surprisingly, IT managers report that lack of human resources is the #1 reason why they don't test the recoverability of backups more often.
One more tidbit: hardware failure is the most common reason (cited by 68% of the survey respondents) why IT organizations need to recover servers and data, followed by general IT problems such as misconfiguration (63%) and user error (56%).
You can register on the Veeam site to receive a full copy of the VMware Data Protection report when it becomes available next month.
Wednesday, September 22, 2010
Afterwards I heard from a lot of data protection vendors wondering if I had ever heard of them (presumably because I hadn't written about them). I was aware of most of them, but one exception was AppAssure, which has been shipping its flagship Replay software suite since 2007.
If you're open to considering backup vendors outside of the Big Four (and given the relatively high readership of that Cofio post, many of you are), AppAssure is worth consideration.
Many of the company's executives hail from Symantec/Veritas (and AppAssure says that about 75% of its customers are former Backup Exec users) with some execs tracing their roots back to W Quinn Associates (which was acquired by Precise Software which was acquired by Veritas which was acquired by Symantec).
AppAssure claims 13 consecutive quarters of growth, and as of last month the company had exceeded its total 2009 revenues (which grew 268% over 2008 revenues), according to Steven Toole, AppAssure's chief marketing officer. The company also claims that Replay has accounted for more than 56 million backups on five petabytes+ of data.
The application-aware Replay4 software backs up the entire application stack, including app objects, files, operating systems and blocks with "near" continuous data protection (snapshots every 15 minutes).
The suite includes standard functionality such as backup and recovery, disaster recovery and cloud recovery, as well as all the specific features you might need in a data protection suite, including the elimination of backup windows, off-host processing, data deduplication and compression, VM failover, bare metal restore, corruption detection, virtualization, replication, fast backups (8GB/minute), etc.
And most of those features are bundled in what appears to be attractive pricing (exceptions include replication and virtualization, which are priced separately). For example, the base price is $899 per protected server, and that's for a perpetual (non-recurring) license. Specific pricing: $2,099 per VMware host with an unlimited number of guests, $1,499 for Hyper-V, $1,499 for SQL Server, and $2,099 for Exchange.
If that sounds intriguing, check out AppAssure's web site where you can get a free trial version of Replay.
Monday, September 20, 2010
Around the same time (early summer), Permabit introduced an OEM version of its data deduplication software, dubbed Albierio, that storage hardware and software vendors can integrate into their products.
Permabit is among only a few "independent" deduplication vendors that are taking the OEM route rather than the acquisition route (although I still say the company is an attractive takeover candidate).
So far, Permabit's strategy seems to be paying off. The company has inked two OEM deals -- with BlueArc about a month ago and, today, Xiotech -- and I understand that Permabit has a few more OEM deals that will be announced over the next couple months.
Permabit claims zero performance degradation with its Albireo deduplication software, and both BlueArc and Xiotech officials cited Albireo's performance, as well as scalability and data "safety" (Albireo doesn't alter data) in announcing their OEM agreements. BlueArc and Xiotech are a good start for Permabit, although it's unclear how long it will take either vendor to integrate Albireo into their systems.
Both the BlueArc and Xiotech deals are focused on data deduplication for primary storage, as opposed to secondary or backup storage. That's where Permabit is currently focused, although Albireo software can be used across all storage tiers, and can be implemented with block, file or unified (block + file) storage.
I recently caught up with Tom Cook, Permabit's president and CEO, to talk about trends in the capacity optimization market.
I asked him about the fact that some of the large storage vendors seem to be embracing multiple capacity optimization techniques for different tiers of storage and, in some cases, carrying a mix of homegrown and OEM'd solutions for data deduplication and/or compression.
"We're still in the very early stages of data optimization, where you have single-function devices such as D2D and VTL appliances," says Cook. "But over time, the big players will try to get to a single data optimization solution that covers all storage tiers."
"Vendors will drive toward total efficiency, where's there's no re-hydration of data," says Cook. "And we believe that the best place to start is on primary storage. It's like New York: 'If you can make it there, you can make it anywhere.'"
Dell to acquire Ocarina for data deduplication
IBM scoops up Storwize
Tuesday, September 14, 2010
IDC recently released its quarterly report on the market, and EMC held on to its #1 ranking with a 24.4% market share on Q2 2010 revenue of $722 million, followed by Symantec at #2 (16.5% share, $488 in revenue), #3 IBM (13.9%, $410 million) and #4 NetApp (8.7%, $256 million).
Rounding out the Top 6 were CA and HP in a statistical tie. CA had a 3.6% market share on revenue of $108 million, and HP had a 3.3% share with revenue of $97 million. The only change in the lineup between Q2 2010 and Q2 2009 was a switch in positions between CA and HP.
In terms of revenue growth over the last 12 months, there were four gainers and two losers. Gainers included EMC (+13.3%), IBM (+10.6%), NetApp (+6%) and CA (+2%). Symantec (-6.9%) and HP (-10.3%) declined year-over-year.
Overall, the storage software market hit almost $3 billion in the second quarter, a 3.3% growth vs. the same period a year ago.
IDC segments the storage software market into eight product categories. Of those, the segments experiencing the most growth over the last year included storage infrastructure (+12.7%), archiving (+8.2%), storage management (+5.8%), and data protection and recovery (+4.9%).
For more details, see IDC's press release, "Storage Software Market Delivers Continued Growth in the Second Quarter."
Related blog post:
The Top 5 array vendors: HP #4, Dell #5
Wednesday, September 8, 2010
For the second quarter of this year, EMC retained its #1 ranking in the external disk systems space with almost twice the market share of #2 IBM. On Q2 revenue of almost $1.3 billion, EMC held a 25.7% market share compared to IBM's 13.6% slice on revenue of $680 million.
NetApp was #3 with an 11.4% share on revenue of $571 million, followed very closely by HP with an 11.3% share on revenue of $567 million (which is essentially a dead heat between the two vendors).
Rounding out the Top 5 was Dell, with a 9.4% share and revenue of $472 million.
Liz Conner, IDC's senior research analyst, storage systems, estimates 3PAR's share of the market at 0.58%.
Depending on how quickly HP can ramp the 3PAR revenue stream, it won't be long before HP is firmly in the #3 spot, followed by NetApp at #4.
Or maybe not. In terms of revenue growth in 2Q10 vs. 2Q09, NetApp was the big gainer, with an impressive 55.3% revenue growth rate, followed by EMC with a 40.6% growth rate. HP only had 20.9% growth year-over-year, and Dell posted a 17% increase. IBM was the laggard at 10.9%.
The "others" category in the external disk array market continues to decline. In 2Q09, "others" accounted for 33.5% of the market ($1.6 billion in revenue), but in 2Q10 that share slipped to 28.6% ($1.4 billion). So the "others" market share is approximately the same as EMC's share.
That stat will no doubt throw more fuel on the speculation fire regarding which disk array vendor(s) will be acquired next (e.g., Compellent, Isilon, Pillar, Xiotech, etc.).
If you add up all (external + internal) storage systems revenue, the market share rankings shift: HP (19.3%), EMC (19%), IBM (15.8%), Dell (12.3%), NetApp (8.4%).
The NAS and iSCSI SAN sectors continue to rack up impressive growth figures. The combined NAS+iSCSI market grew 29.2% year-over-year in the second quarter, to $4.2 billion. EMC had a 28.9% share, followed by NetApp at 13.6%.
The NAS market posted 51.1% growth year-over-year, with EMC taking a 45.6% share followed by NetApp with 25.2%.
And the iSCSI SAN market grew 49%, with Dell in the lead with a 32.9% slice, followed by HP, NetApp and EMC in a statistical tie for second place.
Overall, the external disk storage market grew 20.4%, topping $5 billion in the second quarter.
For more details, see IDC's press release: "Disk Storage Systems Market Sustains Strong Double-Digit Growth Across All Sectors in Second Quarter."
Thursday, August 26, 2010
After what appears to be a lot of under-the-covers negotiations, HP ended the drama with a $33-per-share, $2.4-billion acquisition offer (see "Dell Ends 3PAR Talks After HP's $2.4 Billion Bid" on Enterprise Storage Forum).
After what we've seen over the past couple weeks, I'm hesitant to call this a done deal, but that appears to be the case.
So HP gets 3PAR's crown jewels of virtualization, cloud computing capabilities, storage/data tiering and thin provisioning (not to mention ASIC technology). The question now is: How will HP fold 3PAR's systems into its existing disk array lineup? 3PAR overlaps big time with HP's venerable EVA line, and to a lesser degree with the high-end systems that HP OEMs from Hitachi Data Systems. Something has to give, although it will probably be awhile until HP provides details.
For Dell, the question is: Now what? Many observers have speculated that Dell will go after another disk array vendor (Compellent, Isilon, Xiotech?), but none of those companies are a replacement for what Dell had in mind with the 3PAR bid. Maybe Dell will turn to networking (Brocade?). Or software (CommVault?).
Dell put its cards on the table, and finally folded. For now. Which table will it sit down to next?
In a side show next to the three-ring HP-Dell-3PAR circus, litigious Crossroads Systems yesterday filed a patent infringement lawsuit against 3PAR and, according to an article on PC World, D-Link, Rorke Data, Chelsio Communications, DataCore Software, iStor Networks and American Megatrends. According to that article (see "3Par Faces Lawsuit as Bidding War Continues"), the suit involves a patent for a storage router that provides virtual local storage on remote storage devices.
Tuesday, August 24, 2010
Will Dell counter? At first, I thought this was a done deal for HP. In an acquisition context, a 33% raise leads to an opponent folding. Then again, maybe this “I’ll see your $1.15 billion and raise you $450 million” is just these guys anteing up. This pot could get close to, or exceed, $2 billion.
Yes, I think Dell will counter, but I don’t think it would be wise. And the reason for that leads me to the next question.
Is 3PAR better for HP or Dell? I think it’s better for HP, but this is highly arguable.
HP has a solid high-end lineup with the HDS OEM deal, but I doubt that HP’s Dave Donatelli (formerly with EMC) is a reseller type of guy. 3PAR gives HP its own technology, and possibly a better weapon against EMC. The question would then be: What would HP do with the HDS line? And if HP follows Oracle’s suit on that, What becomes of HDS? (HP execs said that they would continue with the HDS partnership, business as usual.)
HP knows how to sell really high-end gear, which can’t be said about Dell. Plus, if Dell winds up acquiring 3PAR it would have to then follow up with a string of risky, blockbuster acquisitions to really own the IT stack. And even Dell’s war chest would be seriously depleted after a long series of billion dollar acquisitions.
If this deal pushes upward of $2 billion, there would be a lot of pressure on Dell to prove that it can play in the IT stratosphere, which is questionable. Dell obviously had a lot of success with the EqualLogic acquisition, but instead of going right to the very high end with 3PAR (and thus going head-to-head with EMC), it seems like a stepping-stone approach might have made better sense (e.g., acquiring BlueArc or Compellent or Isilon or Xiotech, etc.)
Will other suitors jump in? Doubtful. The only possibilities are IBM, EMC, NetApp or Oracle, and of those only Oracle is rich enough and nuts enough (see “Who will Oracle acquire next?” ). If Oracle wants another storage product line, my money is on (and Larry’s money is in) Pillar Data.
If IBM or EMC made a move for 3PAR it would (a) be admitting that their existing high-end arrays aren’t up to snuff and (b) create too much confusion among customers and overlap in product lines. For $1.6 billion+, there are much better acquisitions for IBM or EMC to make.
NetApp could probably afford 3PAR, but does NetApp really want to get into another big-time bidding war? Been there, done that.
I think we’re in for a bidding war, but there will only be two players at the table.
Related blog post:
HP’s bid for 3PAR not its first
Recommended blog: Stephen Foskett’s “Everyone Loves 3Par: Here’s Why”