When Sun revealed its open source storage push this week, some industry observers wondered about its business model. In other words, how can Sun make money on open source storage products?
Then Sun reported its earnings Thursday night, and it became clear that its storage business isn’t exactly rolling in dough these days anyway.
Sun’s storage products generated $530 million in revenue last quarter, down 5.4 percent from a year ago and $100 million short of its target. Big-ticket items such as tape libraries and high-end disk systems were down in a quarter in which EMC and IBM reported increases. Server revenue also fell short by $100 million, making it a disastrous period for the new combined servers and storage unit.
Overall, Sun lost $34 million in the quarter compared to a profit of $67 million the year before. On the earnings call, Sun execs said they would be restructuring to the tune of 1,500 to 2,500 layoffs.
Can open source save this sinking ship? Sun CEO Jonathan Schwartz seems to think so. Open source was a common theme of his earnings call, with open storage getting its share of attention with statements such as: “We have a great variety of new Open Storage innovations [entering] the market within the next few quarters.”
Schwartz didn’t talk much about how Sun will make money on open storage, except to emphasize that Sun would save money on R&D by having a common open platform for all of its servers and storage systems. Layoffs are expected to save Sun $100 million to $150 million a year, although it’s not clear how much of the reduction will be in storage.
It remains to be seen what the quality of open storage products will be, but Sun has little to lose. It’s tried a lot of things over the years to jumpstart storage sales, including paying $4.1 billion for tape library market leader StorageTek, and nothing has worked. Sun OEMs systems from Hitachi Data Systems, LSI and Dot Hill, and usually has less success than other vendors who sell the same systems. For a while Sun planned its storage future around the 6920 midrange system, which it billed as a virtualization product and an EMC Clariion killer. Customers yawned, and Sun sold the technology to HDS last year.
Now its storage plans revolve around a large DAS system called Thumper and open source software. Considering its track record, things can’t get much worse, can they?
The economy has been on the mind of just about everybody recently, and with good reason. Gas at near record highs, unemployment rising, housing values reportedly dropping, the credit crunch and foreclosures numbering in the bazillions it is easy to see why people are not exactly upbeat about the state of our economy.
In the storage market, however, it’s looking like a blockbuster year. EMC and others are reportedly on track to meet or beat financial analysts’ estimates, and that leads me to today’s blog.
As it turns out, the impetus for this blog post was my recent attendance at a DR seminar put on by Storage Decisions featuring Jon Toigo. Looking around the room, I couldn’t help but think of what it looked like in the early days of “network administrators” when people didn’t think of network pros as any different from the server guys. Today, the storage admin is being called on to be part lawyer, part business analyst, part networking guru and all-knowing about all things storage, but there are very few companies with a dedicated storage team (outside the Fortune 500’s that have Exabytes of storage to manage).
For the most part (and please chime in with your experience) storage folks are still viewed as “server guys”. This is, of course, changing, and I wouldn’t bring it up if there weren’t a bigger point to be made: if you do a quick scan of Monster, Dice or Jobcircle, there are more and more listings specifically calling for a “Storage Administrator”. Storage is fast becoming the segment to be in–the information infrasructure could not function without it, and it is increasingly becoming the focus of much planning and resource allocation, in terms of both time and money. Talk to most companies, and they have storage budgets that are going up even in a down market, and they are hiring people to dedicate to the task of storage. Storage pros are more highly valued, and their pay is going up.
So what does this have to do with DR? DR is, at its basic level, moving data from one place to another, on a regular basis, far enough away that if you had a disaster you could recover your data and continue operations in the face of a disaster. This, in almost every case that I can think of, requires storage, storage networking technologies and someone who knows enough about them to set it all up and keep it working in a changing environment. Hence all the storage pros in the room vs business types that normally involve themselves in DR.
Toigo put on a great presentation. It was filled with a ton of valuable information and even if you have nothing to do with the DR planning and implementation at your company, I would enthusiastically recommend attending one in your area. I walked in thinking I had a passable grasp of DR best practices and walked out realizing I had barely scratched the surface, and that as a storage professional I needed to understand more about business practices as they relate to DR.
For example, Toigo discussed what a data model was and not only how to build one but suggestions on explaining it to non-technical analysts so we could all use it together to ultimately build a workable DR plan around valuable data instead of putting together a set of technologies to make our systems highly available but unable to really recover from a disaster. And it’s the storage guy who should be taking the lead on that.
Think of the value you bring to the table when you can not only provide the information infrastructure, but also assist in developing a DR plan that will keep the business functioning, and generating revenue in a disaster. In the process, you can also create things that have intrinsic value to multiple business units–think of what information security can do if they know what a document or document type is worth as compared to other documents. My fellow storage pros, I’m seeing a bright future for us.
Compellent joined other storage companies including EMC and Data Domain in reporting a strong first quarter despite a down economy. The company’s revenues more than doubled year over year to $18.3 million, growth of 107% over the first quarter of 2007 and 9% over the previous quarter.
The company also is still a ways from profitability, and lost $1.2 million last quarter despite the increased revenue. CEO Phil Soran said on the company’s earnings call that this is because Compellent is growing and is adding operating expenditures such as salaries for new employees. Soran said he expects Compellent to be profitable by the second half of this year.
With the rest of the country in financial turmoil, how are storage companies staying strong? “Storage is the last thing that gets cut from the IT budget,” was Soran’s answer. I would also imagine it’s because storage has always been a conservative market–it doesn’t have as far to fall as some other markets.
Another thing benefitting Compellent, according to Soran, is the acquisition of midrange disk array competitor EqualLogic by Dell. It’s been well-publicized that EqualLogic channel partners have been wary of the deal, if not downright alienated by it, because of Dell’s poor reputation in the channel. Soran declined to give any specific numbers around how many channel partners have defected or how much new business it accounts for, but volunteered anecdotally that Compellent is seeing more large EqualLogic channel partners looking its way as a result of the Dell deal.
Still, Soran says the company has a ways to go when it comes to gaining that mind share. Echoing some of NetApp’s statements when it rebranded itself earlier this year, Soran said Compellent does well when companies look at its products but often doesn’t get brought to the table.
I also asked him whether or not Compellent is seeing significant business as a tier-2 disk array in large shops. He said yes, but also declined to break out any numbers.
Soran attributed Compellent’s growth to the attractiveness of its consolidation and thin provisioning features in a down economy, similar to the power and capacity savings that have reportedly kept money flowing in to Data Domain’s coffers. But Soran said Compellent’s chief competitor remains EMC, which doesn’t yet offer many of the features he was referring to–and EMC also reported a stronger-than-expected first quarter.
“They have a good brand,” Soran said.
John Thompson must have known the question was coming. The Symantec CEO certainly heard the rumors. So when he was asked Wednesday night during his company’s earnings conference call about selling off parts of his company, Thompson couldn’t have been clearer.
“Contrary to popular rumor, we have no plans to divest of anything,” he said. “None.”
The rumors mainly involved the storage products that Symantec acquired from Veritas three years ago. And they were widely circulated. According to an Associated Press earnings preview story that ran this week:
Analysts are particularly interested in the possible sales of backup and recovery software product NetBackup and the company’s non-Windows Data Center Foundation, which comprises of storage and server management products.
Several technology bellwethers, including IBM, Hewlett-Packard and EMC have been named as potential buyers for Symantec’s storage products, including NetBackup.
AP could have added two other bellwethers who have been mentioned as suitors of all or some of the Symantec storage products – Oracle and Microsoft.
From the tone of Thompson’s voice when he answered the question, he’s not happy with the rumors. Yet Symantec is at least partially to blame. There have been frequent reorganizations since it bought Veritas, usually accompanied by layoffs. Symantec admitted a large layoff in April but would not give details. This left the door open for scared Symantec employees, disgruntled former employees and opportunistic competitors to attempt to fill in the details. And Symantec execs have talked about getting rid of poor performing units on previous earnings calls.
But Wednesday’s call was upbeat. Symantec reported outstanding results all around, and storage was front and center. Email archiving, backup, and storage management were among the product segments that posted double-digit year over year growth. Thompson and COO Enrique Salem talked of a bright future for Net Backup 6.5, Backup Exec 12, and Storage Foundation. They emphasized Symantec’s encryption and virtualization capabilities and gushed about three hot storage areas where Symantec has hardly been a pioneer: data deduplication, continuous data protection and software as a service (SaaS).
Symantec’s earnings were impressive in current economic conditions, although with 53 percent of its revenue from international sales, it took advantage of favorable foreign exchange rates against the dollar. Symantec gained share from its major rival EMC on the backup front, with 11 percent year-over-year growth compared to EMC’s 8 percent growth.
The question now is whether the strong storage performance will prompt Symante execs to forget about spinning off any pieces, or will it only add to the value of a possible sale? Thompson’s take is nothing is for sale. Despite what you might have heard.
Even as we continue to debate whether or not tape is dead, indicating at least that its salad days are probably behind it, some of the most interesting innovations in tape technology I’ve seen are happening right now.
For example, there’s Index Engines’ tape indexing and search software. If you’d been able to give backup administrators the ability to do a keyword search across dozens of backup tapes to identify what tapes should be restored, as well as the ability to extract single relevant files from said tapes, we might not have ever heard of a VTL.
I’d put the latest development from ediscovery services provider RenewData into that category as well. Renew says its tape-processing systems now only need to take a single pass through a given piece of linear media. Renew previously needed two or three passes, requiring its admins to mount tapes in proper order and reassemble data as it was ingested. The single-pass process will reduce the time it takes to find relevant information stored on its clients’ tapes.
The single-pass process is made possible by software that allows that data to be reassembled on the back end. Renew is not selling that software, except as part of the back-end of its hosted services. Renew’s VP of marketing Bob Little says the company doesn’t have any plans to offer it as an on-premise product.
But I have to wonder if someone else won’t find a way to develop something similar. I also wonder, if the tape space keeps coming up with finding new ways to access data randomly on linear media, whether this disk vs. tape debate could get much more interesting.
Hardly a day goes by without a new storage service rolling out. On Monday, it was IBM’s turn to launch two storage services as part of its portfolio of services for midsized customers – organizations with 100 to 400 employees and a handful of Windows servers.
The interesting thing about IBM’s Remote Data Protection Express and E-Mail Management Express offerings is they are the first new services IBM has launched from its Arsenal Digital Solutions acquisition in December. The Arsenal brand is gone but the remote data protection service is the same one that Arsenal offered, even still including data deduplication from EMC’s Avamar.
The email archiving service is something Arsenal was working on at the time of the acquisition, pushed to market quicker with IBM technology.
“Email management is a new offering [for Arsenal],” said Arsenal alum Brian Reagan, now an IBM Information Protection Services executive. “Under the covers we’ve started to adopt and integrate more IBM offerings.”
The email service covers Exchange and Lotus Notes, and Reagan said database and unstructured data archiving services are in the works.
Since Arsenal was into managed storage services long before anybody talked of clouds and SaaS (software as a service), I asked Reagan if he thought IBM should have kept the Arsenal brand. He said Arsenal did have a name for itself and partnerships with AT&T and other large providers, but IBM is a pretty recognizable name too.
“We get the benefit of IBM’s brand,” he said. “As Arsenal, we would have to spend twice as much money to get the attention of customers because they didn’t know who we were.” Reagan pointed out that IBM ran advertisements for its services during the PGA Masters broadcast. “That was something Arsenal could only dream of,” Reagan added.
Then again, you don’t have to be IBM to attract attention for storage services these days. Everybody’s getting into the cloud act, and Reagan says the glut of offerings have served mostly to confuse customers.
“There’s a tremendous amount of customer interest,” he said. “The downside is, it’s created confusion. Some of the really low end players that only service the consumer end of the market have clouded the picture. They’ve confused people wondering what the difference is between low end service that’s priced too good to be true and real resilient service.”
In other words, it will take awhile before enough sun shines on cloud computing so we can really know what to expect.
Until now, all we’ve known about Pi Corp., the startup EMC purchased in February, were that it was still in stealth, and that its software was meant to provide access to content on multiple types of Web-enabled devices.
While that description makes sense, it also is vague enough to leave open the possibility of several different directions that type of technology could go in. I had imagined it transcoding things like movies and music for streaming delivery to iPhones as well as PCs, for example, or reformatting Word documents to be read on both laptops and Blackberries by corporate mobile workers.
But a little more information has come out about Pi since the acquisition, at PiWorx.com. You can download a product demo, which unfortunately my system specs on my standard-issue laptop don’t support, but there’s also a walk-through of the software’s features and screenshots.
What it reminds me of most is Flickr — except with music, documents and other content types. Rather than making a “photostream,” it looks like it can create multimedia personal sites or collections for sharing that bring together those different types of content.
It also looks like the product would offer services like those available with network-enabled external hard drives for consumers that allow content to be accessed from the Web, but on a much bigger scale. Those external devices allow access to every piece of data in the repository to the owner of the password/admin rights for the system. This would let people share content selectively with more complex authentication.
So, let’s just say you’re a news writer at a storage conference and you need to see that PowerPoint from last week, but you’re in the middle of the show floor and your laptop’s upstairs. Let’s also say you don’t own an iPhone or PDA. The idea behind PiWorx, if I’m understanding correctly, is that you could go to one of the stations set up for people to check their Webmail, log in and call up the PowerPoint online.
I still wonder if this software will be sold to corporations for internal use as well as deployed as part of EMC’s Fortress? Despite an emphatic PiWorx message about data security, I still think storage admins would be more keen to use this type of thing if the content delivery networks and storage repositories were their own, and access to the information was limited to the employees or authorized partners of the company.
Then again, I’m not a storage admin. Any thoughts from those in the peanut gallery?
In a bit of a Friday surprise, storage systems vendor Nexsan Technologies today filed for an initial public offering (IPO) with the SEC.
The move is surprising not because of Nexsan’s financial situation, but because storage IPOs are so 2007. IPOs in general are down this year with Wall St. fearing recession and waiting to see if the problems financial services firms are having spread to other sectors.
On the storage front, no company has gone public since 3Par last November. NAS vendor BlueArc and consultant firm Glass House Technologies filed for IPOs late last year but neither appears close to actually going public. Neither company has updated their S-1 filings with their latest earnings, which suggests they’re in no hurry to test the market.
Financial analysts and other industry sources say companies such as DataDirect Networks, Copan, and LeftHand Networks are eager to go public and probably would have filed by now if market conditions were better.
But Nexsan is brave enough – if brave is the word – to give it a shot. The systems vendor that specializes in SATA arrays for archiving probably isn’t as well positioned as some of the above mentioned vendors. Nexsan lost $3 million on revenue of $49.8 million for the year that ended June 30, 2007, and dropped another $2.3 million on $30 million in revenue during the last six months of last year. According to its filing, Nexsa has lost a total of $35.1 million.
Histories of losses haven’t stopped other storage companies from going public. Those that took the plunge in 2006 and 2007 – 3Par, CommVault, Compellent, Data Domain, Double-Take, Isilon, and Riverbed – had a quarter of profit at the most and almost all of them had never recorded a profitable quarter before their IPO. The same goes for EqualLogic, which had its IPO short-circuited by a $1.4 billion acquisition offer from Dell.
And while Nexsan‘s finances don’t look much different than those of Compellent’s or Isilon’s before they went public, at least there was a bullish IPO market when Compellent and Isilon went out.
So Nexsan’s progress deserves keeping an eye on. If they do make it public for a decent price, it could spark another rush like the storage market saw in 2006 that lasted until late 2007.
The term “vendor lock-in” is rarely used in a good way by storage buyers. It usually means you’re stuck with products from one vendor, making it difficult to switch if you’re unhappy or something better comes along.
Still, with probably more options for storage products than ever before, most companies still buy all their primary storage from one vendor. That’s according to a Forrester report, “Consolidate Storage Vendors to Reduce Complexity,” released this week.
A Forrester survey of 170 companies ranging from SMBs to large enterprises in North America and Europe found that more than 80 percent bought their primary storage from one vendor over the last year. That includes 64 percent of the companies with more than 500 TB of raw storage.
The report, written by analyst Andrew Reichman, says using more than one primary storage vendor can make it more complex to manage, provision and support the storage environment. And while using multiple vendors can often bring better pricing, buying from one vendor can result in volume discounts.
“You may have tried to contain costs by forcing multiple incumbent vendors to continuously compete against each other, with price as the primary differentiator,” Reichman writes. “This strategy can reduce prices and limit vendor lock-in, but it can also lead to management complexity and poor capacity utilization.”
The report recommends keeping things simple by and using fewer vendors when possible. However, that advice comes with several caveats: buying all storage from one vendor means taking the bad with the good, and some vendors’ product families differ so much “they may as well come from different vendors.”
Of course, I’m sure there are horror stories out there from organizations that have had bad experience with lock-in as well as those who’ve had incompatibility issues with products from multiple vendors.
My daily rounds of the storage industry web this morning brought me to The Storage Anarchist, a blog by an EMCer that I often find interesting. As it turns out, one of my articles was in his sights yesterday following EMC’s earnings call.
Most of the reaction to the first-quarter earnings announcement was rather more negative than I think EMC would like, considering they posted record revenues. All the financial analysts on the call, wild eyed from the fog of battle out in the market as the economy sinks further into doldrums, seemed not to believe that EMC’s forecasts for the year were really remaining unchanged. And they did ask plenty of pointed questions.
TSA’s description is rather more dramatic: “Several of the participating financial analysts inquired about the potential impact that the newly-delivered virtual provisioning for Symmetrix might have on future capacity demands. From the tone of the questions, you could easily imagine a pride of lions circling their prey.”
But I have to say the next sentence surprised me. “And sure enough, by noon Beth Pariseau had her coverage posted on SearchStorage, under the headline EMC’s Tucci: Thin provisioning mandatory but overrated.”
After that there’s some discussion of a Byte and Switch article and there’s no further discussion of my article, so I’m still not precisely sure why it was brought up. A little ways down in the post, though, there’s this reference to a bear that recently killed its trainer that I can’t help but wonder about:
And all I have to say about the bear is: remember, these are wild animals, and they’re driven by instinct and not logic or trust.
Any resemblance between wild animals and industry experts is purely coincidental!
Again, it’s hard to tell exactly where that comment was directed, but I think he compared Mary Jander, Wall Street analysts and me to wild animals? That would certainly be a first for me!
So here’s the perspective from the other side of the coin (or cage, as it were). When the CEO of a major storage company explains to the folks on Wall Street exactly how his company is going to continue to make money on a feature billed by many in the industry as a way to not give vendors like EMC quite so much money in the long run, I think it’s probably important for users to hear that perspective on the technology. I think it’s also probably important for users to have a realistic sense of the benefits of a given technology, one they’re not getting from most vendor marketing. That’s the logic and trust I care about.
Meanwhile, TSA saves most of his scantily-veiled critiques for IBM, though of course he never names names. This in turn prompted IBM blogger Barry Whyte to respond with…the news that IBM is planning thin provisioning for SVC. IBM is giving thin provisioning the title, “Space Efficient Volumes/Vdisks (SEV).”
So lets think about this, if for example you had an appliance that could front all storage types, provide you with online data migration between said storage types, let you manage copy services across them all, soon provide Space Efficient characteristics, natively support any SATA or flash device you decided you wanted, provide many thousands of disks behind a single management interface and integrate with all the ‘Israeli’ products you could imagine… why would you care that just one of your products that has its largest footprint as a Mainframe box didn’t have all of those features, when according to Mr Burke, everything the Mainframe does well it does itself, and by his own admission won’t need or use features like Thin Provisioning.
Interesting. But what’s odd there is that the mainframe box IBM sells is the DS8000, and last I heard, IBM’s planning thin provisioning for that too. Or maybe it will be getting thin provisioning by way of SVC?