Wednesday, July 28, 2010

A backup vendor you haven’t heard of

July 28, 2010 – It’s hard to imagine that there’s a backup software vendor out there that, until recently, I hadn’t heard of. And I’m guessing you haven’t heard of them either.

Cofio Software was formed in mid-2006 and has been shipping its flagship AIMstor data protection software for about a year. The company’s early adopters have been mostly in China, where users are apparently more open to trying something new and aren’t as tied to brand name products.

Some of Cofio’s founders (including Tony Cerqueira and Fabrice Helliker) were formerly at BakBone Software, but Cofio can trace its roots back to AT&T Bell Labs and the NetVault product line.

In addition to backup and restore, the AIMstor suite includes a host of other data protection applications and technologies, including data deduplication (target and source, with Change Byte Transfer technology), snapshots, real-time replication, continuous data protection (CDP), live/batch backup, bare metal recovery, and more.

However, in my opinion Cofio has two key differentiators vs. the more well-known backup vendors:

--All of Cofio’s data protection applications are built on a common code platform, or unified framework. In that sense, Cofio’s primary competitor may be CommVault, which also has a common code base for a variety of data protection applications. This is in contrast to the disparate, often non-integrated point products that vendors include in their data protections suites. It’s also in contrast to the need for users to buy various data protection applications from multiple vendors, which leads to management nightmares.

--AIMstor has a (seemingly) unique interface with drag and drop policies and data flow of physical and virtual machines and groups. It’s called a Workflow User Interface (WUI), and it has a whiteboard feel to it. Since I don’t use enterprise-class backup software, it’s hard for me to describe this interface. It’s the sort of thing that you need a demo for (see link below).

Although Cofio has been shipping AIMstor for about a year, the company just released the 2.2 version, which can be considered the first production release.

Cofio is in the process of recruiting VARs, and only has a handful of them in North America. One of those VARs is Cambridge Computer.

I asked Cambridge Computer CTO Jacob Farmer what he likes about AIMstor, and here’s his response:

"I just started working with them. I like the concept. In a sentence, you define policies and it protects your data. All other backup products just copy data on a scheduled basis.

Effectively, you tell the software:
- what data you want to protect
- what you need for recovery point
- what you need for recovery time
- what you need for retention
- where you want copies of the data stored

It also includes other aspects of data protection
-- immutability (enforcement of write-once)
-- user access audit logs
-- bare metal to disparate hardware "

The recently released version of AIMstor includes a lot of new features. For details, see the AIMstor 2.2 press release.

But like I said, this is the sort of product that you really need a demo on to understand it fully. Check out the short video demo on Cofio’s website to get started.

Or click here for a free trial version of the software.

Of course, this begs the question of whether or not the storage industry needs yet another backup vendor, but if you’re looking for something new and aren’t married to the big brand name vendors, Cofio’s AIMstor might be worth a test drive.

Related article from wikibon.org (Aug. 17, 2009):
Cofio Software wants SMBs on board with their data management platform

Related articles from InfoStor:
CommVault connects Simpana software to the cloud
CA debuts ARCserve r15, adds D2D software
Symantec beefs up SMB backup, security
Arkeia integrates backup with VMware vStorage


Related article from wikibon.org (Aug. 17, 2009):
Cofio Software wants SMBs on board with their data management platform

Monday, July 26, 2010

Attention storage pros in the entertainment industry

July 26, 2010 – There are plenty of storage-oriented trade shows and conferences to choose from, but few focus on a specific vertical market. One exception is the upcoming Creative Storage Conference next Tuesday, August 3, in Culver City, CA – conveniently located near the L.A. entertainment hubs of Hollywood, Burbank and Studio City.

The fourth annual, one-day conference is focused solely on the storage issues faced by creative professionals in the media and entertainment industry, and addresses applications such as pre- and post-production, HD content capture and editing, animation, special effects and non-linear editing.

The sessions, presentations and discussions will cover all areas of storage, from primary storage to content archiving.

Sessions include:

Content Capture: Many cameras, many effects, many storage devices
Soothing the bits, posting the passion: Storage for editing and post production
Delivering the goods: Storage for content delivery
Keeping the good stuff to last: Content archiving and asset management
So that’s what you mean – Entertainment and media users talk about how they use digital storage

A sampling of the storage vendors that will be participating includes NetApp, Isilon, Quantum, DataDirect Networks, SanDisk, Western Digital, AIC and Promise Technology.

View the full agenda and register at the Creative Storage Conference web site.

Thursday, July 22, 2010

Isilon stock soars on Q2 earnings report

July 22, 2010 – Who says that startups can’t crack the market-share stranglehold of the big boys? Isilon appears to be on its way to doing just that.

Isilon presented its Q2 earnings report today, highlighting quarterly revenue of $45.1 million, up 15% over Q1 revenue of $39.3 million and up a whopping 56% over the second quarter of 2009.

The apparent turnaround has once again put Isilon (NSDQ: ISLN) in the billion dollar club, with a market value of $1.06 billion.

The news boosted Isilon to the #1 spot in the InfoStor Market Index today.

At one point today the company’s shares were trading at $16.24, up almost 20%. This year, Isilon’s stock price is up about 125%.

It’s hard to believe that it’s been almost four years since Isilon went public, in September 2006, and it’s interesting to note that the company’s stock topped $27 back in those heady days (after an IPO opening at $13 per share).

You can get the full financial results in Isilon’s press release, but here a few key stats:

--Q2 net income was about $2 million, or twice what analysts had predicted and twice the company’s Q1 2010 earnings. That compares to a net loss of $3.7 million in Q2 2009.

--Gross margin was 62%, up from 57% in Q2 2009.

--Isilon execs are estimating fiscal 2010 revenue growth in the low-to-mid 40% range

And the company seems to be in hiring mode. Isilon now has 398 employees, an increase of 30 people vs. the previous quarter.

I’ve always thought that one of Isilon’s weaknesses was an over-reliance on a single vertical market -- media and entertainment, which is particularly well-suited to the company’s scale-out NAS architecture. However, Isilon seems to making inroads into the more general IT market.

In the second quarter, media and entertainment accounted for about 35% of the company’s revenue. However, revenue from mainstream/traditional enterprises accounted for 28% of its revenue, a figure that is steadily growing.

From what I hear, Isilon's success is in no small part due to its strengths in the virtual server space. Here's what one user emailed to me in response to my original post:

"They are killing it in the VMware market. Their VMware story for NFS is head and shoulders better than everyone else, at least from a management perspective.

"Just imagine, no LUNs or volumes. Which means no more spending half my day balancing VMs across LUNs and volumes. Presto, 1/2 of the stuff that blows about managing VMware goes away. I can vMotion storage whenever with no repercussions. I can take snaps and replicate at a file/dir level (i.e. VM level), and it is monkey easy to manage as a system. "

Look out, NetApp.

Related articles:
Isilon puts multiple tiers under one file system
EMC breaks Q2 revenue record

Wednesday, July 21, 2010

Keep an eye on this startup

July 21, 2010 – Some storage startups anticipate a trend then ride the coattails of the trend. Other startups actually create trends.

One startup that may be in the latter category is Actifio, which announced $8 million in Series A financing today led by Greylock Partners and North Bridge Venture Partners, after an 18-month incubation period.

Actifio is pioneering a concept it refers to as Data Management Virtualization, or DMV (a search acronym that will surely lead you astray).

The company’s press release states that server virtualization technologies have transformed the IT infrastructure but that the storage infrastructure is still a major bottleneck “with data lifecycle management shackled by point tools that are deployed in silos – creating complexity, inflexibility and significant expense . . . Actifio’s DMV technology reduces the cost of managing the application data lifecycle and virtualizes vendor-independent physical or cloud-based storage devices into a private, public or hybrid storage cloud infrastructure.”

That didn’t help me much, so I had a chat with one of Actifio’s key founders – Ash Ashutosh, formerly a vice president and chief technologist in HP’s StorageWorks division and, perhaps more importantly in this context, founder of AppIQ, which was acquired by HP a few years ago.

“DMV is a solution for data protection, disaster recovery, business continuity and lifecycle management of data,” Ash explained. “It simplifies data management. The paradigm we’re applying to data management is analogous to what virtualization did to servers.”

I still don’t get it, but I’m calling attention to this startup because of its management pedigree. In addition to Ash, Actifio executives include:

David Chang, vice president of products (formerly founder and VP of product management at AppIQ)
Steven Blumenau, VP of products (formerly a VP at Iron Mountain and senior director of advanced development at EMC)
Rick Nagengast, VP of sales (formerly VP of channel and partner development at EMC and GM of the Storage Products Division at DEC and Compaq)
James Pownell, customer operations manager (formerly founder and president of Exagrid, founder and VP of engineering at Highground Software, and a development manager at EMC)

That’s a lot of HP/AppIQ/EMC blood.

Actifio is expected to ship its product “early in the fourth quarter” – I’m guessing around the time of Storage Networking World. The company plans to sell though the channel, and is currently in hiring mode.

Monday, July 19, 2010

Dell to acquire Ocarina for data deduplication

July 19, 2010 – Dell dropped a bombshell on the data deduplication market today with the announcement that is has signed an agreement to buy Ocarina Networks. The amount of the deal was not disclosed (but I’m working on that). We’ll be following up on this announcement in the near future as we get more info, but for now:

Ocarina has content-aware storage optimization technology – data deduplication plus compression – that until recently it sold bundled in appliances. Less than a month ago, the company announced an embeddable version of its ECOsystem data reduction technology targeted at OEMs.

Since that announcement came amidst rumors that IBM was about to buy Storwize (an Ocarina competitor) for $140 million, the Ocarina announcement led to speculation about which vendors might adopt the technology, as well as which of Ocarina’s partners might acquire the company. (The IBM-Storwize rumor has, at least for now, fizzled out.)

In a blog on Ocarina’s announcement, I opined that HP might be interested in Ocarina (see “Ocarina: 4 dedupe predictions, and an OEM strategy,” a June 22 blog post). Along with EMC, Hitachi Data Systems, BlueArc and others, HP is one of Ocarina’s partners.

That level of perspicuity is why I don’t invest in storage vendors.

Until the announcement of the embeddable, OEM version of its software, Ocarina was known primarily as a vendor of data reduction technology for primary storage. But the embeddable version is applicable across the storage spectrum, from primary storage to backup and archive storage.

In Dell’s press release on the acquisition announcement, Ocarina CEO Murli Thirumale was quoted as saying “This brings deduplication to not only primary storage, but also to key storage workflows including backup, replication, migration and tiering.”

However, the release also stated that “Ocarina brings a leading deduplication capability to complement Dell’s EqualLogic solutions.” And Brad Anderson, senior vice president in Dell’s Enterprise Product Group, also singled out the EqualLogic product line in a quote: “Ocarina provides an important component of our data management portfolio and our EqualLogic ecosystem . . . .”

So, the Ocarina technology will be applied only to the EqualLogic line of iSCSI arrays? Or will it be applied across the board; in other words, all of Dell’s storage systems? And if the latter is correct, what does that mean for Dell’s existing data deduplication technologies and partners?

Will Dell still need deduplication partners CommVault and Symantec?

What are the implications for Dell’s reseller deal with EMC for Data Domain deduplication systems?

Will Dell eventually have one deduplication technology, or will the company maintain a menu of options?

Where does this leave other data reduction specialists, such as GreenBytes, Exar and Permabit, which recently announced an embeddable version of its storage optimization technology (see “Permabit deduplicates primary storage”)?

And how will the deduplication kingpins – NetApp and EMC – react?

For now, more questions than answers.

Related blog posts:
“Ocarina: 4 dedupe predictions, and an OEM strategy” (June 22)
IBM to acquire Storwize for $140 million? (June 14)
And here's a blog post on the Dell-Ocarina acquisition from David West, CommVault's vice president of marketing and business development: http://news.commvault.com/DavidWest/000049_A_Complementary_Approach_to_Deduplication.asp

Friday, July 16, 2010

InfoStor + Enterprise Storage Forum = Storage Synergy

July 16, 2010 – Many of you are aware that InfoStor is now part of the Internet.com network of websites, which includes 12 sites for IT professionals and 4 sites for developers. Also part of that network is Enterprise Storage Forum which, like InfoStor, focuses solely on the data storage market.

As a result, we are now exploiting the synergies between the two sites, primarily by sharing content. Sometimes we’ll cover breaking stories and trends separately and sometimes we’ll link between the sites to repurpose content.

The net effect for visitors to both sites is a significant increase in storage-specific content.

As part of this move, my long-time colleague and compadre Kevin Komiega has assumed primary responsibility for Enterprise Storage Forum, while I will continue to focus primarily on InfoStor. But we will both be producing content for the two sites.

In addition to exploiting synergies between Infostor and Enterprise Storage Forum, we will continue to share content with other Internet.com IT sites to bring you a wider variety of content, albeit always with the storage professional in mind.

Here’s a recent example from Datamation:
Storage tiering with automated data migration

Feel free to contact us:
Dave Simpson: dsimpson@internet.com
Kevin Komiega: kkomiega@internet.com

Related blog:
Infostor.com is heading to Internet.com

Tuesday, July 13, 2010

It may be a slow boat to converged networks

July 14, 2010 – The hype behind converged networks (or converged fabrics) is heating up, leading one to believe that we’re on the cusp of this game-changer in IT infrastructure. However, I think this is going to be a very slow transition that will take place in two distinct phases.

First, IT organizations will migrate to 10Gigabit Ethernet (10GbE), which is already well underway. As they do so, the underlying equipment (network adapters, switches) may in fact support a converged fabric (Ethernet LAN + Fibre Channel SAN via FCoE) but most users won’t take advantage of that functionality in the first phase of the migration.

For now, end users are primarily concerned about network performance, particularly in I/O-intensive environments such as virtual servers and, soon, virtual desktops (see “Storage considerations for VDI implementations,” by the Evaluator Group’s Russ Fellows and John Webster).

Faster Ethernet is priority #1, not truly converged fabrics. Remember, almost 90% of the server network ports in a data center are Ethernet.

This a central point in an InfoStor blog post from Frank Berry, CEO and senior analyst with IT Brand Pulse (see “3G C-NICs Address Mass Migration to 10GbE”).

In an end-user survey conducted by IT Brand Pulse, more than 70% of the users said they were either too busy to investigate convergence or were not planning to converge their networks (although that does leave 30% with at least a plan to do so).

According to Frank’s blog: “LAN and SAN convergence is happening today, but it’s deployment of LANs and iSCSI SANs that are making it happen. In 2010, less than 100,000 FCoE-enabled host network ports will be deployed, while over 1 million iSCSI host ports will be installed.”

Migration to 10GbE will go quickly, but the transition to a truly converged, single-wire network will go much more slowly.

One reason for this is not technology-related but, rather, cultural. More than ever, a converged network will require very close cooperation between the network and storage teams, which in some cases is like asking the Red Sox to merge with the Yankees.

Who’s in charge of product evaluation and purchasing? Who will “own” the converged network?

These factors, perhaps more than technology-related issues, could delay adoption of converged networks.

Related article:
Fabric convergence: Changing the nature of fabric attach

Tuesday, July 6, 2010

EMC to buy Greenplum for “big data” private clouds

July 6, 2010 – EMC announced today that it plans to acquire Greenplum for an undisclosed amount of cash. The deal is expected to close by September.

For the news story, see “EMC acquires data warehousing vendor Greenplum.”

I had never heard of Greenplum, but after a quick look at their site I gleaned that the company is a data warehouse and analytics vendor. However, EMC wasn’t about to lose an opportunity to fly the cloud flag in its press release on the acquisition announcement, according to which: Greenplum’s “disruptive data warehousing technology [is] a key enabler of “big data” clouds and self-service analytics.” That’s almost poetic.

EMC branching out beyond its storage-centric roots is, of course, nothing new. And the Greenplum acquisition seems to make more sense than the Documentum and RSA acquisitions originally did, although it may not be quite as impressive as the VMware and Data Domain acquisitions are proving to be.

Greenplum has a shared-nothing, massively parallel processing (MPP) architecture for data warehousing and analytical processing, delivered on a virtualized x86 infrastructure (unlike, oh, Oracle’s Exadata).

EMC plans to meld Greenplum’s architecture with its own private cloud vision/architecture. After the acquisition, Greenplum will become part of a new Data Computing Product Division at EMC.

On the surface, this just looks like a solid acquisition at what I assume to be a good price. But to get an idea of what EMC is really up to here, read Chuck Hollis’ blog: “EMC to Acquire Greenplum.”

Noting that the Greenplum architecture is based on a virtualized x86 infrastructure, Chuck notes that: “All of EMC’s storage products are x86 based – this creates a potential pathway where data intensive functions could be run closer to the information, freeing the compute farm to do what it does best.”

“The vast majority of these data warehouses contain sensitive information and produce analysis that is either confidential or otherwise privileged. Think information security and data loss prevention, for example.”

“Much of the higher-order analysis produces rich content that frequently drives a collaborative workflow among knowledge workers. Think about EMC’s assets in content management, collaborative workflows and case management.”

“And . . . let’s not forget the seductive appeal of running on-demand business analytics as yet another fully virtualized workload using dynamic resources in a private cloud model. Like running on a good-sized Vblock, for example.”

Is this a competitive reaction to Oracle? No, according to Hollis. And it won’t disrupt EMC’s relationships with vendors such as Sybase, SAP, Microsoft and ParAccell (Greenplum competitors, as are vendors such as Netezza and Teradata).

It’s impossible to opine about the sagacity of an acquisition if you don’t know how much a company paid, but I’m guessing that EMC (NYSE:EMC) got a good deal here. And it gives them a great opportunity to demonstrate unique use cases for their private cloud vision, not to mention the opportunity to sell a lot of hardware because the data warehouses that Greenplum plays in are measured in petabytes.

Greenplum customers include NASDAQ OMX, NYSE Euronext, Skype, Equifax and T-Mobile.

Friday, July 2, 2010

EMC pulls the plug on Atmos Online

July 2, 2010 – EMC has shuttered its Atmos Online cloud storage service, at least as a commercial service for end users. The company is shifting the technology to its Atmos-based cloud storage service provider partners, which currently include AT&T, Hosted Solutions and Peer 1 Hosting.

And I’m sure there will be many more now that partners won’t have to compete directly with EMC.

On its Atmos Online site, EMC directed customers to its Atmos cloud service provider partners.

Going forward, Atmos Online will be available strictly as a development environment, rather than a paid subscription service.

In my POV, Atmos Online was always a POC anyway.

It was a proof-of-concept designed to lure services providers at a time when the cloud storage concept was still a bit sketchy (although some would argue that it’s still sketchy). Atmos Online was launched a little more than a year ago.

The losers in this announcement are existing customers of Atmos Online (although it’s unclear how many of those there actually were), who will have to shift to one of EMC’s partner’s services or migrate to a non-Atmos cloud storage service.

The winners, of course, are EMC’s Atmos cloud storage partners, which no longer have to compete with EMC. Likewise, EMC no longer has to compete with those partners. It’s a win-win on that front. In fact, on its Atmos web site, EMC is “strongly encouraging” its existing Atmos customers to migrate to one of its partners.

Atmos Online as a commercial service for end users apparently never gained enough ground for EMC to justify the costs associated with hosted services. There’s more money in selling the technology and equipment to third-party providers that, at least in the case of AT&T, are better equipped to handle -- and make money from -- public cloud storage services. (One of EMC’s other partners – Hosted Solutions – launched its Atmos-based Stratus Cloud Storage service last month, which is based on TwinStrata’s CloudArray technology.)

Besides, Atmos Online wasn’t EMC’s only cloud storage play for consumers; the company still has Mozy and online storage services provider Iomega.

The shuttering of Atmos Online should be seen not as a failure on the part of EMC but, rather, as another prescient and tactical move by the company.