Tuesday, September 29, 2009

How to manage 2PB+

September 30, 2009 – I dialed in to a wikibon.org Peer Incite teleconference yesterday. I like these things because they typically feature end users, often representing very large IT facilities.

In yesterday’s meeting, that would be the California Institute of Technology (Caltech) which, among many other things, is the academic home of NASA’s Jet Propulsion Laboratory (JPL).

One facility at Caltech hosts 2.3PB to 2.5PB of data, according to Eugean Hacopians, a senior systems engineer at Caltech and the speaker on the wikibon.org Peer Incite gathering. Since the facility’s files are very small, that 2.5PB translates into about two trillion files, according to Hacopians. Translated another way, the facility has about five million files per TB of storage capacity. (In its Infrared Processing and Analysis Center, or IPAC, the applications are primarily astronomy imaging related to space projects.)

You can listen to Hacopians’ hour-long chat here, but a few things jumped out at me while I was listening.

Instead of using a traditional SAN, Hacopians uses what he refers to as building blocks. In the file-serving area (as opposed to its compute servers and database servers), a typical building block consists of a Sun server (with two 4Gbps Fibre Channel HBAs and one 4Gbps Fibre Channel switch from QLogic) attached to SATA-based disk subsystems. About 99% of the capacity is on SATABeast disk arrays from Nexsan, and up to three SATABeast arrays can be attached to each file server.

“A large, shared SAN would have created more hassles and headaches than the building block approach,” says Hacopians. “A SAN would have introduced 3x to 5x more complexity.” That’s in part because the Caltech facility has a lot of different projects (with 10 to 14 projects going on simultaneously), which poses problems from a charge-back and accounting standpoint, according to Hacopians.

To cut energy costs (which are high when you have more than 2,500 spinning disks) Hacopians’ Caltech facility takes advantage of Nexsan’s autoMAID (massive array of idle disks) technology, which offers three levels of disk spin-down modes. (Caltech uses two of the three modes to maximize the performance/savings tradeoffs.)

Although some IT sites are leery of disk spin-down technology, Hacopians says that Caltech has not had any negative issues with Nexsan’s autoMAID technology.

So how many storage pros does it take to manage almost 2.5PB of storage capacity in a building block architecture? Until about two years ago, Hacopians managed about 1.5PB on his own. Today, he has help from two other people, each spending about one-fourth of their time on storage management.

Wednesday, September 23, 2009

What is Dedupe 2.0?

September 28, 2009 – Dedupe 2.0 is a term used primarily by Permabit, and it involves extending the benefits of data deduplication across the entire storage environment – not just the traditional application of deduplication on secondary (backup/archive) devices but on primary storage devices as well.

And, yes, it has applicability to cloud storage, whether it’s internal/private or external/hosted. (Then again, what doesn’t have applicability to cloud storage these days?)

If reducing your overall storage costs by 10x or 20x via across-the-board deduplication and capacity optimization sounds appealing, you might want to view our recent Webcast, “Dedupe 2.0: The Benefits of Optimizing Primary and Secondary Storage.” It originally aired last week, but is accessible here.

Jeff Boles, a senior analyst with the Taneja Group research and consulting firm, addresses the IT challenges and issues and Mike Ivanov, vice president of marketing at Permabit, explores solutions.

Monday, September 21, 2009

Will network and storage teams merge?

September 21, 2009 – Server virtualization brings up an old question: Within IT, should the network and storage groups combine? The question is valid because virtualization has created an unprecedented inter-dependency within IT teams – and not only between storage and networking, but also the server and applications teams.

The question isn’t new. It first arose when NAS hit the scene, and came up again when iSCSI gained momentum. But for the most part, NAS and iSCSI are still the domain of the storage professionals.

The question is moot at most SMBs, because the same person or small team is responsible for virtually all IT disciplines. But at larger enterprises, merging IT groups is a stickier issue.

I don’t think virtualization alone will lead to a merger of network and storage groups, and a recent survey of Fortune 1000 firms conducted by TheInfoPro (TIP) seems to, partially, support that opinion.

According to the TIP survey, although more than half (54%) of the respondents said that server virtualization has had a ‘significant’ or ‘major’ impact on addressing storage needs, 78% of the respondents said they do not expect storage and networking teams to combine.

In addition, most (77%) of the respondents said they do not have a separate virtualization group, and 60% said their organization sees major operational benefit in having a separate data management (storage) group.

Still, that seems to leave a lot of IT pros on the fence regarding the merger of network and storage groups.

But if NAS, iSCSI and virtualization don’t drive the various IT groups together, one emerging technology could be the tipping point: Fibre Channel over Ethernet (FCoE) and so-called 'converged networks.'

FCoE, together with the other storage/networking technologies, could cause interesting internal IT battles. And if it doesn’t lead to an actual merger of IT teams, it will most certainly lead to the need for an unprecedented level of cooperation between them.

For another take on TheInfoPro’s study, see Kevin Komiega’s blog post, “Come together? Not now . . .in IT”

Wednesday, September 16, 2009

Cloud computing and the US Open

September 16, 2009 – I logged plenty of TV hours watching U.S. Open tennis (not to mention NFL season openers) over the last few weeks, all in attempt to gain some respite from the IT world. However, I was subjected to IBM’s cloud computing ad for what seemed like dozens of times during the Open. (Big Blue seems to be sparing football fans from the same torture.)

A guy comes on the screen and asks, “What is cloud computing?,” piquing my interest, albeit to a very minor degree.

So, with a huge international TV audience at its disposal, IBM’s response is: “A cloud is a workload-optimized service management platform enabling new consumption and delivery models.”

Just as I’m about to say “What?” a young girl comes on the screen and says, “It’s what?”

The remainder of the ad doesn’t de-obfuscate cloud computing any further. (In fact, it makes it cloudier.)

To the degree that tennis fans were in the mood for a good definition of cloud computing, they’re just as confused about it now as they were with the Serena Williams foot-fault call and ensuing diatribe.

I have no idea what the TV networks get for 30-second spots during the U.S. Open, but that didn’t seem to be money well spent.

But it could have been worse: I could have been subjected to TV spots about cloud storage. (Yeh, right, like cloud storage providers have that kind of cash sitting around.)

In any case, if you are interested in cloud storage, check out the recently-posted “Cloud storage opportunities and challenges” by Margalla Communications analyst Saqib Jang. Saqib delves into end-user and cloud-provider requirements in the areas of scalability, privacy, data protection, manageability and security.

You can also find some fresh videos, blogs and articles on cloud computing and cloud storage at Glasshouse Technologies’ Web site.

Friday, September 4, 2009

Subscribe to our new newsletter

September 10, 2009 – InfoStor recently expanded the content in our weekly e-newsletters by about 75%. In addition to the top five news stories, we added blog posts from yours truly and senior editor Kevin Komiega, in-depth feature articles from our staff and independent consultants and analysts, links to Topic Centers focused on specific technologies, and links to recommended white papers and Webcasts.

To subscribe to the newsletter, which is delivered weekly to your email inbox, click here.

If you’re not a subscriber you missed out on the following content in the most recent newsletter:

Latest News
Dell taps Brocade for FCoE, CEE gear
Brocade adds VM visibility to SAN management software
FalconStor extends virtual appliance line for VMware
Storage highlights from VMworld

Latet Blogs
Dave's Blog:
VMware and Cisco and EMC, oh my
Rumors swirled at VMworld regarding the imminent announcement of a deep, formal partnership between VMware, Cisco and EMC (aka VCE).

Kevin's Blog:
EMC scoops up FastScale Technology, Kazeon
The EMC acquisition machine has been busy. . .

Featured Articles
SAN management for virtual servers
Sorting out SSD strengths and weaknesses
Why you need virtual infrastructure optimization
Consider data reduction for primary storage
Market prepped for 6Gbps SAS

Featured Topic Center: Virtualization
For in-depth features and breaking news on virtualization from a storage perspective, visit InfoStor's Virtualization Topic Center

White Papers and Webcasts
D2D2T Backup Architectures and the Impact of Data Deduplication
Backup and Recovery: The Benefits of Multiple Deduplication Policies
Optimizing Performance and Maximizing Investment in Tape Storage Systems
Dedupe 2.0: The Benefits of Optimizing Primary and Secondary Storage

Tuesday, September 1, 2009

VMware and Cisco and EMC, oh my

September 1, 2009 – There’s a rumor swirling at this week’s VMworld show regarding the imminent announcement of a deep, formal partnership between VMware, Cisco and EMC (aka VCE).

The troika already has a loose relationship, but scuttlebutt suggests a much more formidable triumvirate. The announcement might come at VMworld this week, although next week may be more likely.

Paul Mansky, a principal in equity research, data center infrastructure, at CANACCORD Adams, issued a note this week that hints at what the three-way relationship might entail.

Citing industry sources, Mansky says the deal will revolve around Cisco’s Unified Computing System (UCS) platform and, of course, EMC storage systems (with an emphasis on Atmos?) and VMware in bundled configurations that address cloud computing (internal or external) environments. The venture is also expected to include joint testing and marketing, as well as a road show in September or October.

Interestingly, sources also say that EMC plans to compensate both EMC and Cisco field reps for 100% of the value of the combined solution sale.

If anybody doubted whether Cisco’s UCS would gain traction, this deal would put those notions to rest.