Storage Effect

Even best-in-class rebuild times expose data to hours of risk

Blocks and Files points to an Demartek study (sponsored by Pillar) showing that the Pillar Axiom 500 rebuild times are much shorter on high capacity arrays that similar EMC or NetApp systems.

The glaring data beyond Pillar’s performance, though, is the teeth-clenchingly long times that data is one drive failure away from catastrophic loss in every case. 

The tests were conducted with about 50 500GB drives per system using RAID 5 (RAID 4 for NetApp), meaning the arrays can be rebuilt if one drive fails, but not two.  So during the rebuilds of from 3 to 23 hours, if another drive fails, all data is lost. 

Insert 1 TB drives into the equation, and your rebuild time (and vulnerability) doubles.

RAID 6 and other dual-failure protective schemes make this problem go away, but cost a little in capacity. 

How are you dealing with this?  I’ve heard that RAID 6 is gaining traction for 7200 rpm high capacity enterprise drives like Seagate’s Barracuda ES that are less reliable than 15K SAS enterprise drives (see Seagate’s 3.5″ Cheetah and 2.5″ Savvio for reference). 

Does your RAID vary by drive class?  What other magic do you apply to make this work?

You generate much more information than the files and messages you create

EMC now has a digital footprint calculator on their website.  It estimates how much information is being created, stored and replicated by one’s daily life.  This is an eye-opening exercise that points out all the myriad ways we each generate digital information, far beyond the obvious powerpoints, emails and digital movies.

Similarities to Carbon Footprint 

It has similarities with the Carbon Footprint concept, and of course is directly related due to the power needed to store the kept info I caused to exist in the world.  Another similarity: much of the information created on my behalf was not at my bidding.

I highly recommend the related IDC-EMC forecast of worldwide information growth through 2011.

According to EMC, I’m generating about 8 GB a day - over half a TB so far in 2008.  Beth Pariseau’s and Chuck Hollis’s footprints are here.

What’s your digital footprint?  Comment back and we’ll compare notes.

Deduplication prevents businesses from repeating themselves

Clever marketing from Overland Storage: contrasting their de-duplication solution with a copy machine.  For me it brings home the essence of what de-dupe is all about.

As I posted yesterday, a major headache and expense for business data protection today is redundancy, meaning copying the same files over and over and over again.  Deduplication is one of those technologies whose value is pretty easy to explain:  it makes only one copy of everything, reducing the capacity required by 10 to 20X.

Management is limited, data is not 

Rather than reduce capacity requirements, that frees up more data to be created, used, saved, distributed.  Storage demand is not limited by the amount of data created, but by the ability of consumers and businesses to effectively manage it.

Case in point: according to EMC and IDC, 2007 was the first year that the data generated and replicated in the world exceeded the storage available to keep it. 

Less data leads to more storage 

I’ll say that again: Deduplication leads to more storage.  Agree or disagree?  Tell me why.

Two petabyte storage consolidation and virtualization up and running

Byte and Switch covered Ford’s revamp of their corporate data storage strategy, including 2 petabytes of virtualized storage.  They used IBM’s SAN Volume Controller and DS4000 & DS8000 systems. 

Ford’s Vijay Santoran says that they’ve tested EMC gear as well, and are not relying on a single solution.

Consolidation is paying off already, but it’s not a silver bullet 

From the sounds of it, Ford has been able to deliver on the promises of virtualized storage, including fewer chasses, less power, delayed storage growth.  Despite this, I believe Ford will ultimately be able to support more, not less data with this enabling architecture.

I’m a firm believer that storage as a business enabler trumps storage as a cost savings tool every time.

What do you think? I’d love to hear from you.

Will Cloud Computing take EMC from storage maker to storage user?

EMC is mostly known as a company that makes storage.  That may change in the future, according to Data Center Knowledge.  EMC’s Cloud Computing offering could someday rank them with Google, MSN and others as one of the top users of storage. 

EMC’s recent acquisition of Pi, a personal information management startup is a step in that direction, according to Chuck Hollis, EMC’s VP of product alliances:

“Like a diamond being set into a ring by an expert jeweler, Pi is potentially the centerpiece of a very intriguing strategic play. And one that not too many people will initially appreciate, I believe. I think people are having a tough time putting EMC into a neat industry bucket.”

EMC clearly has a unique vertical business model opportunity as a storage provider and consumer.   

But no one’s yet sure exactly what

Byte and Switch talks about the storage impact of social networking on pure play Web 2.0 social networking sites as well as businesses in general. 

The infrastructure seems to be emulating the network it supports, demanding what they call “cloud computing”.  The needs are less about performance or even capacity, but managing the content cloud as it grows and changes.

EMC and IBM are on it.  So are a slew of newer storage technology companies. 

One thing’s for sure about storage for social networking: there’s going to be a lot of it.

A perfect storm for IBM in the enterprise storage space? 

There are a lot of startups in the storage world with mostly excellent technology, all struggling to break out.  IBM may have scored a coup with its acquisition of XIV by providing the business scale and market reach to propel Nextra into mainstream adoption. 

IBM’s marketing plus XIV’s technology, coupled with the fact that IBM doesn’t have EMC’s Symmetrix business to cannibalize, creates a market-changing Pivotal Moment in the enterprise storage space.  If IBM can move fast, they can achieve the market success in this segment that all of those startups mostly just gret to dream about.

Read what Robin Harris has to say on this.

Equallogic understands the power of manageability

Steve Duplessie’s post on Dell’s Equallogic grab includes a great example of the insatiable demand for storage that’s really only limited by manageability.

Business data storage is becoming less and less about the bytes and all about how easy it is to use them.  Equallogic has had to fight its way kicking and screaming into a fiercely competitive market, and they’ve succeeded because they understood this fact and stayed true to it with their products.