This offer really highlights the potential cloud storage has to help companies or local governments get fast and easy access to disaster recovery capabilities they could never afford in the past.  No hardware to buy, just download the software, map the drives and start a copy process.  Robocopy is free and you already have it, so no investment there.  And I’ve used the Nasuni software – I am certainly no system administrator but I was able to figure out how to add files and create snapshot copies, as well as restore from snaps, pretty quickly.  And because the environment is virtual – getting up and running in the event of an actual disaster can be done by spinning up a virtual machine from anywhere, provided you have the proper credentials.

This offer gave me one of those head shaking moments where I sit back and think about just how far technology has advanced in these recent few years – think about what it would have taken just 3 or 4 years ago to create a DR site and actually get back up and running in the event of a disaster - the remote site, hardware (servers and storage), advanced storage arrays capable of creating remote mirrors or host-based replication software, the software licenses and management overhead.  The cost and complexity of creating that type of environment is far beyond what many small to mid-sized companies could afford, never mind what state and local government budgets can pay for.  The cloud changes everything, this can all be bought as a service, in a shared model to lower administrative and capital costs, with near perfect economics (100% utilization).

Other vendors can learn from this example.  Aside from the obvious user benefits, this offer is also a good example of innovative marketing, presenting an offer that can really make a business impact to help users with true business issues – disaster preparedness and, if needed, recovery.  Steve blogged a while back about how technology marketeers are promoting cloud everything without tying solutions to business needs and items that users actually create a budget line item for.  This Nasuni offer is a great example of someone hitting the mark.

I learned much more about Solaris and the SPARC processor roadmap than any human should be subjected to, but it was still worth the trip.  The Sun and STK guys seem really happy to “not be Sun” anymore and be a part of one of the largest technology companies in the world.  They have funding, are hiring, and all the questions about viability are gone.  They have over 300,000 Oracle customers now that are all Sun storage prospects, and a salesforce larger than they could ever have dreamed at Sun.  The storage guys, both disk and tape, seemed like they had a weight lifted off their shoulders.  They were open and talkative about the storage portfolio and where it’s headed, openly sharing product development roadmaps (under NDA, of course) with the analyst community.

These guys have a really good story to tell.  They have the apps, database, server, switches (data center fabric only, Infiniband, not going after Cisco any time soon) and storage.  They have a real converged computing story.  Exadata is pretty impressive, and the management team we saw over the past 2 days was very impressive.

I guess, if I had to net it all out, I’d sum it up as saying these two days certainly took away any doubt I had about Oracle’s commitment to the Sun storage portfolio.

Cloud storage holds tremendous promise as a way for IT to significantly drive down both CAPEX and OPEX, but the reality is that cloud storage is new and unproven.  Most cloud storage service provider solutions on the market today offer basic storage capacity and data protection, such as RAID or mirroring, but we continue to hear user concerns about data availability, security, and performance as well as regulatory compliance.  That’s exactly what TwinStrata addresses–and why we’ll continue to see cloud storage enablement platforms come to market this year.  This is an emerging market category that encompasses companies that address the inhibitors surrounding enterprise IT adoption of cloud storage: data availability is addressed with read-only snapshots; security with encryption; and performance and latency with a local cache and algorithms to ensure active data is stored locally.  To address proprietary interfaces, cloud storage enablement vendors write to the service provider’s proprietary API and present a standards-based interface back to the business while excessive bandwidth costs are addressed through data reduction technology.

TwinStrata has not yet announced the formal product set–but the vision, strategy and intelligent storage cloud ecosystem.  There will be more to come.  The team, Nicos Vekiarides (CEO), John Bates (CTO), and Rob Infantino (Consultant and Advisor) comes from a storage virtualization background (Incipient & StorageApps/HP).  That makes sense–essentially, cloud storage enablement blurs the lines between local and cloud-based storage, creating a virtual (and bottomless) storage pool wrapped with policies for data protection, performance, and placement.  To make an end-to-end intelligent storage cloud a reality requires a strong ecosystem, and it appears TwinStrata has a good head start there with a list of partners that includes cloud storage technology, service, and solution providers.

This is a timely launch.  Cloud storage enablement solutions like those from TwinStrata, Nasuni (just went GA today), and Cirtas (as yet not officially launched) will help dispel user concerns and propel cloud storage into commercial IT.

One of the key challenges associated with being successful in a specific product segment, as NetApp is in NAS, is that you get pigeon-holed into that category and it’s nearly impossible to break out.  EMC would probably still be considered a mainframe storage company if it hadn’t bought DG and invested in growing CLARiiON business – despite its success in SAN and NAS.  But it seems NetApp is breaking the mold somewhat – not by trying to become a SAN storage provider, though it seems to be doing well in the SAN space, but by becoming a storage provider regardless of the access protocol.

In a brief published earlier this year, Unstructured Data in 2010: Trends to Watch,  we predicted that unified storage would gain momentum in 2010, driven by better economics and flexibility.  But what it really comes down to is that it’s not about block or file storage, it’s about supporting applications and the business.  NetApp has its priorities right, and appears to have changed its spots from being a NAS leader to being a storage leader.  Nicely done.

You can watch the webinar right here live on Thursday, March 25 at 2:00 PM, or come back later and watch the archived recording.

ESG research conducted in late 2008/early 2009 showed significant user interest in scale-out NAS solutions thanks to their scalability, business agility, and operational efficiencies. With 2009 spending slowing to a near stop, interest in scale-out mostly stayed just that: interest.  In 2010, ESG expects that interest to translate into actual spending, aided by increased visibility from big-name vendors like EMC, HDS, HP, IBM, and NetApp as they continue to invest in scale-out offerings and validate commodity-based scale-out architectures for enterprise applications.

The interest we saw in late 2008 was pretty high: 38% of those surveyed planned to adopt within 12 months, 37% are interested in the technology!  We wrote a research brief on the topic (this one is password protected, clients only), Scale-Out NAS Adoption & Market Drivers –there is more data in the brief including cuts by industry, key features, adoption drivers, and impact to scale-up architectures.  That data came from our enterprise storage survey, conducted in late 2008.  This new project gives us a chance to dig deeper into the drivers for scale-out storage architectures (both block and file). I am really looking forward to putting more metrics around a number of questions, such as:

• Awareness of, usage of, interest in and plans for new scale-out architectures
• Prioritization of key business/IT drivers (e.g., cost reduction, application drivers)
• Key applications and usage
• Benefits realized/quantified by early adopters
• Plans for integrating scale-out in private cloud or service oriented architectures
• Impact on data center infrastructure
• Primary customer concerns
• Vendor preferences
• Vertical industry dynamics
• Where solutions are deployed (LOB/HQ)
• Impact on scale-up storage deployment
• Most important criteria for choosing scale-out technology & vendor

If you are interested in the research, feel free to contact us.  We’ll be publishing some reports with the new data in the second half of 2010–after the sponsor exclusivity period expires (pretty standard stuff for multi-client reports)–and I’ll share some of the high points here in my blog.

Good blogs on the topic: Steve O’Donnell recently blogged on Internet-scale log files as a driver and Storagebod blogged on industry developments.

It is nearly impossible to unseat an incumbent without a triggering event – this is something you hear from Steve Duplessie (@stevedupe) fairly often.  Certainly the death of the sitting senator qualifies as a trigger for change – in this case it forces change that may not have happened for many years to come.  Humans are by nature change resistant, and that’s reflected pretty well in IT environments.  Once we have technology in place, we get comfortable with it, it’s predictable, and often the devil you know is better than the devil you don’t.  The longer it’s there and the more processes built around it, the higher switching costs become.  But then a compelling event comes around that triggers change.  Today in IT, the economic crisis has largely acted as a trigger event – budgetary pressures combined with technology advances like server virtualization and scale-out storage platforms have IT considering cloud infrastructure alternatives (both public and private).  It will be a long journey, but thanks to the economy and advent of enabling technologies the drive for greater efficiency in the data center has certainly begun.

If you’re coming from a position of weakness or in a vulnerable position, change the discussion.  It was a brilliant move on Brown’s part.  Scott Brown recognized that MA has a pretty strong Democratic party, so he changed the game from Democrat versus Republican to Democrat versus Independent.  Brown understands that despite the fact that registered Democrats outnumber registered Republicans in MA by about 3:1, 51% of the registered voters in MA register as Independents.  And while most of the country thinks of MA as a solid blue state, for the 16 years prior to 2007, a Republican held the top state office.    We are proud independents, and Brown played to that.  We can see tech parallels – about a dozen years ago EMC Symmetrix was in a vulnerable position at the top of the storage market – HDS announced its new high-end systems and Symmetrix was getting long in the tooth.  Over a number of years EMC bought Data General (and a midrange storage business), Legato, Documentum, and changed the discussion to ILM and to building out an information infrastructure.  HDS is still perceived as a high-end storage company.

Leverage all of your available communication channels – including social media. Brown’s campaign used all available communication channels including social media, an area where Coakley’s efforts were pretty much nonexistent.  Momentum builds on itself–a small social medial push can snowball because it leverages the voices of everyone you touch, and fans out from there.  It’s like the old Pantene shampoo commercial that said “I told 2 friends, then they told 2 friends, then so on and so on.”  That was in the world of 1:1 communication.  In the social media world, we all have 1:many channels that give us exponential reach.  While social media did not win the election for Scott Brown, it certainly provided another communication channel that helped him rally his base – I saw a number of friends who are registered Independent start out as Coakley fans but swayed towards Brown in the final weeks by what they were reading (and pointed to) on Facebook!

The list could go on, from ensuring your message resounds with your target audience to complacency, arrogance and passion.  There was certainly a sense of EMC circa 2000 arrogance and complacency in the Coakley campaign, and NetApp circa 2000 passion and drive in the Brown campaign (fortunately for EMC, people are change resistant and that gave them time to respond to vulnerabilities by changing the discussion, and for NetApp, the growth in unstructured data provided a compelling need for big-iron file storage!).

I’ll spare you from more.  And yes, I am sure Brown won for more reasons than I am outlining here – but if there hadn’t been a compelling event, there would not have been a race.  If he hadn’t used all of his available communications channels, would he have effectively gotten his message out?  If he hadn’t changed the discussion, could the sentimentality that so many MA citizens felt about uncle Teddy’s senate seat have played a bigger role?  If Coakley spent more time on communications 2.0 rather than 1.0, would she have mobilized her base?  Post-mortems and what ifs are much easier than setting strategy – but they always provide an opportunity to learn something – we should all make sure to listen.

It’s great to see Microsoft take the chains off of its SLA documentation, but I do have one more request (and I can talk about it now that the NDA is no longer in place).  The SLA is pretty straightforward – Microsoft offers service credits if its uptime falls below 99.9%.  But it is difficult to determine the exclusions, so I ask Microsoft – please provide a pointer within this document to the “technical documentation” to which the SLA refers!  I have searched the site and can’t find it anywhere.  Here’s what I am referring to, directly from the now public SLA, I highlighted the reference with bold font:

“Total Storage Transactions” are all the storage transactions in a given time interval (initially set at one hour) for a subscription, with a few notable exceptions.  Examples of excluded transactions include pre-authentication failures, transactions that are throttled based on suspicion of abusive behavior, authentication failures, attempted transactions for accounts over their prescribed quotas, creation or deletion of containers, tables or queues, or clearing of queues.  These exceptions do not count toward either Total Storage Transactions or Failed Storage Transactions.  Please refer to our technical documentation for further information regarding Total Storage Transactions.

There are other spots where Microsoft refers to the technical documentation for further information, including where it refers to defining failed storage transactions as “Transactions not processed within the time period specified in our technical documentation.”

Microsoft – thank you, you took a big giant step forward, but there is a little more work to do to finish up, the SLA is good, but not complete!

Over the course of business today I was stunned to come across this:

Digging into Microsoft Azure cloud storage services, I went looking for the SLA.  Microsoft posts the SLA as a publicly available Word doc for download, but posts the following on the download page:

“The Service Level Agreements (“SLAs”) posted on this site are the confidential information of Microsoft. You may view the SLAs if you have purchased or are considering the purchase of Online Services from Microsoft. You agree not to disclose the SLAs to any other third party or to make use of the information for purposes not related to the your (SIC) purchase or prospective purchase of Online Services from Microsoft. By downloading or viewing this document, you agree to these terms.”

I am always considering buying online storage services, so of course I downloaded the agreement and read it.  But I can’t talk about what it said.  In this era of social media and transparency, I am (almost) speechless that Microsoft would put this confidentiality language around its services.

Service providers are asking us to trust them with our data and consume storage as a service; how can vendors go about hiding details behind service levels?  If the electric company started out by saying “I’ll tell you what type of service you’ll get from us, as long as you don’t tell anyone else”  we’d probably still be using candles… it just makes me wonder what they are trying to hide!

The cloud storage SLA agreement (and associated confidentiality language) can be found at:  http://www.microsoft.com/downloads/details.aspx?displaylang=en&FamilyID=d32702dd-a85c-464d-b54d-422a23939871

Is this even enforceable?  Does it mean we can’t discuss the information I found publicly available on the web?  Am I the only one that thinks this is whacked?

Microsoft, are you listening?

The top three issues facing CIOs today are avoiding or reducing risk to existing business activities, reducing cycle time to deliver innovation, and reducing cost.  Vendors often focus on only one vector when positioning and selling their products, and this is often a weak spot in vendor strategy.  It is important to keep all three vectors in mind – because missing on one  can cost a vendor a sale and  missing on two creates a big hole the competition can exploit.  Think about it – cost savings is really attractive in this macroeconomic environment – but how many of you as vendors walk in with incredible TCO and ROI models and think your cost proposition is a sure winner – only to lose the deal?  CIOs won’t put money-saving efforts in place if there is a chance of introducing risk into the equation and even the most compelling cost-savings solutions are undone if they increase cycle times.  Understanding how you can help the business by reducing cycle time, without introducing risk, is a compelling value proposition.

On the storage front where I focus, managing data growth is an ongoing challenge for IT. It is also the “low hanging fruit” with which CIOs can make an impact on all three vectors.  Keeping up with data growth has become an ever more costly effort as it has been historically limited by traditionally inefficient and complex processes to manage scale-up architectures.  These stove-pipes make it difficult to respond to changing business conditions and create risk by the sheer complexity created when trying to manage an environment with petabyes of storage.

IT needs to change the storage model – it is starting to collapse under the complexity and complexity’s byproduct, waste.   IT’s ability to respond to business needs must occur in real time, which in turn is driving IT to look at deploying newer technologies (like cloud services and scale-out architectures) that can provide a platform for business agility, consolidation, ease-of use, and availability.  In the next few blogs I’ll be looking at the “golden triangle” (okay – we’re working on naming it and making it prettier – but I didn’t want to wait for the branding to be finished to discuss this!) in relation to scale-out architectures and the interest in cloud services.

UPDATE: Steve O’Donnell expands on his thoughts about the golden triangle in his latest blog – give it a read:  http://www.enterprisestrategygroup.com/2010/01/what-influences-it-buying-decisions/