Overview

Few, if any, storage technologies have attained the hype levels cloud storage reached in 2009.  The poor economic environment contributed significantly as cloud vendors touted the capital and operational efficiencies of outsourcing assorted platform, infrastructure, and software responsibilities to cloud service providers.  But a number of high profile system outages cast a cloud (pun intended) over these offerings as users recognized that some cloud storage service offerings may not be 24x7x365 data center ready, nor do they meet enterprise data availability and security requirements.  Enterprise storage users cite cloud storage availability, security, performance, and compliance as key concerns.  Additionally, ESG’s 2010 IT spending intentions data[1] shows moving applications and storage to cloud service providers is at the bottom of the priority list.  ESG does, however, expect some developments that will address these concerns in 2010, and we’ve already seen the emergence of a few: Nasuni and TwinStrata have rolled out hybrid solutions that mitigate most, if not all, of these concerns; all this is promising to make 2010 an interesting year for cloud storage.

Cloud Storage Trends to Watch in 2010

1. The cloud buzz dies down as enterprise IT users step back to evaluate new and existing cloud storage solutions, trying to sort out what is “marketecture” and what is real.  This is the easiest trend to spot, since it represents a normal market cycle most new technologies go through as they reach heights of buzz that no vendor has a chance of delivering upon.  In 2009, everything cloud hit the peak of the cycle. This year, we all step back and take a breath to sort through what is doable today and what is coming in the next few years—it’s time to get our collective heads around the major transformation from providing tactical solutions to providing a virtual IT infrastructure that grants greater business agility and paves the way for new business opportunities.
2. Lack of standards causes both concern and opportunity. The proprietary nature of vendor solutions is the wet blanket at the cloud party.  Major cloud storage service vendors are building their own proprietary APIs; there are no standards, so once users choose a vendor, they are locked in.  There will be some cloud storage success stories, but broad-based adoption of cloud storage services will be stifled by vendor bad behavior and land grabs in addition to a general lack of both standards and interoperability.  It won’t be all bad news, though: cloud storage infrastructure providers will continue to see growth driven by web-based businesses, MSPs, and small businesses.  Even better news is that companies like Cirtas, TwinStrata, and Nasuni are taking care of “translation” from standard data access protocols—like iSCSI, NFS and CIFS—to enable users to jump into (and between) cloud service providers without changing applications or writing scripts.  These solutions break cloud storage service vendor lock in scenarios and provide data portability in the cloud.
3. Task-specific cloud-based storage services will continue to show growth. Archiving thrives and consumer cloud-backup services grow, but commercial enterprises learned the hard way that enterprise backup is not a good cloud use case—or at least not enough of an improvement over the way backup is done today.  Backup is great for consumers and small businesses where data sets are smaller or if there is an RTO of several days.  For commercial enterprises, whether backing up with tape or to the cloud, the solution involves FedEx to get data restored.  Sending large data sets across the wire is just not a viable option.
4. The global regulatory environment gets more play as the implications of leveraging storage service providers across borders creates uncertainty about jurisdiction.  We’ve seen some of this play out in 2009, but the issue is sure to raise its head in 2010 as enterprises look to outsource cloud storage services.  This provides vendors like Symantec and EMC’s RSA division with an opportunity to step up to provide policy-based information management solutions integrating information security technology that cracks open and examines data packets to ensure data doesn’t go where it’s not supposed to. 
5. More pressure is on cloud storage service providers to provide data availability solutions that complement their online offerings.  Cloud storage service vendors begin to realize the jump from offering consumer-class to enterprise-class services is bigger than they thought.  Considering that none of the pure storage service providers offer data availability guarantees, it’s amazing how little attention this issue received in 2009.  Unfortunately, it will probably take a major data loss event, like someone accidentally deleting a large data set, to bring this to light.  Users need some sort of data protection beyond RAID and mirroring, and even mirroring remotely does not solve the data protection problem.  Dashboards and policies that offer snap capability and read-only copies of data sets would be a good start for commercial grade solutions.
6. Unstructured data builds public cloud traction. Local gateways that create an onramp to a cloud storage “tier” will be increasingly used to augment onsite capacity for non-critical data.  Users remain concerned with regulatory compliance and security for much of their data, but there is a great deal of data that doesn’t require lockdown and audit, especially in industries that are not tightly regulated, making this information well-suited for the cloud.  Education is a great example: a university could easily leverage cloud storage services for student home directories.  Imagine the potential saving for a campus supporting tens of thousands of students.  Long-term archive of non-critical information is also a good use for cloud storage and should drive businesses to cloud archive providers like Iron Mountain Digital that offer policy-based archive protection and management.
7. Block cloud storage services start to emerge as a viable onsite storage augmentation solution that brings enterprise-class capabilities to consumer-class cloud solutions.  Gateways that provide local cache to mitigate latency and take care of translating block data IO into proprietary cloud storage vendor APIs makes cloud a viable option to store block data.  Look for these solutions to offer SAN-like local storage targets and advanced functionality like read-only snapshot and versioning.  Solutions like these can completely change data protection as users start to realize that previous (and painful) data protection practices, like backup, are no longer even necessary if the primary data store is in the cloud.  Finding the data that requires a restore from a read-only snapshot is not as straightforward as finding data in the backup catalog, but the potential costs savings of eliminating tape drives, backup servers and software licenses, and losing the operational overhead associated with backup is huge.
8. We’ll be saying goodbye to some cloud storage service vendors unless the business model and offerings evolve to meet enterprise needs.  The SMBs and web-based businesses making up the bulk of cloud adoption today won’t generate enough revenue to pay the bills—and unless cloud vendors address data availability and regulatory concerns, they may not see enough enterprise adoption to pay them either.  The business models, for the most part, don’t support customizing offerings for end-users.  Look for some providers to evolve their business models to offer more in the way of private cloud and storage management services, and to partner with some of the emerging vendors that offer local gateways to enable SAN and NAS tiering to the cloud service as well as data protection services.
9. User organizational issues become both a driver and an inhibitor. Line of business managers become enamored with the easy availability and speed with which new capacity can be deployed and will push for outsourcing.  But IT is in charge of protecting corporate data and will jump in to try to get ahead of those efforts.  The legal department will align with IT.  But IT could be facing a change-or-die mandate if service providers can cost effectively offer advanced data protection and compliance services.  This may be bad news as corporate politics heat up, but smart IT managers will get ahead of the curve and proactively bring proposals for building a private cloud and augmenting onsite with offsite capabilities—and come armed with the associated long term ROI calculations.  IT has a window of opportunity (while service providers are transitioning from commercial-grade to consumer-grade offerings) to start making new data center capacity buys with the goal of building flexible storage services capabilities within the four walls of IT.
10. Users will begin to categorize applications that are cloud-suitable. Realizing that data transfer costs add significantly to cloud storage costs, users will begin to realize that long term storage of reference data that is relatively inactive but needs to be at their fingertips is a good use case. Every cloud provider bills for each and every IO operation, as well as for total amount of data transferred, to cover bandwidth costs.  That needs to be a big consideration for what type of data is stored in the cloud and could inhibit adoption.  Long-term archive for static data is good; anything else will run up the bills.  Leveraging cloud storage for archive alone could really add up to significant savings.

The Bigger Truth

Storage vendors and cloud storage service providers have not done users any favors with all the cloud hype.  Ask any two of them to define what cloud storage is and you’ll get two different answers. Storage vendors are cloud-washing everything in their portfolios to take advantage of the hype wave.

On the end-user front, buying cloud storage services continues to be at the bottom of the list of storage priorities for IT.  Those end-users we’ve spoken to are just too spooked by concerns about meeting auditing requirements to prove regulatory compliance, security, and data availability.  Will there be some enterprise public cloud success stories in 2010?  Absolutely, but it won’t be the norm unless cloud service providers address compliance and data availability.  ESG believes we’ll begin to see these dynamics change throughout 2010.  This year will be one of normalization for cloud storage as the buzz dies down and users realize that cloud storage is a delivery model, not a product; that private clouds are services oriented architectures deployed in their own data centers; and some of the functionality and fluidity of data storage and services that the vendors have been touting is a vision that’s still a few years out.  While there have been incredible advances in networking bandwidth and big price declines in bandwidth costs over the past decade, no one has overcome the speed of light.  And, in order to protect data to meet enterprise needs, cloud vendors will need to offer advanced services—and increase pricing to cover the costs of offering those services.  Some emerging technologies will be introduced in 2010 that will make many of these points moot and change the game.  The biggest challenge will be  IT’s unwillingness to change.

Humans are change resistant by natureand that’s reflected pretty well in IT environments.  Once we have technology in place, we get comfortable with it because 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 are built around it, the higher switching costs become.  If the cost savings from moving to cloud storage are not incredibly compelling (and cost models suggest they can be, when operational costs are considered in addition to capital savings), users will stick to the tried and true.  That’s where we are today.  But nothing stands still, especially in the technology industry.  During 2010, expect some new technology to come to market to help get users over that hump—like gateways that bring data control, portability, protection, and security capabilities into play—and keep the cost equation intact.  No matter how 2010 plays out in the cloud space, one thing is for sure: it will be interesting!