Cloud Storage Strategy

The Big Sky Theory is a leftover mindset from the days of aviation. In essence, it states that "two random flying bodies are very unlikely to collide as the three dimensional space is so large relative to the bodies.". What they mean is since there's so much space up there in the sky, we don't have to worry about collisions. This sort of thinking is obsolete for obvious reasons; there's nothing quite like a close encounter with space debris, for example. Further, if the Big Sky Theory was working, we would not need TCAS (the Traffic Collision Avoidance System).  I guess the sky is not as big as we thought it was.

I believe that there exist at least three inherent advantages for major converged Telcos in the cloud market, including:

Building on existing user needs
The bandwidth consumption and revenue potential associated with moving data is a way to capitalize on your existing network assets that traditional SaaS suppliers do not have access to, as they buy bandwidth. Add this to driving mobile device sales (smartphones and tablets) and use, as well as the "stickiness" of hosting the user's data, and a second area of benefit is obtained; neither of these monetization strategies is associated with an actual fee based file service, but are instead realized by simply having an offering with significant take up.

Consideration
The large existing customer base of established network service providers gives your service offering an opportunity to move up in the "consideration" race. You will likely have a chance to have your offering "tried" at least once, which is huge. Winning the mind share and having the offering get adopted after initial use is the game changer (which is no different from any other of these sorts of services). Along these lines, the initial customer experience is absolutely critical. Ease of use, GUI design, performance, security and differentiation that is immediately obvious, is the game changer.

Lock-in from previous subscription services
There are several advantages that large network service providers have for large scale deployments. One includes the capabilities that they already have in place today; provisioning, identity management and access control and billing. Another is the experience and control systems they have for service deployments. Therefore,  a significant success for this offering can lead to additional revenues, better utilization of existing network and system assets, and create greater customer loyalty.

For these three reasons alone, incumbent Telcos have a built-in competitive advantage.  The key is taking decisive action. In the end, the decision to enter cloud storage is a leadership decision, not a technical one.  Here's to the leaders.

In the first quarter of 2010, we created a 4-phased Cloud Storage Maturity Model which still makes complete sense in 2012. The market is finding the way forward, following the blueprint we described almost to a T.

With our cloud storage maturity model in mind, here are some thoughts on the trends we're seeing this year. What we're describing is the State of Cloud Storage in 2012:

• The rise of a new application delivery architecture

We think this year everyone will figure out that cloud storage is not just cheap storage via the web, but the delivery of a new application architecture, whereby applications are fully contained on a variety of devices (smartphones, tablets and PCs) and the backend is a cloud data store with REST APIs. 

• The consumerization of everything

Enterprises as well as large institutions will seek alternatives to allowing their employees to use consumer tools like DropBox as well as alternatives to tools like box that still store data (albeit encrypted) in a public cloud (i.e. Amazon S3). This trend, aptly named the "consumerization of IT" is creating many alternatives that provide more security, and integration with key corporate support tools.  Cloud Storage is a key enabler of the consumer tools that have found favor with early and wide adoption, so look for the successful entrants with enterprise class security to be based on cloud technologies, and stored behind private firewalls so that the solutions will be as attractive to end users as the consumer tools, but that IT can control the security of their confidential corporate data and compliance initiatives.

• Disintermediation of services
  

The disintermediation of telcos and device manufacturers will continue, with the most cogent example being Apple delivering text messaging "off network" via iCloud.   Already, with devices using WiFi, from multiple suppliers, a storage cloud is a key enabler of these services.

• Manageability

Storage cloud manageability will grow as a requirement as the storage clouds get larger.  Storage clouds will become a delivery vehicle for highly complex management schemes focused on storage tiering. Let's talk ITIL in the Cloud.

Ladies and gentlemen, fellow citizens of the Cloud, the enterprise is ready for cloud storage. Let us know what you think.

Let's discuss the rapid adoption of "App Stores", and the types of applications that are being enabled for a variety of devices, primarily tablets and smartphones. While these applications are each written to deliver unique capabilities and take advantage of the platform they are designed for, as well as being written for different platforms (for example, Android apps are developed in JAVA, Apple IOS apps in Objective C) they share a common enabling technology: REST API access to a storage cloud.  

This approach is rapidly enabling new applications, which differ from older approaches like client server, virtual desktop and Web browser enabled applications in very important ways.  The backend infrastructure ( a REST API storage cloud) is a simple, massively scalable server-based solution, that can reside within the enterprise data center (a private cloud), or at an IT Service Provider (either a hosted private cloud or a public cloud). It is easy to scale, easy to manage, and by providing a consistent storage backend, it allows developers to focus on their applications and associated business logic versus worrying about the backend architecture. This provides for a "separation of concerns" in that Web, Windows, MAC, Android or IOS applications can be substantially different, but they all use the same REST API of the storage cloud they depend on. As your application use and portfolio grows, you simply scale the storage cloud, versus having to make significant changes to the server infrastructure, as is the case with client server, virtual desktop or Web browser enabled applications.

The disruptive wave is coming fast, and you will see it bringing new capabilities and applications for that large screen in your automobile (watch for car makers offering applications in App Stores), a touch screen on your refrigerator, as well as enterprise applications that will support tablets and smart phones.  Think about estimating at a job site with an application on a tablet, or estimating a property insurance damage claim with a tablet, the application is on the tablet, and the data could reside in multiple storage clouds.

Expect to see more enterprise applications taking advantage of this low cost, and easy way to deliver the kind of applications their employees and partners need.

There are multiple productive and desirable use cases for cloud storage for the enterprise.  This spans many use cases, including specialized applications that use programmable API cloud storage, file server replacement solutions, backup and archive solutions and what we call "Personal Cloud Storage."  (Personal Cloud Storage, or PCS, is the use of PC, tablet, mobile clients and web browsers by an individual to access a pool of storage.)  As opposed to a long discussion about all these, let's focus for a moment on the phenomena of the "consumerization" of IT and what if really means in the context of enterprise IT, and specifically as it relates to PCs.

The enterprise continues to register significant concerns about using public, multi-tenant storage offerings, primarily because of the "triple threat" of risk:  unauthorized access, data loss, and the opportunity for enforced access by government institutions.  In addition to the triple threat, employees are beginning to expect the enterprise IT organization to be as responsive to their needs as they know public storage service providers can and will be.  This includes easy and immediate provisioning of whatever amount of storage is required, on a pay as you go basis, at a fair price, on all your devices, and what is now represented as the consumerization of IT.

So, what's the appropriate response for the enterprise IT organization?  I believe there are three responses forming up:  

1)    We can't stop it, so let's publish acceptable usage policies and hope for the best;
2)    Just say no - you cannot do that - and then try to control it; and,
3)    Obtain an alternative that is better controlled and delivered by a trusted service provider as a "private" or "hybrid" cloud solution for your institutions use, and/or implement an in-house solution.

At Mezeo, we saw this coming and we strongly believe that the third option listed above is the appropriate response.  It recognizes the need for self service provisioning of required storage solutions along with multiple access points, but puts in place appropriate controls to avoid the triple threat.  

In Standardized Cloud APIs? Yes, Don Macvittie takes on the opposing view held by editor Mike Fratto in Standardizing Cloud APIs Is Useless.

Gentlemen, you are both right.

In a previous blog entry, I noted the following:

One last comment on this business of vendor lock in and cloud storage APIs (another focus of the OpenStack announcement).  I would submit that, while a specific set of APIs has the potential to create vendor lock in, this is a much smaller problem than what is experienced in other technologies.  If you are really worried about it, you probably have never actually written a REST API call.  It is written in many languages, and we have seen cases where applications that run on S3 run unchanged on Mezeo.  Others need very minor modifications, and some are excited to take advantage of some of the unique Mezeo API based services.  It just is not a problem, and this is much more related to FUD (fear, uncertainty and doubt) and marketing zealotry than it is associated with technological reality.  The APIs of choice will shake out, and it is far too early to say if it will be S3, OpenStack, CDMI or a combination of all of these and others yet unknown. 

At Mezeo, we have never believed there will be one winner, and instead focused on architecture to enable easy and effective delivery of whichever APIs stand the test of time. The Mezeo Cloud Storage Platform API enables advanced services and programmatic access to Mezeo enabled storage clouds.   The Mezeo Interoperability API enables interoperability of applications developed for Amazon S3, Google and Eucalyptus based storage clouds. 

(Note:  we are soon announcing our first addition to the Interoperability API that will deliver the SNIA CDMI data management capability).

The interesting view that seems to be missing here is that marketplace competition by service providers already serves to drive down the price of cloud storage, so a commoditized stack embraced by most is unlikely to yield extraordinary incremental savings.  At the same time, while the competitive market conspires to drive cloud storage costs ever lower, the need to differentiate, and deliver solutions as well as a programmable storage to enable multiple new and exciting types of applications will rapidly replace the pure cost and scale focus of current cloud storage offerings.  Sometimes, the "new" application is simply enabling it in the cloud, to produce the same result at a lower cost!  This requires significant cloud storage functionality in order to make this easy and productive.  Amazon continues to prove this with their many additions and capabilities which differentiate their service.  Mezeo sees much the same view on the part of our customers.  Their focus is on what cloud storage can do, the problems it can solve, what business opportunities it creates and what new applications it enables.  All of these views assume it will be competitively priced.

So, should we ultimately and will we ultimately achieve some sort of "standardized" API for cloud storage and cloud computing?  YES! 

What benefits will it bring?  The usual ones we expect from standardization. 

Will it be a panacea and a reason to buy?  No, but it will ultimately be a reason not to buy.

Is it the most important thing all of us in the cloud space can do right now?  No, but supporting the standards bodies is, and your suppliers should have a position on that. 

I like practical business approaches.  That is why we try to look at cloud storage through the lens of what it can be used for, why that makes sense and why it is a big business opportunity.  We developed our Interoperability API to make it easy for cloud storage solution providers to use Mezeo enabled clouds.  We built the Mezeo API to provide a robust development environment for programmable storage.  Both are important.  We did not do either one to get engaged in the "which API will prevail" struggle for becoming a standard.  We support CDMI, and are beginning to utilize it where it makes sense and adds real business value. 

I would like to close by saying that standardization will be important, it will never be perfect and it is going to take some time.  In the meantime, what I do know is that there are excellent ways to use cloud storage to solve real business problems in new ways that will yield substantial business improvements, including savings.  If you are worried about vendor lock in, get your top technical people together and do two things:

1)    Make sure that you understand that converting from one API to another, while not the most productive or best use of time, is simply not that big a deal and;

2)    Remember that there are big market forces at work here, and that your cloud storage service suppliers will be bound as much by those versus being protected by their proprietary APIs!

Check out our FREE TOOLKIT >>
CLOUD STORAGE: A BUSINESS MODEL for THE ENTERPRISE

Enterprises are increasingly looking to on-premise private storage clouds as a cost-effective way to share information.  Why?  A cloud behind the firewall enables users to easily access, share and collaborate without compromising data security, integrity and availability.  But there are still a few points every enterprise needs to consider before making the decision.

1. Consider your security needs.  How private is a private cloud?  How confidential is the data that you may ultimately store in the cloud?  Do you want the cloud behind your firewall, and if so, how will you access it?  Will you open a port and depend on a user id and password for access control, or will you also require that your cloud is accessed via a VPN?  Remember, today an employee can decide to email a confidential file to someone outside of the firewall, so what do you really accomplish with a "private cloud"?  If the cloud is available thru a port, can you monitor the cloud and shut it down when (not if) it comes under attack?  If the data being stored cannot be compromised then it cannot be on a network, at all.
   
   
2. Make sure you have adequate internal resources.  Do you have the scale, expertise and data center space to host your own private storage cloud and will you save money versus public, multi tenant storage clouds?  Assuming that your analysis of number (1) above suggests that a private cloud is a desirable approach, then our experience is that you can host your own cloud for an identified operational cost that is at least as low as the public provider.  Now, if it forces you to accelerate capital consumption in order to build out a data center, it may be a poor decision.  But all those sorts of things being equal, you can do it.  The intangible costs, like unanticipated headaches, dilution of focus on your scarce resources, and or a poor choice of cloud infrastructure can quickly turn a winner into a loser.
   
   
3. Identify a use case before starting.  A common mistake that we see is the extraordinary focus on the technology as opposed to a focus on your cloud storage use case and the business case that surrounds the use case.  The project should start with an analysis of the use case and its resulting impact on the business, and operate on the assumption you can source a cloud of appropriate size, scalability and costs.
   
   
4. Research technology solutions that are most appropriate for your use case. When you take the use case approach, you will quickly understand that a private cloud is not just a storage infrastructure; but rather an ecosystem of cloud storage clients, backup and archive solutions, special purpose data movers, management and support, that, when combined with a cloud storage infrastructure gives you a complete solution.  Once again, a use case focus will flush this out, versus a platform technology led process.
   
   
5. Integrate cloud storage with your overall cloud computing strategy.  A storage cloud is simply one layer of a cloud computing stack.  How does this cloud fit within your cloud computing stack?  Does the way in which you integrate it support the other cloud computing decisions you have made?  Evaluation of the cloud storage solution and how it will interact with, support and/or be the infrastructure associated with your overall computing cloud is a critical part of your overall evaluation.  You may also choose to use a service provider hosted "private cloud" to solve issues associated with your deployment, and not ever deploy the storage part of a compute cloud within your own data center.   

And finally, you may want to check out our FREE TOOLKIT >>
CLOUD STORAGE: A BUSINESS MODEL for THE ENTERPRISE

After looking at this snippet I decided to read the full paper from Appirio.  It actually says that 68% of "cloud adopters" expect to have the majority of their data and applications in the cloud within three years. 

What we are not told is the actual % of cloud adopters amongst medium and large enterprises. So, 68% of some subset of medium and large enterprises are racing to the cloud - not bad - but not the marketplace as a whole either.

Since "cloud adopters" are also early adopters, it comes as no surprise that they are rapidly moving to the cloud and will be done in three years.  They are also doing it because they intuitively understand that the cloud is the right thing to do, and they know that even if it is at best a break even proposition today, they will be the first to arrive at the true benefits.  Or, perhaps, they are solving something more problematic than others may be enduring, like poorly designed home grown applications, legacy application that are ready to cycle out, lack of capital for data center expansion, or any other of the many drivers that sends you to the cloud for a solution.

 Also, cloud adoption is slowed primarily by security concerns, and two concurrent activities are solving this problem on a daily basis.  One, IT organizations are winding up their early evaluations and figuring out what they can immediately live with in a cloud solution.  Second, improved security solutions are racing to market to solve these concerns associated with cloud based solutions, so more and more cloud capabilities are now deemed secure.  This trend will continue until security is no longer the blanket excuse for not becoming cloudy, and instead wide spread adoption will be governed by capacity, conversion capability, business case process and analysis, and the orderly march to the next generation of computing infrastructure.

I have a golden rule of IT technology adoption:  It always happens, and it always happens slower than the industry pundits suggest.  Cloud computing is no different.  As a matter of fact, it is very supportive of my golden rule of IT Technology, and is behaving in the usual, predictable fashion. Every time someone tries to convince me otherwise, I think about IBM, the mainframe, and the many occasions it was pronounced dead.  I went to that funeral twenty years ago and it is still growing strong.  How about the AS/400? Same story.  

Now is an exciting time to be in information technology.  What excites me about cloud computing is that it is an approach to computing service delivery that is driving significant technology investment and the promise of significant returns.  And, it is being delivered in the crucible of the open market, not by government edict, or with central planning, but with the chaos and creative destruction that only capitalism can provide.

Dan Decasper co-founded Cirtas with the vision of enabling the enterprise to utilize cloud storage as if it were a local array.

CloudStorageStrategy.com asked him a few questions about "Cloud Storage Controllers" and what the technology means for the enterprise. 

Note: Cirtas is a Mezeo Ready Solution Partner.

What was the original idea behind Cirtas?
Cirtas' co-founders previously built a WAN optimization company that was acquired by Citrix.  Together, they saw the coming of the cloud and recognized that broad scale adoption of cloud storage would require a "catalyst" technology to make the experience seamless and high performing.  They had the expertise to develop such technology and founded Cirtas to address the market need.

What is a Cloud Storage Controller?
A Cloud Storage Controller (CSC) is analogous to the array controllers found in enterprise storage systems, except instead of providing data protection, security, advanced virtualization features, and performance for an array of locally attached disk drives, a Cloud Storage Controller provides these capabilities on top of public cloud storage.  The net result is that public cloud storage can be utilized by the enterprise as if it were a local storage array - quickly, easily and for any application.

Who uses them Cloud Storage Controllers? Why do they need them?
Cloud Storage Controllers are used by mid-to-large enterprise - typically organizations with 10TB to several petabytes of data that are looking to reduce their overall cost of storage and simplify their storage environment.

These types of customers need Cloud Storage Controllers because leveraging public cloud storage services is simply not feasible without them.  The sheer number of applications running in a typical enterprise precludes them from being re-written to cloud APIs, and other concerns such as security and performance must be overcome.  Cloud Storage Controllers address all these concerns in a very robust way, allowing cloud storage to be used like local storage - but with superior economics and simplicity.

What benefits does a CSC offer?
A Cloud Storage Controller offers three main benefits:

1) It creates a seamless and highly robust connection to cloud storage, while requiring no changes to applications running in the data center.  Applications are able to access the cloud using standard block and file access protocols on what appears to be a standard logical volume.  The magic of the Cloud Storage Controller is in presenting a standard storage interface to servers and applications on the SAN, while connecting to the cloud over the WAN, and requiring nothing of the user to make this happen.

2) It accelerates the performance of applications using cloud storage through advanced WAN acceleration techniques including caching, deduplication, compression, and protocol optimization.  These techniques make cloud storage perform like local storage - even over relatively limited Internet connections (e.g. a DS3 at 45Mbps).  Furthermore, these same techniques help customers reduce their cloud storage service fees (and total cost of ownership) since they are not storing as much data in the cloud or moving as much data back and forth to it.

3) The Cloud Storage Controller provides the same features and capabilities expected of local storage arrays, such as thin provisioning, automated storage tiering, volume management, snapshots, and more.  These capabilities enhance the native functionality of the cloud, making it easier to use and providing superior data protection capabilities.

What applications/data sets make sense for the cloud?
Any Tier 2 or Tier 3 application or data set makes sense for the cloud through a Cloud Storage Controller.  The applications that don't yet make sense are highly mission critical or performance sensitive applications like large databases, ERP systems, and data warehousing.  Applications that make perfect sense are home directories, unstructured content storage, SAN offload, archiving, backup/tape replacement, and large-scale data collection (e.g. log storage)

What are your impressions of the cloud storage market today, and where it's going?  
It's an exciting time to be in the cloud storage market.  There's so much potential and things are just getting going.  Cloud storage has been very successful for web applications, and in conjunction with cloud computing.  However, it hasn't yet transcended into a tool for the mainstream enterprise.  At Cirtas we've got an opportunity to change that and we're seeing tremendous traction.  When we started two years ago, IT managers were aware of the cloud, but were fairly uneducated about it.  Today there are increasing numbers of cloud storage initiatives being driven from the CIO level, along with much more recognition of the technology among the media and analyst community as a potential game changer.  So these days it's much easier to engage with customers.  Keep an eye on the statistics published by cloud storage providers - you'll be amazed at the growth rates you see them achieve.

Las Vegas, Nevada - The Mezeo team is at the Tier 1 Research Hosting & Cloud Transformation Summit this week.  The Summit has been very well attended this year, and we are seeing an increased focus on the cloud (evident by the name change of the summit itself) and much discussion on the enterprise adoption of the cloud.  This is our third time to sponsor the summit, and we're seeing new faces like CA and Nimsoft this year.

In the opening keynote, Antonio Piraino, VP of research for Tier1 Research, talked about "phase two" of the cloud computing market. In this phase, Piraino says we'll see enterprises begin to seek hybrid solutions and telcos enter the cloud market.  He noted that the entry of the telco cloud is a significant development in that it reflects that enterprises are driving this demand for cloud-based solutions.  This change is already evident in the marketplace with many of the recent telco acquisitions of cloud based providers.

On Tuesday, I participated in a panel discussion with CA, Microsoft and Verizon Business on Enterprise Cloud Adoption.  The consensus from these big players is, like Tier 1, is that they are seeing enterprises begin to move to the private cloud, and that the public cloud adoption was coming next.

For more insights, see our cloud storage business maturity model >>