We see a lot of coverage about cloud storage these days – and why it is or is not being adopted. One way to look at cloud storage adoption is to view it as an evolutionary process which changes over time, as both the organization matures and becomes adept at leveraging the new technology, and as the technology itself evolves to meet the real needs of the end-user. The common name for this sort of thinking is a “maturity model.”
With that in mind we developed this simple maturity model for cloud storage, based on the actual cloud storage adoption process we’re witnessing in the industry. We’d like to hear your thoughts – are you seeing the same trends?
PHASE ONE: Public Cloud Storage
Description
There remains significant marketplace confusion about what constitutes cloud storage. Cloud storage is a persistent storage for unstructured data accessed via Web services APIs over a network (LAN or WAN), with the additional characteristics of rapid provisioning of both new accounts of any size as well as rapid provisioning of increases (or decreases) in account size, along with a pay for use model, Some believe that cloud storage is just the provisioning and pay for use model with access method being varied between older technologies (CIFS/NFS) and http (Web services API access). Public, multi-tenant storage clouds as delivered by service providers clearly meet our definition, as traditional access methods like CIFS/NFS are not useful over the Internet.
Many technologists and almost all non technologists, make the initial mistake that cloud storage is simply the storage used when using cloud computing. In fact, a cloud computing image (CCI) may very well be provisioned and stored when not in use on traditional iscsi type storage systems, and is often dependent on very high speed access associated with a locally attached device. Many times, the data needed for the application supported by the CCI is often stored on shared storage devices within the same data center as the CCI, for application performance reasons. The data for these CCIs may also be block, or data base data. This is storage for cloud computing, but it is not “Cloud Storage”! This confusion permeates the marketplace in Phase One. Many vendors, particularly traditional storage vendors, have confused the marketplace by claiming to be cloud storage based on “thin provisioning” attributes with traditional data center access versus HTTP access. Cloud storage may also be accessed and utilized by CCI based applications, but that is not a defining attribute of cloud storage. Cloud storage is accessed by applications on both CCIs and dedicated servers, as well as clients on PDA’s and PC’s, wherever they are and whenever they need access. The use cases are very tolerant of the latency associated with the Internet. The thin provisioning and pay for use model of cloud storage does deliver the important cloud storage attribute of transferring storage costs from a CAPEX to an OPEX basis, if you are acquiring your cloud storage form a service provider on a pay for use basis.
The IT service provider space is the earliest adopter of cloud storage, for both offensive and defensive purposes. Many service providers are hosting workloads on dedicated or virtual servers (CCIs), and the workloads are new applications that utilize cloud storage from companies like Amazon S3, Rackspace Cloud Files, Nirvanix, and SoftLayer CloudLayer. Since the amount of data can be very large, it is difficult to move without downtime. And since the processing is relatively easy to move, IT service providers recognize the need for their own cloud storage service in order to provide a complete offering to their customers and to promote retention. Without the associated cloud storage, the application server workload can easily move, usually to the provider who provides the storage cloud. This is the defensive argument for service providers to offer their own storage cloud. On the offensive side, cloud storage is growing rapidly in terms of adoption, provides a new revenue stream, can attract new hosted workloads (cloud or otherwise), and drives increased (and very profitable) bandwidth use.
The web hosting industry also saw the initial development activities associated with adoption of Web services APIs, which provide many programming capabilities that are now resident in the storage, and easily enabled new applications that are delivered via the Web. These services, including tagging, searching and filtering, sharing, publishing, and collaboration, all exist within the APIs of a storage cloud, and are easily implemented within the application. While the enterprise has not yet adopted this new functionality, it has become quite pervasive within social networking apps, enabling new apps on mobile devices, file sharing services, and online file services, and backup and archive services.
Cloud storage is currently offered by only a few service providers including Amazon (S3); SoftLayer (CloudLayer); Rackspace (CloudFiles), Nirvanix, and is only available as a service. Enterprise adoption is limited to development only, primarily testing, and enterprise adoption has not yet occurred, primarily because of security concerns.
Key attributes
Adoption Drivers:
Business drivers: low cost, rapid scalability and on-demand capacity
Technology enablers: New programming capabilities
Adopters:
– SMBs/ SMEs
– Developers
– Consumers
Use Cases:
– Testing and application development
– SaaS (Consumer & SME/SMB users: Backup, file sharing, additional device storage, rich media)
Differentiators:
– SLA variability
– Pricing elements
—–
PHASE TWO: Public & Private Cloud Storage
Description: As large enterprises start to fully comprehend the benefits of cloud storage, their interest grows. While security concerns keep them from adopting the public cloud, they begin building private clouds behind their firewall. A private cloud provides them with the level of control and security that they are comfortable with and improves the utilization rates of their existing storage infrastructure, because of thin provisioning and potential for technology reuse. Enterprises start to roll out advanced capabilities such as file sharing and collaboration to their employees and their partners. The initial use of storage cloud services allow the enterprise to begin initial development of storage cloud based applications. They also start to move backup and archives into their own clouds. Since these applications do not require the highest performing storage, enterprises are able to reuse decommissioned hardware. This effectively starts the process of “tiered storage.”
At the same time, the public cloud storage offerings continue to grow. The availability of deployable solutions to create your own storage cloud begin to arrive in the market, enabling IT service providers to quickly implement storage clouds versus being faced with a roll your own development effort. Public storage cloud service offerings become more pervasive and better accepted as security and awareness increases.
Key attributes (Private Cloud Storage)
Adoption Drivers:
Business drivers: low cost, rapid scalability, high security and control
Technology enablers: new programming capabilities, cloud gateways (such as Blue Thread, Entropy)
Adopters:
– Enterprises
Use Cases:
– Application Development
– Testing
– Backup
– Archiving
– File Sharing and Collaboration
Key attributes (Public Cloud Storage)
Adoption Drivers:
Business drivers: Low cost, rapid scalability, on-demand capacity
Technology enablers: new programming capabilities, cloud gateways generating multi-cloud usage
Adopters:
– SMBs/SMEs
– Developers
– Consumers
– Enterprise Evaluators
Use Cases:
– Testing and application development
– Backup
– SaaS (Consumer & SME/SMB users: Backup, file sharing, additional device storage, rich media)
– Personal cloud storage with access clients
– Backup and archiving using cloud gateway
– Special use cases enabled by cloud gateway
– File server replacement
– Availability of CIFS/NFS access within the data center
Differentiators:
– SLA variability
– Pricing
– Scalability and performance
– Access options
—–
PHASE THREE: Public, Private and Hybrid Cloud Storage
Description: The maturity of the cloud (both private and public) has enabled many new applications which now require all the advanced services of a storage cloud (Web services API access, tagging, search, sharing, collaboration, etc). Capabilities such as Geo Access (accessing files from a repository closest to the requester) and Geo Replication (policy driven replication across geographies to facilitate disaster recovery) are realized. As Internet latency is constantly improving, more and more applications become “cloudy” in terms of storage, and cloud location becomes slightly less important as associated with performance. Cloud storage is now a requirement of developers and development platforms. Most SaaS applications also expect the availability of cloud storage. Everyone is storing everything! Most importantly, the improved security in public storage cloud offerings begins to blur the distinction of importance of security as being where data is stored (in public or private clouds). Instead, applications utilize both public and private clouds, for reasons associated with location of data, disaster recovery and backup, and CAPEX versus OPEX. Only the most sensitive data still retains a private cloud requirement. Performance is a more salient driver of where the data is stored, does it need to be on a LAN in the same data center as the application?
This use of both public and private clouds as solutions for storage, often by the same application, becomes what we refer to as the Hybrid Cloud.
Key attributes (Private Cloud Storage)
Adoption Drivers:
Business drivers: Low cost, high security and control, rapid scalability, compliance and forensics
Technology enablers: New programming capabilities, cloud gateways
Adopters:
– Enterprises
Use Cases:
– Application development
– Backup
– Archiving
– File sharing and collaboration
– Geo access
Key attributes (Public Cloud Storage)
Adoption Drivers:
Business drivers: low cost, rapid scalability, on-demand capacity, clouds become more pervasive
Technology enablers: new programming capabilities, cloud gateways generating multi-cloud usage
Adopters:
– SMBs/ SMEs
– Developers
– Consumers
– Enterprise evaluators
Use Cases:
– Testing and application development
– SaaS (Consumer & SME/SMB users: Backup, file sharing, additional device storage, rich media)
– Personal cloud storage with access clients
– Backup and archiving using cloud gateway
– Special use cases enabled by cloud gateway
– File server replacement
– Availability of CIFS/NFS access within the data center
Differentiators:
– SLA variability
– Pricing elements
– Scalability and performance
– Access options
– Multiple clouds vs. single cloud
Key attributes (Hybrid Cloud Storage – a mix of Public and Private Cloud Storage)
Adoption Drivers:
Business drivers: lowered average cost obtained via a mix of public/private cloud, reduction of DR/BC costs, optimized mix of capex and opex
Technology enablers: improved security
Adopters:
– Enterprises
Use Cases:
– Incorporates use cases for private and public clouds
Differentiators:
– SLA variability
– Pricing elements
– Scalability and performance
– Access options
– Multiple cloud vs. single cloud
—–
PHASE FOUR: Federated Cloud Storage
Description: With the advent of greater security, flexibility and interactivity, users will demand applications that provide real time dynamic interaction within their supply chain. Regardless of where their data may reside, partners, customers, employees and consumers will want a seamless, transparent access capability. Enter the Federated Cloud. Through a common management layer, Federated cloud will connect private and public clouds exposing all storage as a single name space. Through federated identity management and creation of trust relationships amongst various vendors and enterprises, authorized users (human or programmatic) will be able to authenticate to their cloud and be able to access information that resides anywhere across the globe. Excess capacity will be easily pushed over a grid and be sold and consumed as a true utility. Ultra-high utilization rates will be achieved, and within the trust circle security and compliance requirements will be defined and met. Interoperability will be ensured by continued maturity and standardization of APIs and applications.
This truly will culminate in a meaningful internet of knowledge and commerce. The “Semantic Web” has arrived! Note that, for matters of very high security, agencies and enterprises will continue to use private clouds.
Key attributes (Federated Cloud Storage)
Adoption Drivers:
Business drivers: need for real time dynamic interaction with partners/customers on different clouds, ability to sell excess capacity within the trust circle, optimized infrastructure utilization, establishment of trust relationships
Technology drivers: federated authentication and provisioning across clouds, streamlined cross-cloud management, standardized APIs
‘
Adopters:
– Service providers
– SMEs/SMBs
– Consumers
– Enterprises
Use Cases:
– Supply chain management
– Ad-hoc capacity capacity enhancement
– Non-sensitive and sensitive data hosting
Differentiators:
– SLA variability
– Pricing elements
– Scalability and performance
– Access options
– Security
– Governance and regulation compliance
—–
Based upon our experience in the marketplace, a large majority of the organizations are still in the first two phases. There is an undeniable appetite by the early adopters to be at the forefront, however, unlike many other emergent technologies, cloud storage comes equipped with a very compelling economic model and that is really helping justify the move into the cloud.
There are relatively few options for early adopters to implement private clouds that deliver the appropriate capabilities. This is why Mezeo focused on a deployable platform versus only offering cloud storage as a service. With the deployable platform, enterprises can implement their own in house cloud, and also take advantage of a “private” cloud hosted on their behalf at a service provider. See my discussion of this topic in my post: Cloud Storage for the Enterprise – Part 2: The Hybrid Cloud
In summary, those of us who hail from the IT service provider industry are very comfortable with cloud storage. We see the adoption as proceeding, and the issues are being knocked off as they arise. We are in an early technology cycle but with innovative early adopters we see a bright future.
According to a recent Gartner press release, 20% of businesses will own no IT assets by 2012:
Several interrelated trends are driving the movement toward decreased IT hardware assets, such as virtualization, cloud-enabled services, and employees running personal desktops and notebook systems on corporate networks.
The need for computing hardware, either in a data center or on an employee’s desk, will not go away. However, if the ownership of hardware shifts to third parties, then there will be major shifts throughout every facet of the IT hardware industry. For example, enterprise IT budgets will either be shrunk or reallocated to more-strategic projects; enterprise IT staff will either be reduced or reskilled to meet new requirements, and/or hardware distribution will have to change radically to meet the requirements of the new IT hardware buying points.
This is a bold statement. If we believe Gartner, it means that we are at the beginning of an explosion in cloud-based services managed by trusted providers on behalf of the enterprise. Of course not all businesses will choose this path, but a substantial number of industries can and will. As I blogged about earlier, the message from the CFO office is clear. We will see adoption rates rise dramatically as the benefits of cloud services become more obvious to business leaders.
A second point of interest is the prediction that by 2012, India-centric IT services companies will represent 20 percent of the leading cloud aggregators in the market (through cloud service offerings).
Here’s the take-away:
Gartner is seeing India-centric IT services companies leveraging established market positions and levels of trust to explore nonlinear revenue growth models (which are not directly correlated to labor-based growth) and working on interesting research and development (R&D) efforts, especially in the area of cloud computing. The collective work from India-centric vendors represents an important segment of the market’s cloud aggregators, which will offer cloud-enabled outsourcing options (also known as cloud services).
We are witnessing examples of what GE innovation consultant Vijay Govindarajan calls reverse innovation in IT. Natarajan Chandrasekaran, the CEO of Tata Consultancy Services notes:
I’ve seen the new cloud-based computing models for applications and processes gaining currency in emerging markets. Rural cooperative banks and small and medium businesses in India are actually far ahead of their western counterparts in adopting these models. In fact, companies from emerging markets, buoyed by strong domestic revenues and revival in growth, have been making adjustments to their global strategies and fine-tuning their investments in order to be part of the recovery process in the west and build on their global expansion plans.
As the enterprise embraces the cloud, they’ll need a maturity model to help them on their journey. My next post will explore what the maturity model for cloud storage looks like.
Leave a Reply