Wikibon Users Say IBM's SAN Volume Controller Leads Way for Heterogeneous Tiered Storage

Share Article

Recent findings by the Wikibon community indicate that for every $1 spent on hardware and software acquisition, an additional $.50 is expended on storage migration and provisioning related costs. Users report being able to lower this additional expense to less than $.10 by implementing automated tiered storage strategies largely relying on server and storage virtualization.

for every $1 spent on storage hardware and software, another $.50 is expended on provisioning and migration costs.

Wikibon, the open source community for unbiased IT research, advisory and best practices released results today from its latest study Minimizing Migration Pain with Automated Tiered Storage. Based on case studies with Wikibon users in the Global 2000 and input from consultants in the Wikibon Storage Professional Network, the research unveils the following trends:

*SAN-creep has gripped many organizations where islands of incompatible SANs have emerged with no easy way to share data between servers and storage arrays;

*Expensive tier 1 hardware, software and maintenance have forced customers to seek alternative strategies to accommodate hyper-storage growth cost effectively;

*Costs related to moving data and making changes to storage infrastructure are substantial. Migration and provisioning complexities often require shutting down applications to move data or provision new storage capacity. These costs often exceed $50,000 for each array migrated;

*Users indicate over the life of an array, on average, "for every $1 spent on storage hardware and software, another $.50 is expended on provisioning and migration costs." This additional outlay is being reduced dramatically to less than $.10 for each $1 spent by implementing virtualization and automated tiered storage;

*Server virtualization is becoming if not a prerequisite, a highly desirable strategy prior to or in conjunction with implementing heterogeneous storage virtualization; VMware was cited most frequently by Wikibon users as the dominant server virtualization approach;

*The best practice tiered storage strategy cited within the Wikibon user community is to take a simple approach by first establishing and communicating clear service level guidelines for tier 1. All other storage is then defaulted to tier 2 and migrated to tier 3 based on the retention policies of the organization. This ensures tiering is determined by policy, administered by storage professionals and not driven by 'squeaky wheels' within the line of business. Customers indicate they are virtualizing as much storage capacity as possible to implement these strategies;

*IBM's SAN Volume Controller (SVC) was cited by Wikibon users as the most mature and frequently the preferred solution to virtualize heterogeneous storage assets. Users cited IBM's 2,500 installations, strong references and leverage of Tivoli automation software as key drivers in their decision process;

*Users were reticent to aggressively utilize in-box tiering within tier 1 arrays as a primary strategy citing expense as the primary objection;

*Customers also cited using or evaluating EMC's Invista, Hitachi's USP and other emerging virtualization products from companies such as 3Par and EqualLogic.

Users indicate that these solutions, including IBM's SVC offer varying degrees of complexity, functionality and cost all of which bring certain advantages and drawbacks. The Wikibon community cited several adoption barriers and trade-offs with various approaches, specifically:

While supporting heterogeneous storage arrays, the use of virtualization appliances such as IBM's SVC and EMC's Invista had several drawbacks cited by users, including limited LUN support and service level concerns for some tier 1 applications. Users report having had to purchase additional appliances in order to support larger installations, meaning that storage behind separate appliances could not be shared. As well, users had to carefully test application performance and availability prior to moving applications behind the appliances into production. This was especially acute for certain tier 1 applications which require the highest levels of performance and availability. Also, automated classification and dynamic data movement remain major challenges for users.

Commenting on the results of the study, David Vellante, Co-founder of Wikibon said: "The Wikibon community concludes that migration and provisioning pain points are substantial and several solutions on the marketplace are bringing needed relief. However, there are still some adoption hurdles including the perceived lack of complete solutions to manage heterogeneous tiered storage assets. Many users continue to wait for a clear storage leader to emerge and articulate a compelling vision for the virtualized data center."

About The Wikibon Project
Wikibon is a worldwide community of practitioners, consultants, and researchers dedicated to improving the adoption of technology and business systems through an open source sharing of free advisory knowledge. The community uses a wiki to capture, manage, and distribute world-class, "how-to" storage knowledge. Founded by executives from market research firms like META Group (now Gartner) and IDC and launched in February, 2007, Wikibon's research is fully available for anyone to use and edit. All content is vetted by peers and 'super-peers' within the Wikibon community.

About the Research
This all new research is being released as part of Wikibon's Business Execution Service - BXS for Enterprise Storage. BXS is a monthly mashup of user best practices, case studies and analyst insights gathered by Wikibon members and made available to private clients. Wikibon produces only unbiased information. The community does not produce vendor-sponsored research that attempts to represent any company, products or trends in a manner that favors or penalizes any companies.

David Vellante, Co-founder


Share article on social media or email:

View article via:

Pdf Print

Contact Author

David Vellante
Visit website