Social Security Considers Private Cloud Storage

The Social Security Administration (SSA) is considering a private cloud as one option for a new enterprise disk storage device infrastructure and services that would support the agency's entire data-storage needs.

A private storage cloud with the ability to house multi-tenants and report on storage use by application--as well as isolate data from the perspectives of management and performance--is one of the topics of interest in an agency Request for Information (RFI) posted on FedBizOpps.gov.

The SSA also is seeking ideas for a system to support other needs, including a wide variety of agency computers and OSes, including IBM MVS, ZOS mainframe systems, Solaris, HP-UX, Linux, VMware, and Microsoft Windows, according to the RFI.

Support for clustered servers; high-performance enterprise and departmental storage subsystems located in various offices across the United States; and open-systems storage connectivity also are topics of interest in the RFI.

[ The Social Security Administration has been through its share of other IT overhauls recently. See Social Security Told To Improve Online Services. ]

Others include an interest in external heterogeneous disk storage virtualization that can pool physical storage from multiple devices into what appears to be one device managed from a single console; automatic tiered storage; and storage resource management for configuring, managing, and reporting on usage and performance for all types of storage use at the agency, according to the RFI.

The RFI does not outline a specific data-storage requirement for the proposed project because it has not yet been defined, said Scott Marko, the SSA's point of contact on the RFI, in an interview Thursday. He said to his knowledge, the new system likely will be a replication of the SSA's current systems, which are outlined in detail in the RFI.

Though the SSA is now just seeking information on the new storage system, "it's highly likely" there will be a specific requirement and a reward following the RFI, Marko added.

The data-storage infrastructure the SSA currently is using is spread out across two data centers on the east coast. The agency has 39 VMAX, DMX3 mainframe subsystems; 35 subsystems on other OSes; 50 VMAX, DMX3, and CLARiiON open systems; and a virtual tape library that supports 3.2 petabytes of usable data, according to the RFI.

The total amount of data on its mainframe systems is 8 petabytes of raw data, with 2.7 petabytes of that characterized as usable, according to the RFI. On other OSes, the SSA stores 12.4 petabytes of raw data and 8.8 petabytes of usable data.

In addition to its main data center, the National Computer Center (NCC) in Baltimore, the SSA also has a second support center in Durham, N.C. and plans soon to build a new national support center in Urbana, Md. The total amount of data replicated between the two data centers in use is 2.7 petabytes on mainframe systems and 50 terabytes on other OSes, according to the RFI.

Replacing its strained NCC, which was built in 1979, has been a challenge for the SSA and is a chief concern for the agency.

Last year, the SSA Inspector General said the NCC was approaching collapse and could cease to function this year; however, the SSA said it won't have the new national support center ready to replace it until 2015. To help ease the NCC's load, the agency built the Durham support center to back up the SSA's most critical data.

In addition to its data centers, the SSA also has seven remote operations communication centers in the following locations: Birmingham, Ala.; Chicago; Kansas City, Missouri; Jamaica, N.Y.; Philadelphia; Richmond, Calif.; and Wilkes-Barre, Pa.

InformationWeek is conducting our third annual State of Enterprise Storage survey on data management technologies and strategies. Upon completion, you will be eligible to enter a drawing to receive an Apple 32-GB iPod Touch. Take our Enterprise Storage Survey now. Survey ends Jan. 13.