December 02, 2005
Silicon Graphics announced that Lockheed Martin Aeronautics Co. has purchased an SGI Diskless Operation Server Solution to help manage multi-level classified data in lab environments. Lockheed Martin's flight simulation laboratory in Fort Worth, Texas, will now be able to reduce the costs and risks associated with changing its lab environment from one classification level to another.
Government and contractor labs often contain hundreds of computers, each with their own system drive. To change classification levels, lab technicians must wheel around carts with the disk drives and swap the drives to switch from one classification level to the next -- an approach that is costly, time-consuming, and risky. For instance, a change of classification from an Unclassified level to a Secret level of processing mandates that all disks connected to every compute system be swapped out. In this kind of computing environment, it can take several hours or even days to change classification levels with attendant risks for hardware damage and errors in security procedures.
The SGI Diskless Operation Server Solution helps to ease the burden involved in managing data at multiple levels of security in a lab environment by simplifying the process of changing classification levels. Traditional approaches of physically changing hard drives is not very practical as the number of computers in government and contractor labs continue to grow, as does the task of managing and securing an ever-increasing volume of classified data. In a lab with several simulators operating at multiple classification levels like the Lockheed Martin flight simulation laboratory, lab technicians must manage hundreds of disk drives for multiple simulators.
"The ability to rapidly change classification levels in these environments is critical. By integrating diskless boot technologies with SGI's InfiniteStorage NAS and SAN technologies, we have developed a solution that takes the time required to change classification levels in large installations like Lockheed Martin's flight simulation laboratory from hours or days to minutes," said David Egts, Senior Systems Engineer, SGI Federal. "Built on SGI InfiniteStorage technology at its core, this Diskless Operation Server architecture allows our customers to choose between a high-performance SAN or a low-cost NAS to access shared data on a system by system basis, ensuring the right balance of performance and solution cost."
The Lockheed Martin flight simulation laboratory supports the F-16, F/A-22, F-35 and other advanced aircraft programs. The laboratory provides a virtual interface for piloted evaluation of current and future military aircraft using multiple simulators, host computers, graphics systems, and databases. To swap out all the hard drives to change security levels could take days. With the SGI Diskless Operation Server Solution, as many computers as possible in the lab will now be diskless and will boot and access data from a fixed SAN assigned for each security level, reducing the maintenance costs and down time of the simulation computers and disk controllers. The goal is to be able to swap security levels in the lab in less than half an hour.
The SGI Diskless Operation Server Solution implemented by SGI Professional Services at Lockheed Martin Aeronautics will include almost 84 TB SGI InfiniteStorage Total Performance 9700 (TP9700) RAID storage array, a Fibre Channel storage array equipped with 4 Gb/second interfaces, SGI metadata servers, and the SGI InfiniteStorage CXFS shared file system to share data within its Storage Area Network.
The SGI metadata servers control file permissions and mediating shared access. Unlike network file sharing, where all data goes through the file server (which often becomes a bottleneck), once the metadata server grants access, systems with CXFS read and write data directly over the SAN to and from disk. In addition, CXFS supports heterogeneous platforms -- IRIX, Sun Solaris, Windows, 32-bit Linux, IBM AIX, Mac OS X and 64-bit Linux for SGI Altix and Silicon Graphics Prism.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?