September 18, 2012
Sept. 18 — Panasas, Inc., the leader in high performance parallel storage for technical computing applications and big data workloads, today announced Panasas® ActiveStor™ 14, the world’s fastest fully integrated parallel storage solution. Through its unique application of Solid State Drive (SSD) flash technology, ActiveStor 14 achieves a dramatic leap forward in small file performance without the prohibitive cost of SSD-only solutions. The single platform will satisfy the bandwidth requirements of the most demanding high-performance storage applications while opening new opportunities for Panasas parallel storage technology in industries with I/O intensive workloads, including finance, life sciences, energy and manufacturing.
“As big data workloads become pervasive within the enterprise, a parallel architecture is a necessary weapon to address critical analysis in a timely manner in order to drive competitive advantage,” said Faye Pairman, president and CEO at Panasas. “ActiveStor 14 broadens the appeal of our fully integrated parallel file system appliances by supporting a diverse set of workloads and applications—all from a single scale-out NAS solution.”
ActiveStor 14 leverages cost-effective enterprise SATA drives to deliver an extremely affordable, highly scalable storage appliance providing more than eight petabytes of easily deployed, easily managed capacity in a single global namespace. The Panasas file system efficiently incorporates SSD technology to accelerate small file performance and metadata operations while minimising SSD wear and expense. This intelligent blend of cost-effective, high density SATA and SSD storage assures industry-leading throughput performance for mixed workloads consisting of both large and small files while delivering fast rebuilds for always-on performance.
“Scale-out storage systems like ActiveStor 14 are needed to keep pace with the increasingly challenging big data workloads found in technical computing environments.” said Steve Conway, HPC research vice president at IDC. “The use of SSD technology to accelerate namespace operations and small file performance is a solid approach aimed at benefiting a wide variety of customers by addressing their mixed workload requirements.”
Panasas demonstrated ActiveStor 14 performance for mixed workloads by publishing compelling SPEC SFS benchmark results measuring network-attached storage NFS performance. A single ActiveStor 14T shelf containing only 27 data drives delivered 20,745 SPECsfs2008_nfs.v3 operations per second with an overall response time of 1.99 milliseconds—a stellar 768 ops/s per data drive. Scaled out to two shelves, ActiveStor 14T produced 41,116 SPECsfs2008_nfs.v3 operations per second with an overall response time of 1.39 milliseconds. This showed near-linear scaling of performance as you would expect from a Panasas system.
Panasas and its global partners are taking orders for ActiveStor 14 effective immediately with initial shipments in November 2012. Multiple product configurations are available with varying SSD and cache capacities. Panasas will demonstrate ActiveStor 14 this week at the HPC for Wall Street conference in New York City and the Trading Architecture Europe conference in London.
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?