Panasas, Inc. has secured a partnership with Penguin Computing, Inc. to provide high-performance parallel storage that will power Penguin Computing’s 'Penguin on Demand' (POD) high-performance cloud computing service.


                        HPC as a Service inside the Cloud

POD extends the concept of cloud computing by making optimized compute resources designed specifically for high-performance computing available on demand. POD is targeted at researchers, scientists and engineers who require surge capacity for time-critical analyses and organizations that need HPC capabilities without the expense and effort required to acquire HPC clusters.
 
“When we looked at the computing resources that our POD customers required, we knew we needed to eliminate bottlenecks and deliver a fast and seamless end user experience,” said Charles Wuischpard, CEO at Penguin Computing. “After extensive research, we settled on Panasas because their parallel storage solution offers unprecedented performance and delivers the high-level results we demand.”
 
“Cloud computing is a hot market and is gaining interest from companies of all sizes and industries due to its many business benefits.  To be a successful cloud provider, you need to provide a storage foundation that is massively scalable and easy to use, like Panasas ActiveStor,” said Terri McClure, senior analyst at analyst firm Enterprise Strategy Group. “Because Panasas has a long and impressive track record in providing storage for HPC environments, the company is well-positioned to take advantage of cloud opportunities like the one with Penguin Computing.”
 
Penguin on Demand Cloud Computing Service
POD provides a computing infrastructure of highly optimized Linux clusters with specialized hardware interconnects and software configurations tuned specifically for HPC environments. Rather than using machine virtualization, as is typical in traditional cloud computing, POD allows users to access a server’s full resources at one time, enabling maximum performance and I/O for massive HPC workloads. Comprising high-density Xeon-based compute nodes coupled with high-speed storage, POD provides a persistent compute environment that runs on a head node and executes directly on the compute nodes’ physical cores. Both Gigabit Ethernet (GigE) and double date rate (DDR) high-performance InfiniBand network fabrics are available. Jobs typically run over a localized network topology to maximize inter-process communication and bandwidth while minimizing latency.
 
Panasas ActiveStor Parallel Storage
Panasas provides the most comprehensive family of parallel storage solutions available today. Every Panasas ActiveStor system is based on the leading PanFS Parallel File System and delivers exceptional performance, scalability and manageability. Leveraging its DirectFLOW parallel storage access technology, Panasas now supports more than 500,000 servers with cost-effective, data-intensive, and linearly scalable storage solutions. This innovative architecture enables multiple parallel data paths between computing nodes and Panasas ActiveStor storage, thereby eliminating the inherent performance and capacity bottlenecks of today’s file systems and file servers. Panasas also dramatically lowers the cost of managing data storage by supporting petabytes of data capacity growth within a single, easily managed namespace and framework.