Terascala Optimizing HPC Workflows

Terascala, Inc. has unveiled its vision for making HPC more efficient for organizations that rely upon simulation, analysis, and modeling techniques to bring new discoveries and products to market faster.

The company’s goal is to promote a holistic HPC approach that values application and workflow productivity in addition to the performance of individual components.

HPC components form complex systems that interact with applications that have unique characteristics. Conventional HPC solutions focus on overcoming the I/O bottleneck that occurs between the compute cluster and the storage subsystem caused by the inability of scale-out NAS to deliver the multiple gigabytes per second of throughput performance that HPC requires.

While high throughput is important, it is only part of the solution. Terascala’s vision expands on throughput performance to include the basic tenet that HPC is a connected system of hardware, data, applications and workflows that must be optimized in combination to achieve the best results.

Terascala’s vision addresses these challenges:

• Enterprise functionality: HPC is moving into the enterprise and no longer is confined to large supercomputing centers for government and universities. CIOs want assurances that their HPC investment can deliver enterprise results. As a result, demand has grown for HPC storage appliances-which offer fast deployment and easy management-along with RAS (reliability, availability, and serviceability) functionality.
• Performance measurement: Both IT administrators and HPC users lack the data necessary to tune the storage appliance and optimize application workloads and job schedules. Compute cluster management tools track CPU, memory, and network resources but lack the depth needed for application-oriented storage performance monitoring and analytics.
• Stranded data: Often data resides within an organization’s research or product development team, where it remains indefinitely on fast scratch storage, unprotected and unavailable to the rest of the organization. In order for today’s enterprise users to realize HPC’s potential, organizations need to address how HPC data is processed, stored, distributed, preserved, and made available for re-use throughout the organization.

“Some time ago, we recognized that for HPC to be effective in the enterprise, the industry would need to adopt a holistic approach to the HPC lifecycle,” said Steve Butler, CEO, Terascala. “The fastest storage is useless if the user applications are architected in such a way that they cannot take advantage of it. Terascala’s vision of optimizing application workflows will maximize the ROI organizations gain from their large investments in HPC infrastructure.”

“Our vision involves delivering a Terascala product suite that gives organizations the capabilities and information they need to optimize their HPC workflows,” concluded Butler. “Our products provide greater performance and better information to tune applications and workflows for long-term ROI.”

Terascala sells their products through OEM partners to industries where HPC figures prominently, such as seismic processing, manufacturing, genomics sequencing, and financial services. It offers TeraOS, the OS and analytics that transform block storage and controllers into a storage appliance that delivers multiple gigabytes per second of throughput for HPC workloads.

TeraOS is integrated and offered with these partner products:

• Dell/Terascala HPC Storage Solution
• NetApp High-Performance Storage Rack