Where Big Compute Meets Big Data
Enterprises have long understood the need to deploy their infrastructure, particularly storage, in an appliance form factor. Parallel file systems are no different. Researchers, scientists, and engineering must innovate quickly and with confidence.
Bringing new ideas to market places unique demands on the compute environment. Today in manufacturing, for example, modeling scenarios containing over a billion cells are becoming commonplace to deliver high levels of fidelity for pinpoint accuracy. Further, multiple simulations must be run for each project, so the storage system needs to be able to scale linearly to handle the throughput requirements. To achieve this agility and deep level of insight into the design, the applications require larger data sets at higher network throughput.
Terascala works with industry storage leaders to provide the only solution that allows multiple R&D users or applications to work with terabytes or petabytes of data at an aggregate read and write performance of multiple gigabytes per second. All of this is packaged in a simple, complete, and scalable appliance.
By providing modular components that address performance and management, Terascala abstracts the complexities of deploying and managing a parallel file system. A Terascala-powered solution can be up and running in hours—a significant improvement over the many months that enterprises spend trying to develop their own solution or paying for consultants to do it for them.
All the hardware and software required to deliver reliable performance and system availability are included in the Terascala appliance. The appliance is tightly integrated to our partners’ controllers and storage to deliver a predictable read/write performance. Once up and running, enterprises can operate, monitor, and manage the system without the need for specialized skills.
In order for R&D to deliver high fidelity simulations, analysis, and modeling, Terascala has ensured that the user interface provides application-aware visibility and insight. Enterprises can have multiple projects working in parallel and immediately see the performance and also understand how to tune performance for the given workload mix.