DriveScale provides hardware and software technology that allows the separate purchase and scaling of commodity compute and storage in the form of storage light servers and commodity JBODs full of commodity disks, and the flexible binding of storage units to compute elements in any ratio required by an application. As the applications change, these bindings can be dissolved and new bindings applied all via software control.
DriveScale’s technology is designed to serve the Infrastructure requirements of Big Data/Hadoop and other Scale-Out applications, by providing a composable architecture where disk drives are disaggregated from servers with the ability to re-compose servers with disks on demand. We do this in a performance neutral way while maintaining all the architectural principles and design criteria for such infrastructure including data locality, the use of commodity hardware and unmodified software.
With DriveScale, you can use diskless or disk-lite servers and commodity external direct attached storage products to create rack-scale solutions that provide the operational benefits of:
Server to disk ratios and cluster creation are done from a fluid resource pool. DriveScale’s management system presents a unified view of the available resources and that of utilization and consumption. This helps to break down silos of hardware and reduces wasteful consumption.
With DriveScale, you can: