Download now Free registration required
As data sizes continue to increase, the concept of active storage is well fitted for many data analysis kernels. Nevertheless, while this concept has been investigated and deployed in a number of forms, enabling it from the parallel I/O software stack has been largely unexplored. In this paper, the authors propose and evaluate an active storage system that allows data analysis, mining, and statistical operations to be executed from within a parallel I/O interface. In the proposed scheme, common analysis kernels are embedded in parallel file systems. They expose the semantics of these kernels to parallel file systems through an enhanced run-time interface so that execution of embedded kernels is possible on the server.
- Format: PDF
- Size: 398.5 KB