The multicore era has been in full-swing for a decade now, yet exploiting all that parallel goodness remains a prominent challenge. Ideally, compute efficiency would scale linearly with increased cores, but that’s not always the case. As core counts are only set to proliferate across the computing spectrum, it’s an issue that merits serious attention. Researchers from Read more…
In object storage applications, cluster sizes run the gamut from just a few nodes built into a medical imaging modality to thousands of nodes spanning multiple datacenters. Fulfilling the economic and manageability promises of a unified storage approach means using the best protection scheme for each particular use case. The proposed solution leverages both replication and erasure coding to ensure scalability, while optimizing footprint efficiency.
Distributed data grids have been receiving more attention lately due to the scalability provided by the cloud and the increasing need for users to do more with record amounts of data in near real-time. We talk with Dr. William Bain, founder and CEO of ScaleOut Software about scalability, the history of parallel computing and of course, distributed data grids–and how the cloud is changing all three.