Unstructured Data Accelerator (UDA)
Enterprise and research data sets are on a steep rise in terms of volume, velocity and variety. Hadoop, one of the dominant Big Data frameworks, helps organizations to store and analyze contemporary data amounts and features.
The UDA plug-in software package provides a novel shuffle approach for Hadoop’s MapReduce framework. RDMA based networks, with its low latency and high bandwidth features, build the most efficient shuffle provider for MapReduce. Compared to a 1GbE network, benchmark results show nearly double performance of Hadoop® clusters using UDA with 10GbE networks and quadruple the performance using FDR InfiniBand. UDA is a free software package, available under the Apache 2.0 License.
UDA is jointly developed by the Parallel Architecture and System Laboratory headed by Dr. Weikuan Yu from Auburn University and Mellanox.
- Accelerating Big Data Analytics over Hadoop Clusters with InfiniBand
- Storage Solutions for Big Data Applications
- Taking the Fear Out of Big Data Solutions Deployment