Accelerating Big Data Processing and Framework Provisioning with OpenStack Heat-based HadoopSpark

Apache Hadoop and Spark are gaining prominence in handling Big Data andanalytics. Default Hadoop and Spark can not leverage the high-performancenetworking features (e.g., RDMA, SR-IOV) on modern clouds efficiently. Theseframeworks are traditionally written with sockets and do not deliver the best performance on modern high-performance networks. In this talk, we will providean in-depth overview of the architecture of Hadoop and Spark. We will examine the challenges in re-designing networking comp