RDMA for Apache Spark Change Log
--------------------------------
This file briefly describes the changes to the RDMA for Apache Spark software
package. The logs are arranged in the "most recent first" order. 

Release 0.9.5 - 2018-02-19

  NEW FEATURES

    - Initial support for POWER architecture

    - Performance optimization and tuning on OpenPOWER clusters

    - Support for various communication thread binding policies

    - Support for RDMA Device Selection

    - Tested with
      - Various multi-core platforms (e.g., x86, POWER)
      - OpenJDK and IBM JDK

  BUG FIXES 

      - Fix an issue for running spark-shell with RDMA-Hadoop 1.3.0
        - Thanks to John Garbutt@StackHPC for reporting the issue


Release 0.9.4 - 2017-03-26

  NEW FEATURES

    - Based on Apache Spark 2.1.0

    - Compliant with Spark 2.1.0 APIs and applications 

  BUG FIXES

    - Fix issues for Spark SQL workloads


Release 0.9.3 - 2016-11-12

  NEW FEATURES
    
    - Built with Apache Hadoop 2.7.3

    - RDMA support for Spark SQL

    - Integration with HHH in RDMA-Hadoop
   
    - Support pre-connection mechanism

    - Support for Mellanox EDR HCA

  BUG FIXES

    - Fix a hang issue in finalize stage


Release 0.9.1 - 2016-01-05

  NEW FEATURES

    - Based on Apache Spark 1.5.1

    - Built with Apache Hadoop 2.6.0

    - High performance design with native InfiniBand and RoCE support at the
      verbs level for Spark
      - RDMA-based data shuffle
      - SEDA-based shuffle architecture
      - Efficient connection management and sharing
      - Non-blocking and chunk-based data transfer
      - Off-JVM-heap buffer management

    - Compliant with Spark 1.5.1 APIs and applications 

    - Easily configurable for native InfiniBand, RoCE, and the traditional
      sockets based support (Ethernet and InfiniBand with IPoIB)

    - Tested with 
      - Mellanox InfiniBand adapters (DDR, QDR, and FDR)
      - RoCE support with Mellanox adapters
      - Various multi-core platforms
      - RAM Disks, SSDs, and HDDs