Skip to content

Milestones

List view

  • Replication

    If a DataNode dies, RDFS is able to re-replicate the dead DataNode's block onto the rest of the DataNodes. Acceptance test: In a cluster of 3 DataNodes and 1 NameNode, create a file with multiple blocks and then close it. Bring up a fourth DataNode, and then kill one of the original 3. If replication occurs successfully, we can kill the remaining two original DataNodes and read the file from the fourth DataNode.

    Overdue by 9 year(s)
    •
    Due by November 18, 2016
    •7/12 issues closed
    58% complete5 open7 closed

  • Demonstrate Error Simulation and Recovery

    Demonstrate error reliability and/or recovery under the following conditions: failure of k - 1 of k NameNodes, failure of 50% of DataNodes. Failure conditions must be reasonably simulated during demo.

    Overdue by 9 year(s)
    •
    Due by November 23, 2016
    0% complete0 open0 closed

  • Demonstrate Native Filesystem Functionality

    Implement use of native filesystem to pre-allocate storage on disk

    Overdue by 9 year(s)
    •
    Due by November 16, 2016
    •1/1 issues closed
    100% complete0 open1 closed

  • Completed DataNode & Zookeeper Interaction

    All interactions between DataNode and Zookeeper complete. Such interactions include functional Block reports, Heartbeats, and replication command processing between DataNode and ZK.

    Overdue by 9 year(s)
    •
    Due by November 9, 2016
    0% complete0 open0 closed

  • Full Functionality: Standalone Benchmarking MapReduce Tools

    Passes the following tests: Basic functionality tests found in hadoop-*test*.jar Basic MapReduce examples found in hadoop-*example*.jar Stand-alone benchmarking tools: Apache Pig, Apache Spark, and Apache Hive.

    Overdue by 9 year(s)
    •
    Due by December 2, 2016
    0% complete0 open0 closed

  • Basic MapReduce Functionality

    Complete basic MapReduce tasks supported by HDFS such as (AggregateWordCount, MultiFileWordCount). Demo these tests on Nov 23rd. Tests: https://github.com/apache/hadoop/tree/trunk/hadoop-mapreduce-project/hadoop-mapreduce-examples/src/main/java/org/apache/hadoop/examples

    Overdue by 9 year(s)
    •
    Due by November 23, 2016
    •1/1 issues closed
    100% complete0 open1 closed

  • End to End Filesystem Operations

    Demonstrate end to end file system operations. Should be able to create a file, write to it, close it, open it again, read out the written contents.

    Overdue by 9 year(s)
    •
    Due by November 2, 2016
    •4/4 issues closed
    100% complete0 open4 closed