MapReduce Forum

Parallel Reducer execution never goes above two…

  • #50719
    Rupert Bailey

    I can’t seem to get my HDP instance to run any more than two reducers in parallel, even though I’ve got 4 processors 8GB ram and small JVM’s allocated to the instance. I am running a 10M row teragen’ed terasort requesting 4 reducer instances with the following command:

    hadoop jar /usr/lib/hadoop/hadoop-examples.jar terasort -Dmapred.reduce.tasks=4 ./terasort-input ./terasort-output

    Values that I have set in the mapred-site.xml file are:

    (4 reducers appear on localhost:50030/scheduler for this default queue)

    I have the following specifications:

    Hadoop Version: 1.3.2
    Install Method: Ambari
    Nodes: 1
    Operating System: Centos 6.5 Desktop
    Virtual Machine: VMWare.vmx 8GB allocated ram, 4 allocated virtual CPU’s
    Physical Machine: 16GB Ram 2 Core Hyperthreaded i5-3320 (4 threads)

    What do I need to tweek to get this to run 4 reducers at once? I have been able to make it do 4 reducers, but only two at once.

to create new topics or reply. | New User Registration

  • Author
  • #50771
    Rupert Bailey

    Okay the solution was to increase mapred.cluster.reduce.memory.mb

    mapred.cluster.reduce.memory.mb=768MB #allowed 2 reducers to run at once
    mapred.cluster.reduce.memory.mb=1536MB #allowed all 4 reducers to run at once.

    So it seems this is a cluster wide memory setting that sets the limit of reducer task memory across the cluster of each slot allocation. But that’s a guess Feel free to comment if you understand why this is the case. :)

You must be to reply to this topic. | Create Account

Support from the Experts

A HDP Support Subscription connects you experts with deep experience running Apache Hadoop in production, at-scale on the most demanding workloads.

Enterprise Support »

Become HDP Certified

Real world training designed by the core architects of Hadoop. Scenario-based training courses are available in-classroom or online from anywhere in the world

Training »

Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.