Pig Tutorial – performance problems?

to create new topics or reply. | New User Registration

This topic contains 3 replies, has 3 voices, and was last updated by  tedr 1 year, 10 months ago.

  • Creator
  • #26110

    Hello, just recently downloaded and installed the sandbox (deployed on VirtualBox) on my laptop and am at the Pig tutorial. I have created the script, syntax check is ok, explain plan shows no issues, and I am executing the job.

    The blue progress bar gets to about 5-10% and does not continue.

    I noticed in Windows task manager that my disk I/O is pegged at 100% so i decided to reduce the data set from the default amount to only about 200 “IBM” records equaling 15kb. However, the issue persists…disk is pegged at 100%.

    Also, I’ve observed via Ambari that various host components will restart. Specifically I’ve seen TaskTracker and the HBase RegionServer.

    Hardware: Lenovo T410 with 8gb of RAM. i5 CPU @ 2.4 ghz.

    Any ideas on what may be occurring that’s keeping my Pig process from completing? The Hive query tutorial worked fine even on the larger data set.

    Thanks in advance for any feedback or ideas.

Viewing 3 replies - 1 through 3 (of 3 total)

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

  • Author
  • #26646


    Hi Bryan,

    It is good to hear that the extra memory solved the problem.



    Good suggestion. I went to change the memory settings and noticed it was at 2gb. I bumped it to 4gb and that solved the performance problems. Thanks for the help!



    Hi Bryan,

    How is your day so far. If I may recommend to add extra the amount of memory to used by the VM. Possibly 1 or 2 additional G of RAM. Also , Decrease the amount of M-R in mapred-site.xml. Please keep in mind that the VM in Sandbox runs as a single node and it is not in distributed mode.

    Hope this helps.


Viewing 3 replies - 1 through 3 (of 3 total)
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.