HDP on Linux – Installation Forum

Mixed hardware install

  • #46981
    Feng Meng
    Participant

    Hi, can we use mix machines with different hardware specification to setup YARN e.g. one node with 12 core and 256G memory and another node with 16 core and 512G memory?

to create new topics or reply. | New User Registration

  • Author
    Replies
  • #47019
    Dave
    Moderator

    Hi Feng,

    We recommend that all nodes in your cluster are of the same hardware specification.

    Thanks

    Dave

    #47155
    Feng Meng
    Participant

    I see. What’s on some downside of mixed hardware?

    #47163
    Dave
    Moderator

    Hi Feng,

    There can be a decrease in expected performance due to the amount of slots differing over nodes.
    As this is only going to be a 2 node cluster, I expect it is going to be a dev / pre-prod?

    Thanks

    Dave

    #47241
    Feng Meng
    Participant

    Thanks Dave. The spec I give you is just an example. We do have a small number of very powerful machines to start with, but we’d like to eventually scale out by adding more less powerful ones. It is good to know at least this can be done.

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.