Can you suggest me the mandatory hive parameters and clustering configuration steps

On Thursday 17 January 2013 12:56 PM, Nitin Pawar wrote:
looks like a very small cluster with very limited memory to run mapreduce jobs also number of map/reduce slots on nodes are less so at a time only one map is running.

but still 15 min is a lot of time for 600MB memory


On Thu, Jan 17, 2013 at 12:47 PM, venkatramanan <venkatraman...@smartek21.com <mailto:venkatraman...@smartek21.com>> wrote:

    Below details are the cluster configuration

    Configured Capacity         : 82.8 GB
    DFS Used                          : 1.16 GB
    Non DFS Used                  : 31.95 GB
    DFS Remaining                : 49.69 GB
    DFS Used%                      : 1.4 %
    DFS Remaining%              : 60.01 %
    Live Nodes <http://localhost:50070/dfsnodelist.jsp?whatNodes=LIVE>
                          : 2
    Dead Nodes <http://localhost:50070/dfsnodelist.jsp?whatNodes=DEAD>
                        : 0
    Decommissioning Nodes
    <http://localhost:50070/dfsnodelist.jsp?whatNodes=DECOMMISSIONING>
    : 0
    Number of Under-Replicated Blocks : 0

    My Select Query is:

    "select * from tweet where Id = 810;"

    This query takes 15 min to complete



    On Thursday 17 January 2013 12:29 PM, Nitin Pawar wrote:
    how many number of nodes you have for select query?
    whats your select query?

    if its just a select * from table then it does not run any
    mapreduce job
    so its just taking time to show data on your screen if you are
    using that query


    On Thu, Jan 17, 2013 at 12:24 PM, venkatramanan
    <venkatraman...@smartek21.com
    <mailto:venkatraman...@smartek21.com>> wrote:

        I didnt set any hive parameters and my total table size is
        610 MB only



        On Thursday 17 January 2013 12:11 PM, Nitin Pawar wrote:
        a bit more details on size of table and select query will help
        also did you set any hive parameters ?


        On Thu, Jan 17, 2013 at 12:12 PM, venkatramanan
        <venkatraman...@smartek21.com
        <mailto:venkatraman...@smartek21.com>> wrote:

            Hi All,

            Am Newbie in apache hive. I have create a table and
            thats points to the HDFS Folder path and its takes 15
            min to execute the simple "*select*" stmt, Can anyone
            suggest me for a best practices and performance
            improvement on hive.

            Thanks in Advance

            Venkat




-- Nitin Pawar




-- Nitin Pawar




--
Nitin Pawar

Reply via email to