You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by hm...@tsmc.com on 2011/07/28 05:39:10 UTC
Reading Performance
Hi,
There are 500 concurrent clients request which trying to fetch data from
HBASE through our 10 application servers.
These requests will be equally dispatched to 10 application servers.
In my application server, I am using HBASE java API to fetch hbase data
by scan key and secindary index.
It took about 60 seconds if only one request.
Increasing to over 1000 seconds, while 500 concurrent request coming.
Requests in a application server are all sharing a HBASE configure and a
HTable instance for each table.
It seems no use after I increased my region server to 20.
Any suggestions?
Thank you very much
>Our cluster with 20 machines(1U server, 4 cores, 12G RAM)
>Hadoop 20 machines
>Zookeeper 3
>RegionServer 10
>Hadoop version hadoop-0.20.2-cdh3u0
>HBase version hbase-0.90.1-cdh3u0
Fleming Chiu(邱宏明)
Ext: 707-2260
Be Veg, Go Green, Save the Planet!
---------------------------------------------------------------------------
TSMC PROPERTY
This email communication (and any attachments) is proprietary information
for the sole use of its
intended recipient. Any unauthorized review, use or distribution by anyone
other than the intended
recipient is strictly prohibited. If you are not the intended recipient,
please notify the sender by
replying to this email, and then delete this email and any copies of it
immediately. Thank you.
---------------------------------------------------------------------------
Re: Reading Performance
Posted by seven garfee <ga...@gmail.com>.
I think you'd better test your cluster independently without application
server.
And Get the performance of the hbase cluster.
Be sure your cluster is running gracefully first.
Then check if something is wrong on application servers.
在 2011年7月28日 上午11:39， <hm...@tsmc.com>写道：
>
>
> Hi,
> There are 500 concurrent clients request which trying to fetch data from
> HBASE through our 10 application servers.
> These requests will be equally dispatched to 10 application servers.
> In my application server, I am using HBASE java API to fetch hbase data
> by scan key and secindary index.
> It took about 60 seconds if only one request.
> Increasing to over 1000 seconds, while 500 concurrent request coming.
> Requests in a application server are all sharing a HBASE configure and a
> HTable instance for each table.
> It seems no use after I increased my region server to 20.
> Any suggestions?
> Thank you very much
>
> >Our cluster with 20 machines(1U server, 4 cores, 12G RAM)
> >Hadoop 20 machines
> >Zookeeper 3
> >RegionServer 10
> >Hadoop version hadoop-0.20.2-cdh3u0
> >HBase version hbase-0.90.1-cdh3u0
>
>
> Fleming Chiu(邱宏明)
> Ext: 707-2260
> Be Veg, Go Green, Save the Planet!
>
> ---------------------------------------------------------------------------
> TSMC PROPERTY
> This email communication (and any attachments) is proprietary information
> for the sole use of its
> intended recipient. Any unauthorized review, use or distribution by anyone
> other than the intended
> recipient is strictly prohibited. If you are not the intended recipient,
> please notify the sender by
> replying to this email, and then delete this email and any copies of it
> immediately. Thank you.
>
> ---------------------------------------------------------------------------
>
>
>
>