https://community.cloudera.com/t5/Archives-of-Support-Questions/Hadoop-Performance/m-p/143259#M48314 <P><A rel="user" href="https://community.cloudera.com/users/12065/oulaalshiekh.html" nodeid="12065">@oula.alshiekh@gmail.com alshiekh</A>add datanode if you are running out of storage capacity of cluster, add computation node when you see bottleneck in processing, by adding more computation nodes you can launch more mapreduce/spark task. you can also use your node to store data as well as to add more processing capacity(in terms of more no mapreduce tasks)</P> Thu, 08 Dec 2016 19:42:28 GMT rajkumar_singh 2016-12-08T19:42:28Z https://community.cloudera.com/t5/Archives-of-Support-Questions/Hadoop-Performance/m-p/143258#M48313 <P>We know that hadoop main purpose is increasing Performance through adding more data nodes </P><P> but my question is </P><P>if we want to retrieve the data only with out the need to process it or </P><P>analyze it )</P><P>is adding more data nodes will be useful or it doesn't increase the performance at all because we have retrieve operations only with out any computations or map reduce jobs </P> Thu, 08 Dec 2016 16:51:58 GMT https://community.cloudera.com/t5/Archives-of-Support-Questions/Hadoop-Performance/m-p/143258#M48313 oula_alshiekh 2016-12-08T16:51:58Z https://community.cloudera.com/t5/Archives-of-Support-Questions/Hadoop-Performance/m-p/143259#M48314 <P><A rel="user" href="https://community.cloudera.com/users/12065/oulaalshiekh.html" nodeid="12065">@oula.alshiekh@gmail.com alshiekh</A>add datanode if you are running out of storage capacity of cluster, add computation node when you see bottleneck in processing, by adding more computation nodes you can launch more mapreduce/spark task. you can also use your node to store data as well as to add more processing capacity(in terms of more no mapreduce tasks)</P> Thu, 08 Dec 2016 19:42:28 GMT https://community.cloudera.com/t5/Archives-of-Support-Questions/Hadoop-Performance/m-p/143259#M48314 rajkumar_singh 2016-12-08T19:42:28Z