I am utilizing Spring + Datanucleus JDO + Hbase. Hbase is on a completely dispersed mode with two hubs. I am confronting genuine execution issues here.
My webapp can be considered as a pinger which just continues pinging URLS and stores their reaction. Hnce my application runs numerous strings for INSERT into db. I have watched that once the quantity of simultaneous composes surpasses around 20 , the supplements begin taking a considerable measure of time (some take even 1000 secs). What's more, when this happens READS begin flopping as well and my webapp is not ready to separate any information from the db (my webapp hangs). I am a sorry NoSQL db fellow and subsequently don't know where to begin searching for execution.
My real arrangements are: Zookeeper majority estimate: 1 Hbase regionservers: 2 Data Nodes: 2 hbase.zookeeper.property.maxClientCnxns: 400 replication factor:3
Do I have to build the store measure for Hbase ? Should a high WRITE throughput have impact on READ ?
Am I accomplishing some kind of problem with the setup? It appears written work to a document would be speedier that composition information to Hbase . This is my last shot at Hbase. If it's not too much trouble offer assistance
... View more
Simply need to request a few contributions on the issue that we are experiencing at this moment. The fare work all of a sudden is not working. Whenever activated, it simply conjure the show screen however does not do anything. This was working some time recently. No progressions were made on anything.
We checked the fly up blockers and it is killed. Restarted the server and furthermore checked tablespace and datafiles are all OK. Checked the virtual memory amid the exportation of information is additionally OK. There is no blunder message on the screen or in the log documents.@tekslate
Any idea what may be the cause of the problem?
Thanks in advance,
... View more