Hi, In HBase, the complete key-range is split into regions (which is a sub-range of keys). A particular region is served by a single machine (called region server). If all the inserts are sequential, then they will probably fall into the same region and will be handled by the same regionserver, thus limiting the performance & write throughput. But if the inserts are random, then the keys will fall into different regions handled by different regionservers thus giving better performance. Also, there are other factors like splitting of regions during inserts.
-regards Amit ----- Original Message ---- From: rajgopalv <[email protected]> To: [email protected] Sent: Mon, 29 November, 2010 6:00:38 PM Subject: Sequential Inserts In HBASE. Hi All, I'm new to HBASE. I understand that HBASE keeps its data sorted in the filesystem. So when we insert randomly, it takes time to sort. Where as when we insert sequentially, there is no need for HBASE to sort. But, i keep hearing from some of the users that, sequential inserts to HBASE is the worst case thing. Why is that ? -- View this message in context: http://old.nabble.com/Sequential-Inserts-In-HBASE.-tp30329923p30329923.html Sent from the HBase User mailing list archive at Nabble.com.
