Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
HBase, mail # dev - ByteBuffer usage


Copy link to this message
-
ByteBuffer usage
Matt Corgan 2011-09-20, 19:03
Ryan said this on an earlier email:

-----
So if the HCell or whatever ends up returning ByteBuffers, then that plays
straight in to scatter/gather NIO calls, and if some of them are DBB, then
so much the merrier.

For example, the thrift stuff takes ByteBuffers when its calling for a byte
sequence.
---------

I've also heard others mention trickiness getting the KeyValues from server
to client.  What is the general method right now?  Does it use ByteBuffers
or OutputStreams, operate in batches, achieve zero-copy, stream things to
the client, etc...  What would the ideal be?

If the KeyValue interface supported these 3 methods, would that be good?

int appendToV1ByteArray(byte[] output, int offset);
int appendToV1ByteBuffer(ByteBuffer buffer);
int appendToV1OutputStream(OutputStream os);

Ryan - i guess i'm not sure how the scatter/gather NIO calls work or where
they come into play.  Can you shed some light there?

Thanks,
Matt