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

Switch to Plain View
HBase >> mail # user >> Utility of client gateways vs client libraries connecting directly to region servers & zookeeper ?


+
Simon Majou 2013-06-05, 13:03
+
Ted Yu 2013-06-05, 14:21
+
Simon Majou 2013-06-05, 15:03
+
Stack 2013-06-06, 01:29
Copy link to this message
-
Re: Utility of client gateways vs client libraries connecting directly to region servers & zookeeper ?
I was checking the code for the RPC, and it seems that compression /
decompression is always managed by the RpcServer. Am I correct ?

So we can't skip the uncompression step (for example if we want to send to
the user the compressed data directly) ?
Simon
On Thu, Jun 6, 2013 at 3:29 AM, Stack <[EMAIL PROTECTED]> wrote:

> On Wed, Jun 5, 2013 at 4:21 PM, Ted Yu <[EMAIL PROTECTED]> wrote:
>
> > For thrift, there is already such support.
> >
> > Take a look at (0.94 codebase):
> >
> src/main/java/org/apache/hadoop/hbase/regionserver/HRegionThriftServer.java
> >
> >  * HRegionThriftServer - this class starts up a Thrift server in the same
> >  * JVM where the RegionServer is running. It inherits most of the
> >  * functionality from the standard ThriftServer.
> >
>
>
> The embedded thrift server is abandoned functionality and should be
> deprecated [1].
>
> The AVRO server has been removed because it was unmaintained.
>
> Regards REST, we already run a webserver in each regionserver so it would
> make sense attaching the REST context to the already-running http server
> context.  If you could make that happen Simon, I for one would be
> interested in getting it in.
>
> vertx.io looks nice.  Our server currently has a bunch of what it calls
> 'Shared Data' that we'd have to make fit the verticles (doesn't look too
> easy -- smile).
>
> Thanks,
> St.Ack
>
> 1.
>
> http://search-hadoop.com/m/GAFt6B9CRZ/karthik+thrift&subj=Re+Thrift2+interface
>