Well, it depends. Could you please provide some more details?It will help
us in giving a proper answer.
Warm Regards,
Tariq
cloudfront.blogspot.com
On Mon, Apr 14, 2014 at 11:38 AM, Li Li wrote:
> I have a full table scan which cost about 10 minutes. it seems a
> bottleneck for our application
I have a full table scan which cost about 10 minutes. it seems a
bottleneck for our application. if use map-reduce to rewrite it. will
it be faster?
hi all
I want to use hbase as a global cache. I need some advice .
getFromCache(String key){
get from hbase;
if exist return result;
value="working...";
boolean res=checkAndPut;
if(res){
value=doWork(key);
Put(key, value);
Hi Bogala,
I am assuming that you are using 0.94. Sorting Coprocessor (
https://issues.apache.org/jira/browse/HBASE-7474) is specifically built for
that. Working patch is available. However, the patch is not commited to
HBase yet so you will need to apply the patch yourself if you need it
quickly.
You can think of Endpoint Coprocessor as stored procedures in SQL DBs.
In order to use an Endpoint Coprocessor you need to build and install it on
the server side and then it invoke through HBase RPC.
You can make use of
https://blogs.apache.org/hbase/entry/coprocessor_introduction.
2014-04-14 5
For 0.96+, you can refer
to
hbase-client/src/main/java/org/apache/hadoop/hbase/client/coprocessor/AggregationClient.java
as the java client example.
Cheers
On Fri, Apr 11, 2014 at 6:20 PM, Bogala, Chandra Reddy <
chandra.bog...@gs.com> wrote:
> Thanks Yu. My understanding is, this coprocessor
You can also use "bin/hbase classpath" to see what the classpath should be.
2014-04-12 10:55 GMT-04:00 Ted Yu :
> Assuming you deploy HBase from tar ball, bin/hbase would add lib jars
> to CLASSPATH at runtime.
>
> See the following from bin/hbase:
>
> # Add libs to CLASSPATH
> for f in $HBASE_H
