Mailing-List: contact hbase-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hbase-user@hadoop.apache.org
Received-SPF: pass (athena.apache.org: domain of sallonchina@hotmail.com
 designates 65.54.246.99 as permitted sender)
Message-ID: <BAY121-DS453AB7D896D91B304CF5AB1E80@phx.gbl>
From: "Liu Xianglong" <sallonchina@hotmail.com>
To: <hbase-user@hadoop.apache.org>
References: <BAY121-DS576F2A5F0195BE151D8E2B1EF0@phx.gbl>
 <31a243e70909030838q44447ae9r3c427542910c60c8@mail.gmail.com>
 <BAY121-DS75E064B8035A26844BA06B1EF0@phx.gbl>
 <31a243e70909030852s627bb3d5va976a3de4c1e467@mail.gmail.com>
In-Reply-To: <31a243e70909030852s627bb3d5va976a3de4c1e467@mail.gmail.com>
Subject: Re: questions about hbase tables operations
Date: Thu, 10 Sep 2009 22:36:46 +0800
MIME-Version: 1.0
Content-Type: text/plain;
	format=flowed;
	charset="gb2312";
	reply-type=original
Content-Transfer-Encoding: 8bit
Importance: Normal

Hi, Jean-Daniel Cryans. Thanks for you reply. You mean live web sites use 
three APIs. So I wander that how a search engine like Google works to 
correspond to the query. Do engineers implement a Map-Reduce program based 
these APIs to search in HBase and return the results to the user?

Thanks

--------------------------------------------------
From: "Jean-Daniel Cryans" <jdcryans@apache.org>
Sent: Thursday, September 03, 2009 11:52 PM
To: <hbase-user@hadoop.apache.org>
Subject: Re: questions about hbase tables operations

> I know that PS use the REST API, probably others, and they have MR
> jobs. Stack could tell more.
>
> So a typical system using HBase has a lot of data that doesn't fit
> will inside a RDBMS. You have to shard, denormalize, break the joins,
> etc. So HBase can be used for a live web site via one of its API: the
> native Java, the Thrift and the REST API. You can do operations like
> increments, gets, puts, deletes and scans. If you need to batch
> process your data, you do it with MR. The Bigtable paper gives good
> examples of that kind of usage.
>
> J-D
>
> 2009/9/3 ������ <sallonchina@hotmail.com>:
>> I want to find out how the applications like powerset use HBase? Should
>> these applications work based on HBase's API and map-reduce or they use
>> other tools like pig/hive to process data?
>> Furthermore, what about live queries?
>>
>> --------------------------------------------------
>> From: "Jean-Daniel Cryans" <jdcryans@apache.org>
>> Sent: Thursday, September 03, 2009 11:38 PM
>> To: <hbase-dev@hadoop.apache.org>; <hbase-user@hadoop.apache.org>
>> Subject: Re: questions about hbase tables operations
>>
>>> This questions belongs to the user mailing list btw.
>>>
>>> I do not use Pig so I can't comment on it, but what will really do
>>> with HBase is making it easy to do batch processing on HBase tables.
>>> No live queries.
>>>
>>> What do you want to do with the shell? The commands proposed there are
>>> for admin purposes. What operations are you searching for?
>>>
>>> J-D
>>>
>>> 2009/9/3 ������ <sallonchina@hotmail.com>:
>>>>
>>>> Hi, I want to make sure which tools can be used to query HBase tables,
>>>> like HBase shell. Is pig integrated with Hbase, so it can  support some
>>>> query and other operations like create\write\append to HBase tables.
>>>> I found PIG6(http://issues.apache.org/jira/browse/PIG-6) tried to solve
>>>> similar problems, but I don't know when this new feature and more 
>>>> similar
>>>> features will be released in new version of pig.
>>>>
>>>> Also I want to be clear that how powerful of HBase shell. It seems that
>>>> there are not many operations, so if lots of data expected to store 
>>>> into
>>>> HBase, does it can work effectively. I am surveying on this problems. 
>>>> Does
>>>> anyone can help me? Thanks a lot!
>>>>
>>>> Best Wishes!
>>>> _____________________________________________________________
>>>>
>>>> Liu Xianglong
>>>>
>>>>
>>>
>>
>