hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Edward Capriolo <edlinuxg...@gmail.com>
Subject Re: hadoop with db SQL Server
Date Tue, 28 Jul 2009 15:17:57 GMT
On Tue, Jul 28, 2009 at 4:26 AM, Enis Soztutar<enis.soz@gmail.com> wrote:
> Probably, the MS SQL Server does not support LIMIT OFFSET constructs. In
> which case, you can plug a MsSQLDBReader similar to the ones at
> https://issues.apache.org/jira/browse/MAPREDUCE-716.
>
> It would be very pleasant if you could open an issue, generate a patch and
> attach it, so it is committed to trunk, thanks.
>
> Enis Soztutar
>
> po po wrote:
>>
>> Ok, I will try it.Thanks
>>
>> On Tue, Jul 28, 2009 at 1:29 PM, Amandeep Khurana <amansk@gmail.com>
>> wrote:
>>
>>
>>>
>>> On Mon, Jul 27, 2009 at 9:24 PM, po po <popo6190@gmail.com> wrote:
>>>
>>>
>>>>
>>>> Hello,I try to use DBInputFormat with database sql server 2000 and got
>>>> error
>>>> "incorrect syntax near LIMIT".
>>>> Does hadoop support sql server 2000 ?
>>>>
>>>>
>>>
>>> DBInputFormat splits up the table to be read into chunks and each mapper
>>> gets to read a chunk. The splitting into chunks has different syntax in
>>> different databases. What you get by default supports MySQL. It didnt
>>> support Oracle and I tweaked it a little bit for that. I'm not sure if
>>> the
>>> same query will work for SQL Server but you can try it out. The patch is
>>> available on the HADOOP-5616 jira. Else, you can tweak the
>>> createQuery(..)
>>> routine in DBInputFormat to work with your DB.
>>>
>>>
>>>
>>>
>>>>
>>>> Thanks
>>>>
>>>>
>>
>>
>
>
I believe SQL server uses 'select top X columnlist'.  'select bottom X
columnlist'

Rather then using limit and top which are not available across all
databases is it possible to create queries based on an already
existing column that might hash out to small manageable sets per
mapper?

Mime
View raw message