William Vaughn
3/3/2008 7:34:00 PM
Yes, of course. My migraine is getting to me.
Thanks
--
__________________________________________________________________________
William R. Vaughn
President and Founder Beta V Corporation
Author, Mentor, Dad, Grandpa
Microsoft MVP
(425) 556-9205 (Pacific time)
Hitchhiker's Guide to Visual Studio and SQL Server (7th Edition)
____________________________________________________________________________________________
"Michel Posseth [MCP]" <MSDN@posseth.com> wrote in message
news:%23cLbyLWfIHA.1132@TK2MSFTNGP06.phx.gbl...
> Huh ? Bill did you made a typo :-|
>
> Shouldn`t this be
>
> >I recommend doing
>> this processing on the sever using TSQL if possible or CLR procedures if
>> not. THIS is invariably faster.
>
> Or am i completely missing something ? :-(
>
>
> Michel
>
>
>
>
>
> "William Vaughn" <billvaNoSPAM@betav.com> schreef in bericht
> news:56F13648-9F2A-4D4D-9B4A-0E2324ABE507@microsoft.com...
>> The DataReader provides only the pipe, not the bucket. That is, it's just
>> a stream of data that must often be stored locally. Some developers think
>> they can out-write the coders at MS who implemented the DataTable Load
>> method to suck the data out of the DataReader--most fail in the attempt.
>> Others use the DataReader to bring a mountain of rows to the client to
>> extract the gold. The best solution in this case is to setup a ore
>> processor at the quarry--not in Denver 100 miles away only to ship it
>> back. I recommend doing this processing on the client using TSQL if
>> possible or CLR procedures if not. THIS is invariably faster.
>>
>> --
>> __________________________________________________________________________
>> William R. Vaughn
>> President and Founder Beta V Corporation
>> Author, Mentor, Dad, Grandpa
>> Microsoft MVP
>> (425) 556-9205 (Pacific time)
>> Hitchhiker's Guide to Visual Studio and SQL Server (7th Edition)
>> ____________________________________________________________________________________________
>> "Michel Posseth [MCP]" <MSDN@posseth.com> wrote in message
>> news:eVnD3$RfIHA.1132@TK2MSFTNGP06.phx.gbl...
>>>> Will you tell us what is the overhead.
>>>
>>> All the datatable ,dataset and there constraints , validation that you
>>> might not need
>>> A datareader is connected , and is the equivalant of a old fashioned
>>> Firehose cursor ( read only, forward only cursor )
>>>
>>> So by using the datareader`s property`s you can skip the dataset , table
>>> stuff if possible in your design and this gives you extra perfomance
>>> and is especially handy for retrieving small blocks of data ,, or in
>>> projects were you process large amounts of data ( as the data is not
>>> preloaded as is common in disconnected architecture ) ofcourse this is
>>> only suitable for readonly data
>>>
>>>
>>>>
>>>> In my idea will a datareader not make it faster, my general expirience
>>>> is even oposite from that.
>>>>
>>>
>>> I wonder in wich special situation you encountered this, cause normally
>>> a datareader should be superior in speed ( for obvious reassons )
>>> however without the strong typing etc etc
>>>
>>> Michel
>>>
>>>
>>>
>>> "Cor Ligthert[MVP]" <notmyfirstname@planet.nl> schreef in bericht
>>> news:3223643E-D70F-47C5-AC8D-38A9EF0C5AF6@microsoft.com...
>>>> Michel,
>>>>
>>>> Will you tell us what is the overhead. Now your answer look to me the
>>>> same as insert 64Gb in your 32bits computer.
>>>>
>>>> In my idea will a datareader not make it faster, my general expirience
>>>> is even oposite from that.
>>>>
>>>> Cor
>>>
>>>
>>
>
>