|
|
Posted by boa on 08/21/06 14:43
* Robert Klemme wrote, On 21.08.2006 14:01:
> On 21.08.2006 14:26, boa sema wrote:
[snip]
>>> I see these issues with your theory:
>>>
>>> - Disks with cache on them will optimize accesses to minimize
>>> latency created by head moves.
>>>
>>> - With a RAID since disk != physical disk there is not a single head
>>> that moves and read and write requests are scheduled by the RAID
>>> controller to optimize IO operations.
>>
>> I agree that these issues are real, OTOH isn't "my theory" also the
>> reason one places the transaction logs on separate drives? If not,
>> what's the difference?
>
> Access patterns are completely different for data files and TX log. In
> an OLTP environment TX log is continuously written and you need a RAID
> level that supports fast writing and high reliability. Data files are
> read and written with arbitrary (although optimized) access patterns.
I meant during import. I haven't read the sql server algorithm for page
and extent allocation when inserting new rows in a table, but assuming
that sql server will start adding rows at page 0,extent 0 (also assuming
proper clustering of table data and the order of the data inserted) and
then just go from there, the access pattern should be pretty similar,
shouldn't it?
Boa
[snip]
Navigation:
[Reply to this message]
|