Hello,
I’ve tried passing TRUE for DisableWriteBehind to this API, but still see
the WriteBehinds occur as usual (via the AcquireForLazyWrite, PagingWrite,
and ReleaseForLazyWrite sequence). My expectation was that I could
*completely* disable Cc driven write-behinds via this API (but the
modified/mapped-page-writer driven flushes could still occur). My intention
is do my own batched cache-flushing, if possible.
Am I missing something. Thanks.
Vipul.
Hi Vipul,
Your analysis is correct. Setting DisableWriteBehind to TRUE does not
completely disable the lazy writer’s write behind. In fact, it just
makes the lazy writer’s write behind deprioritize this stream - i.e., it
won’t try to flush dirty data on the file each time the cache manager
loops through the files with dirty data to select which dirty data to
flush, but instead will only try to flush the data once every 16
iterations. Another condition that can cause the lazy write to flush
this file is if the file has a large amount of dirty data (greater than
256K of dirty data). Of course, these limits chould change in future
Windows releases.
The bottom line is that you cannot do what you would like to do. The
best you can do it so ask the cache manager to deprioritize flushing
your data stream, but the cache manager must reserve the ability to
flush the file as it determines to ensure that the system does not
become overwhelmed with dirty data.
Thanks,
Molly Brown
Microsoft Corporation
This posting is provided “AS IS” with no warranties and confers no
rights.
-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Vipul
Sent: Tuesday, August 31, 2004 11:55 AM
To: Windows File Systems Devs Interest List
Subject: [ntfsd] Disabling WriteBehinds via
CcSetAdditionalCacheAttributes
Hello,
I’ve tried passing TRUE for DisableWriteBehind to this API, but still
see the WriteBehinds occur as usual (via the AcquireForLazyWrite,
PagingWrite, and ReleaseForLazyWrite sequence). My expectation was that
I could
*completely* disable Cc driven write-behinds via this API (but the
modified/mapped-page-writer driven flushes could still occur). My
intention is do my own batched cache-flushing, if possible.
Am I missing something. Thanks.
Vipul.
Questions? First check the IFS FAQ at
https://www.osronline.com/article.cfm?id=17
You are currently subscribed to ntfsd as: xxxxx@windows.microsoft.com
To unsubscribe send a blank email to xxxxx@lists.osr.com