The primary difference between traditional progressive incremental backup
and journal based backup is the method
in which the client uses to determine what objects to backup or expire.

Progressive incremental backup obtains the backup candidate list by
building and comparing the list of active previously
backed up objects stored on the TSM server with the list of objects
currently residing on the local filesystem.

The local object list is built by scanning the entire file system, the
server list is received over the network, and the candidate list is
built by comparing the two lists.

All three lists reside in memory until the backup is completed.

Journal based backup derives the backup candidate list by contacting the
TSM journal daemon and processing the contents
of the particular file system journal.

The journal daemon is local process (service) which maintains journal
databases for specified filesystems by monitoring
the file system for change activity.

The performance saving in terms of both memory, i/o, and network traffic
for journal based backup is that only objects which
are actually backed up/expired are processed, whereas progressive
incremental has to process the entire file system.

So  the benefit is really only realized when backing up file systems with a
very large number of objects with a small to moderate
amount of daily change activity (for example, a file server with 2 million
objects of which only about 5% to 10% change daily).

Smaller file systems which have high volumes of change activity probably
will not see that dramatic of an improvement
over traditional progressive incremental backup


Pete Tanenhaus
Tivoli Storage Solutions Software Development
email: tanenhau AT us.ibm DOT com
tieline: 320.8778, external: 607.754.4213

"Those who refuse to challenge authority are condemned to conform to it"

---------------------- Forwarded by Pete Tanenhaus/San Jose/IBM on
06/03/2002 01:40 PM ---------------------------
06/03/2002 01:40 PM ---------------------------

John Bremer <jbremer AT LANL DOT GOV>@VM.MARIST.EDU> on 06/03/2002 12:58:56 PM

Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>

Sent by:    "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>


To:    ADSM-L AT VM.MARIST DOT EDU
cc:
Subject:    Re: Journaling



Just trying it on a couple of Win2K machines, I don't see such a dramatic
performance boost.  However these are clients with only several thousands
of files, not millions.


At 09:43 AM 6/3/02 -0500, you wrote:
>Can anyone share their thoughts, opinions, war stories on Journaling for
>Windows NT4.0 and W2K clients??
>
>Thanks,
>
>Matt Adams
>Tivoli Storage Manager Team
>Hermitage Site Tech
>Deloitte & Touche USA LLP
>615.882.6861
>
>- This message (including any attachments) contains confidential
information
>intended for a specific individual and purpose, and is protected by law.
-
>If you are not the intended recipient, you should delete this message and
>If you are not the intended recipient, you should delete this message and
>are hereby notified that any disclosure, copying, or distribution of this
>message, or the taking of any action based on it, is strictly prohibited.