Re: [Bacula-users] Plans for support block-based dedupe?

Subject:

From:

Radosław Korzeniewski <radoslaw AT korzeniewski DOT net>

To:

"Gary R. Schmidt" <grs AT mcleod-schmidt.id DOT au>

Date:

Thu, 3 Jan 2013 15:17:05 +0100

Hello,

2013/1/3 Gary R. Schmidt <grs AT mcleod-schmidt.id DOT au>

> Keep in mind that ZFS block-level deduplication is very expensive in
> terms of RAM.

I'd put it as *any* block-level de-duplication is RAM intensive... I'd
be very interested if someone had an algorithm that was not RAM
intensive - and didn't take forever to accept data! :-)

If you proper design a deduplication storage engine then you can do it without a lot of RAM.

We've got only 15% of performance loss in our lab. I think it is not so bad and it is faster then standard gzip compression. In this case the main performance problem is a client reading a data not a storage daemon handling dedupes.

best regards

--
Radosław Korzeniewski
radoslaw AT korzeniewski DOT net

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122712

_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users

<Prev in Thread]

Current Thread

[Next in Thread>

Previous by Date:

Re: [Bacula-users] Plans for support block-based dedupe?, Phil Stracchino

Next by Date:

[Bacula-users] restore from expired tapes, Rao, Uthra R. (GSFC-672.0)[ADNET SYSTEMS INC]

Previous by Thread:

Re: [Bacula-users] Plans for support block-based dedupe?, Phil Stracchino

Next by Thread:

Re: [Bacula-users] Plans for support block-based dedupe?, Silver Salonen

Indexes:

[Date] [Thread] [Top] [All Lists]