Hi,

Regarding Client-side Dedup:
this then would have impact to a dedup cache on client-side. It should be 
disabled or the data be sent directly to the bigger, slower pool, isn't 
it?

Regards,
Alex



Von:    Erwann Simon <erwann.simon AT FREE DOT FR>
An:     ADSM-L AT VM.MARIST DOT EDU, 
Datum:  18.02.2014 11:58
Betreff:        Re: [ADSM-L] Dedup Question
Gesendet von:   "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>



Hi Alex,

If deduplicated data are moved or copied from a deduplication enabled 
storage pool to anoter deduplication enabled storage pool, only the unique 
missing chunks are moved or copied.

During client side deduplication, chunks are only evaluated against just 
the target pool.

-- 
Best regards / Cordialement / مع تحياتي
Erwann SIMON

----- Mail original -----
De: "Alexander Heindl" <alexander.heindl AT GENERALI DOT AT>
A: ADSM-L AT VM.MARIST DOT EDU
Envoyé: Mardi 18 Février 2014 10:45:10
Objet: [ADSM-L] Dedup Question

Hi,

I have questions regarding Client- and Serverside Deduplication.

If I have 2 primary pools (file dedup of course), where the smaller and
faster is the target for clients and the bigger and slower is the next
pool for it, how does deduplication behave?:
Clients that don't use client-side dedup: Sends all data, of course. That
data is deduplicated on Serverside, and migrated later to the bigger,
slower pool. Is the data rehydrated during migration, or will it just move
the missing chunks?
Clients that use client-side dedup: Are the necessary chunks to send are
measured against all or just the target pool? Because in the target pool
(smaller, faster) is just a small part of the client's data..
If I moreover copy all data to a dedup pool: would then the data be
rehydrated?

Regards,
Alex