Reducing the record size would negatively impact performance.  For rational why, see the
section titled "Match Average I/O Block Sizes" in my blog post on filesystem caching:


Brad Diggs | Principal Sales Consultant | 

On Dec 29, 2011, at 8:08 AM, Robert Milkowski wrote:

Try reducing recordsize to 8K or even less *before* you put any data.
This can potentially improve your dedup ratio and keep it higher after you start modifying data.
From: [] On Behalf Of Brad Diggs
Sent: 28 December 2011 21:15
To: zfs-discuss discussion list
Subject: Re: [zfs-discuss] Improving L1ARC cache efficiency with dedup
As promised, here are the findings from my testing.  I created 6 directory server instances where the first
instance has roughly 8.5GB of data.  Then I initialized the remaining 5 instances from a binary backup of
the first instance.  Then, I rebooted the server to start off with an empty ZFS cache.  The following table
shows the increased L1ARC size, increased search rate performance, and increase CPU% busy with
each starting and applying load to each successive directory server instance.  The L1ARC cache grew
a little bit with each additional instance but largely stayed the same size.  Likewise, the ZFS dedup ratio
remained the same because no data on the directory server instances was changing.
However, once I started modifying the data of the replicated directory server topology, the caching efficiency 
quickly diminished.  The following table shows that the delta for each instance increased by roughly 2GB 
after only 300k of changes.
I suspect the divergence in data as seen by ZFS deduplication most likely occurs because reduplication 
occurs at the block level rather than at the byte level.  When a write is sent to one directory server instance, 
the exact same write is propagated to the other 5 instances and therefore should be considered a duplicate.  
However this was not the case.  There could be other reasons for the divergence as well.
The two key takeaways from this exercise were as follows.  There is tremendous caching potential
through the use of ZFS deduplication.  However, the current block level deduplication does not 
benefit directory as much as it perhaps could if deduplication occurred at the byte level rather than
the block level.  It very could be that even byte level deduplication doesn't work as well either.  
Until that option is available, we won't know for sure.
Brad Diggs | Principal Sales Consultant
On Dec 12, 2011, at 10:05 AM, Brad Diggs wrote:

Thanks everyone for your input on this thread.  It sounds like there is sufficient weight
behind the affirmative that I will include this methodology into my performance analysis
test plan.  If the performance goes well, I will share some of the results when we conclude
in January/February timeframe.
Regarding the great dd use case provided earlier in this thread, the L1 and L2 ARC 
detect and prevent streaming reads such as from dd from populating the cache.  See
my previous blog post at the web site link below for a way around this protective
caching control of ZFS.
Thanks again!
Brad Diggs | Principal Sales Consultant
On Dec 8, 2011, at 4:22 PM, Mark Musante wrote:

You can see the original ARC case here:

On 8 Dec 2011, at 16:41, Ian Collins wrote:

On 12/ 9/11 12:39 AM, Darren J Moffat wrote:
On 12/07/11 20:48, Mertol Ozyoney wrote:
Unfortunetly the answer is no. Neither l1 nor l2 cache is dedup aware.
The only vendor i know that can do this is Netapp
In fact , most of our functions, like replication is not dedup aware.
For example, thecnicaly it's possible to optimize our replication that
it does not send daya chunks if a data chunk with the same chechsum
exists in target, without enabling dedup on target and source.
We already do that with 'zfs send -D':
             Perform dedup processing on the stream. Deduplicated
             streams  cannot  be  received on systems that do not
             support the stream deduplication feature.
Is there any more published information on how this feature works?
zfs-discuss mailing list

zfs-discuss mailing list
zfs-discuss mailing list

zfs-discuss mailing list

Reply via email to