Re: [ceph-users] recommendation on ceph pool
tim taler 于2019年3月13日周三 下午11:05写道：
>
> Hi all,
> how are your experiences with different disk sizes in one pool
> regarding the overall performance?
> I hope someone could shed some light on the following scenario:
>
> Let's say I mix an equal amount of 2TB and 8TB disks in one pool,
> with a crush map that tries to fill all disks to the same percentage.
>
> Assuming that all disks have roughly the same speed of let's say 100MB/s,
> wouldn't that hurt the performance?
>
> As a thought experiment let's say the pool consists of only two disks,
> one 2GB, one 8GB disk - both at 100MB/s
>
> If I put a 1GB file onto it that would would write
> 250MB to the small disk and
> 750MB to the big disk.
>
> leading to an overall write time of 7,5 sec.
>
> If my pool would consist of disks with the same size,
> than on both disks 500MB would be written,
> leading to an estimated time of only 5sec.
>
> Am I right here - in principle, not in exact numbers - or am I missing some
> hidden magic ('cause even the cache operations would take different
> times for different disk sizes, right?)
If your pool size is 1, and you have 2 disks with 2GB and 8GB, then
you write 1GB file,
so the 2GB disk will use 200MB and 8GB disk will use 800MB, crush
think the bigger osd will
distribute more data than small osd that controlled by crush weight;
So the best practice is deploy same capacity and performance disk in
one pool, and if you
have more than one hosts, you best have same osds in each host.
> TIA
> and best regards
> ___
> ceph-users mailing list
> ceph-users@lists.ceph.com
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
--
Thank you!
HuangJun
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] recommendation on ceph pool
I may be wrong but CEPH won’t split it into % based on disk size.
A block is wrote of 4MB to each PG that CEPH decided to use for that I/O
and replication. Yes the 8TB ones would be used more as the size / crush
algorithm will put them higher in the “chance” list but every write will be
the same size no matter the size of the disk.
So to your answer I Guess, yes the 8TB would hold more data and do more
reading. But when it comes to a write each disk will get the same amount of
data on a write. And not 25/75 or another variation.
,Ash
On Wed, 13 Mar 2019 at 11:05 PM, tim taler wrote:
> Hi all,
> how are your experiences with different disk sizes in one pool
> regarding the overall performance?
> I hope someone could shed some light on the following scenario:
>
> Let's say I mix an equal amount of 2TB and 8TB disks in one pool,
> with a crush map that tries to fill all disks to the same percentage.
>
> Assuming that all disks have roughly the same speed of let's say 100MB/s,
> wouldn't that hurt the performance?
>
> As a thought experiment let's say the pool consists of only two disks,
> one 2GB, one 8GB disk - both at 100MB/s
>
> If I put a 1GB file onto it that would would write
> 250MB to the small disk and
> 750MB to the big disk.
>
> leading to an overall write time of 7,5 sec.
>
> If my pool would consist of disks with the same size,
> than on both disks 500MB would be written,
> leading to an estimated time of only 5sec.
>
> Am I right here - in principle, not in exact numbers - or am I missing some
> hidden magic ('cause even the cache operations would take different
> times for different disk sizes, right?)
>
> TIA
> and best regards
> ___
> ceph-users mailing list
> ceph-users@lists.ceph.com
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] recommendation on ceph pool
Hi all,
how are your experiences with different disk sizes in one pool
regarding the overall performance?
I hope someone could shed some light on the following scenario:
Let's say I mix an equal amount of 2TB and 8TB disks in one pool,
with a crush map that tries to fill all disks to the same percentage.
Assuming that all disks have roughly the same speed of let's say 100MB/s,
wouldn't that hurt the performance?
As a thought experiment let's say the pool consists of only two disks,
one 2GB, one 8GB disk - both at 100MB/s
If I put a 1GB file onto it that would would write
250MB to the small disk and
750MB to the big disk.
leading to an overall write time of 7,5 sec.
If my pool would consist of disks with the same size,
than on both disks 500MB would be written,
leading to an estimated time of only 5sec.
Am I right here - in principle, not in exact numbers - or am I missing some
hidden magic ('cause even the cache operations would take different
times for different disk sizes, right?)
TIA
and best regards
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
