I was just corrected on an important detail: this is not our 100 SSD
setup, but a SSD and HDD mix. The index pool is on a replicated SSD
pool. The actual data pool is erasure coded and on HDDs. Everything else
remains true though.
On 6/3/22 12:40, J-P Methot wrote:
Forget my last reply, our client does split backups into several
objects. It also does use the DeleteObjects call though.
On 6/3/22 11:29, J-P Methot wrote:
Our client does use DeleteObjects when it has to delete many objects.
However, we're never close to delete 1000 at a time and our terrible
speed issue also happens when we delete objects one by one. These are
backups, so we're talking 100 GB objects.
On 6/3/22 11:02, David Orman wrote:
Is your client using the DeleteObjects call to delete 1000 per
request?:
https://docs.aws.amazon.com/AmazonS3/latest/API/API_DeleteObjects.html
--
Jean-Philippe Méthot
Senior Openstack system administrator
Administrateur système Openstack sénior
PlanetHoster inc.
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
