On 7/13/12 3:02 PM, Tommi Virtanen wrote:
On Fri, Jul 13, 2012 at 12:07 PM, Matt Weil <mweil@xxxxxxxxxxxxxxxx> wrote:
Is it possible to get IO sizes 1024k and larger?
Can you be a bit more explicit in what you're asking?
Are you talking about submitting IO to RBD (via the kernel module?
inside a qemu vm?), are you talking about using the Ceph Distributed
File System, or are you talking about the IO ceph-osd submits to the
physical disk?
I think it is related. But this may not be the right list.
in 3.2 kernel with ext4 you can create file systems with a larger then
4k block size.
When a file system is created with a 1024k block size something is
limiting the IO size to 512k.
I was wondering if any one has stumbled on this and if so how did they
got kernel to use a larger IO size.
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
