- From: David Schwartz <davids@xxxxxxxxxxxxx>
- Date: Mon, 26 Nov 2007 21:39:40 -0800 (PST)
On Nov 23, 9:26 pm, golden <i...@xxxxxxxxxxxxx> wrote:
My user is complaining that his writes to SAN are slower than local
His evidence doesn't support this.
He provided me a program to test this, I modified it and am
including it below. I ran the test writing 800 bytes 50000 times.
When I wrote to SAN, it took ~ 60 seconds. Local disk took ~40
seconds. If I remove the fsync, the overall operation is faster to
SAN than to local disk.
What's missing is any evidence or proof that the affect of 'fsync' is
the same on the SAN as it is to the local disk. If, for example, the
local disk is ignoring the 'fsync' or only committing the data to the
disk's cache while the SAN is committing it to permanent storage, your
user is comparing apples to oranges.
In any event, his test is kind of silly. If you use 'fsync' that much,
it would be because performance was not important.
- Prev by Date: Re: module license taints kernel.
- Next by Date: How does free() know how many bytes to free???
- Previous by thread: Re: lseek/write
- Next by thread: Running an application on startup in embedded linux