On Fri, Oct 06, 2000 at 08:18:57PM +1000, Robert Cohen wrote:
>
> I wanted to write it using standard IO paths as much as possible. If I
> use esoteric technolgies like the NWFS stuff, then its not clear if
> performance problems found are in the kernel or in the unusual libraries
> used.
The
Your point is taken to a certain extent.
Its true that the files here are not necessarily going to be laid our
sequentially on disk.
However, they will be laid out far enough apart to cause some seeking
which will put load on the elevator.
And even if this program isn't putting incredible stress
Your point is taken to a certain extent.
Its true that the files here are not necessarily going to be laid our
sequentially on disk.
However, they will be laid out far enough apart to cause some seeking
which will put load on the elevator.
And even if this program isn't putting incredible stress
On Fri, Oct 06, 2000 at 08:18:57PM +1000, Robert Cohen wrote:
I wanted to write it using standard IO paths as much as possible. If I
use esoteric technolgies like the NWFS stuff, then its not clear if
performance problems found are in the kernel or in the unusual libraries
used.
The NWFS
Grab NWFS at vger.timpanogas.org. It has a really good ASYNCH I/O
abstraction in kernel that is pluggable and will allow very agressive
testing of the elevator code in 2.4.0. Check the file BLOCK.C for the
2.4 support and ASYNC.C. Theres a nice way to pump ons of AIO requests
into Linux with
Greetings, Robert.
Looking over your test program, I don't think you are actually testing
the elevator algorithm at all. There are a couple of key flaws:
* The reads and writes are synchronous, so the elevator algorithm
at _most_ gets to effect the blocks within a single read or
Here are the latest results for my elv_test elevator benchmark.
This benchmark gives three numbers, a baseline figure for writing
sequentially and write and read results for writing and reading in a
pattern designed to give the elevator a hard time.
The source code is available at
Here are the latest results for my elv_test elevator benchmark.
This benchmark gives three numbers, a baseline figure for writing
sequentially and write and read results for writing and reading in a
pattern designed to give the elevator a hard time.
The source code is available at
Greetings, Robert.
Looking over your test program, I don't think you are actually testing
the elevator algorithm at all. There are a couple of key flaws:
* The reads and writes are synchronous, so the elevator algorithm
at _most_ gets to effect the blocks within a single read or
Grab NWFS at vger.timpanogas.org. It has a really good ASYNCH I/O
abstraction in kernel that is pluggable and will allow very agressive
testing of the elevator code in 2.4.0. Check the file BLOCK.C for the
2.4 support and ASYNC.C. Theres a nice way to pump ons of AIO requests
into Linux with
10 matches
Mail list logo