Messages in this thread |  | | Date | Thu, 20 Jun 1996 14:14:15 +0100 | Subject | ext2fs "performace" | From | Stephen Tweedie <> |
| |
Hi,
In article <199606201225.OAA16689@ceres.tat.physik.uni-tuebingen.de>, koenig@tat.physik.uni-tuebingen.de (Harald Koenig) writes:
> Hi, > running Linux-2.0.0 on a 83 MHz Pentium OverDrive with fast SCSI disk > using a 2.6 GB ext2fs partition with only two files with 1GB size each > I removed these files. removing only one file took 100 seconds real time !!
> in writing the file CPU and memory bandwidth seem to be the bottleneck > (32 bit DRAM in ASUS SP3G board; 486 Saturn chipset; disk has >= 7MB/sec > formatted media transfer rate connected to fast/sync NCR 53C810).
> but why does removing this file takes so much time ? > that's only a "removage rate" of ~11 MB/sec :-(
A 1 GB file on a 1k block ext2 filesystem will have 4096 indirect blocks and a few dindirect blocks. Deleting the file will involve essentially doing a random-access seek and read of each of these blocks, so if it takes 100 seconds you are getting over 40 seeks/reads per second.
Just out of interest, how many seeks/second do you get running bonnie on the same setup? Bonnie _ought_ to give better rates, because the command setup latency for SCSI is quite significant, and bonnie forks multiple readers when measuring seek rates. If you only have one process doing the reads sequentially, then the command setup latency will be much more significant.
Cheers, Stephen. --- Stephen Tweedie <sct@dcs.ed.ac.uk> Department of Computer Science, Edinburgh University, Scotland.
|  |