In response to "Anders Boström" <[EMAIL PROTECTED]>: > I did some new performance-tests: > > All operations are against a directory-tree with 7,255,659,224 bytes > data in 98,025 files. > > | test1 | test2 | test3 | > --------------------------------+-------+-------+-------+ > bacula-fd, no compression, md5: | 10:25 | 10:42 | 10:15 | > bacula-fd, GZIP, md5: | 16:09 | 15:46 | 17:02 | > tar, local (1): | 8:37 | 8:53 | 8:54 | > tar + nc (2): | 9:48 | 9:52 | 9:43 | > tar + gzip + nc (3): | 14:11 | 14:26 | 15:03 | > --------------------------------+-------+-------+-------+
OK. This indicates to me that bacula is doing a damn good job. Only 15% overhead to add checksumming and cataloging features to backup. If you ask me, that's a hell of a deal. > (1) time /bin/sh -c "tar cf - directory | cat >/dev/null" > (2) time /bin/sh -c "tar cf - directory | nc -q 0 backup_server 4711" > (3) time /bin/sh -c "tar czf - directory | nc -q 0 backup_server 4711" > > This round of tests is more in line with what I expected, and the > bacula performance is quite good. The only major difference compared > to my previous tests is that the file-server disc-performance is much > better. It seems like bacula suffers much more than tar from slow > disc-performance on the file-server. backup-server and network > performance don't seem to be an issue at all in the tests, even if > write to TCP is a bit slower than /dev/null. > > However, both tar and bacula suffers from quite large slow-down when > gzip is used. This is on an Athlon 64 X2 3800+ (2-core), running > >50% idle during backup, leading me to believe that there are room for > improvement. But part of the problem might be in the linux-kernel > (2.6.17.8). At least when tar was running, the gzip process seemed to > move from one CPU-core to the other very frequently. Improvement, maybe, but not for Bacula, as far as I can see. If a dual-core system is running at 50%, then 1 core is maxed out. Since the gzip process is serialized, it can only run on one core at a time, which means the CPU is the limiting factor at this time. If you really need more speed, try using GZIP1 or turning compression off. I suppose it'd be possible to make the FD parallelize the compression, but doing that is beyond my skills, and I'm not sure you'd want it anyway. I mean, do you want a backup system that can totally monopolize all your CPU and starve other applications? > Anyway the performance of bacula is good enough for us at the > moment. Well ... I'm so glad you're please. What was your performance goal anyway? If you actually thought you'd get backup throughput at wire speed on 1g network, that was your first mistake. I don't know of any disks that can feed data that fast. Hell, from your experiment above, those disks can feed data at about 13M/sec, which is closer to 100mb than gig, and that's the absolute fastest you're going to get. > I can share one experience with you regarding > disc-performance: > > Both our two Seagate ST3500641AS discs (500Gb Barracuda 7200.9, SATA) > never completes the SMART extended self-test. It worked fine for ~6 > months, and now both run forever. The drives report 30% remaining > of the self-test, then ~2 hours later 10% remaining. After that it > goes up to 40% remaining and the cycle repeats. Also when running the > SMART extended self-test, the IO-performance is more than 10 times > lower than normal (leading to very long backup-time). > > We have been in contact with Seagate about this, and have upgraded to > the latest firmware, without any success. So I guess we will have to > RMA the discs. Are these the disks you're using to test Bacula? Please tell me you're not using hardware known to be broken as a test bed. -- Bill Moran Collaborative Fusion Inc. ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
