Message boards :
Number crunching :
Linux for a 'Doze guy........
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
I would agree with Crunch3r. Best to start with a simple to setup distro like OpenSuse. And from reading msattler's posts I KNOW he is of the character that will be wanting to compile his own kernel and optimise everything just so, meaning FEDORA CORE will be the best later on. I personally like Ubuntu, because you get so much 'stuff' with the distro, but it fits into neither simple nor state of the art category. [because you get so much stuff it can get confusing as which stuff to pick] "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
Sutehk Send message Joined: 11 Jun 99 Posts: 42 Credit: 1,443,674 RAC: 0 |
Is running BOINC under linux any faster than running under windows? |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
Is running BOINC under linux any faster than running under windows? Don't know. But this is more important for applications than for the core client. I am running SETI, Einstein and QMC and I've read posts in the Einstein forums complaining that Einstein was slower in Windows than in Linux environment. Since I only run Linux I cannot confirm. Tullio |
Palomar Jack Send message Joined: 9 Nov 04 Posts: 44 Credit: 503,405 RAC: 0 |
Just curious...... I have something that might help. As most people know that by moving your swap file to another drive it helps speed up disk access. What they don't know is this also decreases Windows susceptibility to system file corruption. I found this out back in the days of Win 3.x while using it to speed up disk access on 386s and up. That one would corrupt it's system files, it seemed, daily. It all came to a screeching halt when I went to two drives. Now, there will be those that will say it's hogwash. Fine, if they're comfortable in being wrong and don't mind having to repair their file system at every turn, that's ok with me. It is also one of the reasons, I suspect, that Linux is more stout. What I suspect is happening is that during a crash, either from a hardware or software error the kernel looses track of or control of the hard drives heads. They can only do what they're told, and if they're told by malfunctioning kernel code to write data to the swap file and it's at another location on the disk it will corrupt the data file it wrote the swap file data over. This is made worse because XP is very swap file usage intense. For safety purposes only, you don't need the fastest "swap drive", just something that will play well with your existing system drive if they're on the same controller. There are many publications describing the procedure, so I won't go into it. But in Win XP if you want to be able to hibernate and write log files after a crash you need at least a 2Mb swap file in the same partition as your system files. Also, you should in general set your swap file maximum size at twice the amount of memory you anticipate getting. If you've only got one drive you can still make this work. During installation partition your drive into three parts. First, decide decide how much swap file you need. Then how much temporary space you need. I'd suggest at least 2Gb. Add the two together and subtract that from the total drive space and make your first partition that size. Make a tmp partition and finally your swap partition. The reason you put the tmp in the middle is to provide a "DMZ" between it and the system partition so if the kernel looses it's mind the heads will be farther away from the system files, and who cares if you wipe out some temp files. You can move the temp directories to the tmp partition with X-Setup from xteq.com. That way you don't have to muck about in Reg Edit. It's a bit of extra work, but I hope it's worth it. Hasta... |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.