My problem that I don't know a good practic way to measure performance of my system I need fileserver for Win9x machines with files share with dos applications Friendly speaking I need to choose between WinNT and Linux I know there are a lot of parameters in for ex in smb.conf but how to play with it? may be there are some utilities or etc. Any help will be appreciated Thanks in advance Danila Vologdin
Danila Vologdin wrote: | My problem that I don't know a good practical way | to measure performance of my system | I need fileserver for Win9x machines with files share with dos | applications | Friendly speaking I need to choose between WinNT and Linux The "best" way is to use a real-world load in a test of both NT and Linux/Samba. Plan on spending about a man-week, spread out over two or more calendar weeks... To start, look at what the existing users are doing, record the file transfers involved, and then write a little script to reproduce the load. I assume they use local filesystems, so you can just sit there and watch a heavy user and write down the time and filename. After observing for a reasonable period (10 minutes, and hour, a day???), look to see how big the files were and make a script to use smbclient to get them at the recorded intervals. Now amend the script so you can run multiple copies of it (i.e., each script uses a different set of files) and run it against both an NT and a Linux/Samba server while recording the response time and kb/sec for each file. Increase the number of scripts until the response time starts to increase rapidly towards a value the users wouldn't like and write it up for your management. --dave [Feel free to send me email: I do this sort of thing for a living] -- David Collier-Brown, | Always do right. This will gratify some people 185 Ellerslie Ave., | and astonish the rest. -- Mark Twain Willowdale, Ontario | http://java.science.yorku.ca/~davecb Work: (905) 415-2849 Home: (416) 223-8968 Email: davecb@canada.sun.com
Peter J. Holzer wrote: | This is not enough. Many applications do not read and write files | sequentially. Yes, but that's probably too much to ask Ms. Vologdin to do on the first go-round... | his is evident for MS-Access (which reads and writes | single records), I'm on the record as recommending one never try to run a multi-user database against a remote filesystem: search for "free sample of crack cocaine" in the archives (;-)) | and also for MS-Word (which only ever loads the part | you are currently editing), but also for MS-Excel which does load and | save a whole spread-sheet at once, but reads and writes the file in some | seemingly random order (which is much slower). | There are test suites like the BapCo suite which run the actual | applications (but we couldn't find much difference between a local disc | and one shared over 10 Mbit ethernet, so at least Bapco is probably more | CPU-bound). Actually, oplocks cause local caching of the files, so the non-sequential accesses are mostly to local disk. That's possibly what fools bapco. [Andrew has a newish load-generation suite, specifically for SMB: I expect folks will eventually start using and reporting based on it...] Simple models and simple benchmarks aren't good for comparison, but in the presence of oplocks, can do a good job solving simple customer problems. --dave -- David Collier-Brown, | Always do right. This will gratify some people 185 Ellerslie Ave., | and astonish the rest. -- Mark Twain Willowdale, Ontario | http://java.science.yorku.ca/~davecb Work: (905) 415-2849 Home: (416) 223-8968 Email: davecb@canada.sun.com