Re: Re: Linux/Windows Universal Benchmark
On Thu, Apr 15, 2004 at 12:13:59AM +0100, Andy Morris wrote: Just been browsing and came across this dicussion, and thought i might be able to give some input. I'm a software grad (well this summer) and use both linux(gentoo 2.6.xxx cant remember specifically) and xp pro sp1 as home os's (games, dev etc) and when building a fairly small application i felt that it was taking too long to build my code using linux (approx 70 secs). So I logged into my windows and it did it in under 40 secs, quite a difference. You are also testing compilers here and gcc isn't a very fast compiler. The resulting code is good though. If you are comparing it to msvc you also need to make sure that its not using precompiled headers which would affect the compilation speed. However, these two machines were of different spec xp:2500xp(1.833mhz) barton 512cache,1gig ram, 7200 sata hdd (high spec) linux:1900xp(1.6mhz) thoroughbred 256cache, 1gig ram, 5400 ide hdd (fairly low spec) I would check your setup since for me linux is usually much faster then xp. I did a test once with matlab doing some heavy computation that took several minutes. Running matlab on a laptop running linux with a 1.3G amd athelon cpu with 256M ram and 5400 rpm ide disk, finished at the same time as a 2G pIII desktop with 2G ram and 7200 rpm disk running XP pro (it didn't hit the disk though so I don't think the amount of ram was a factor). I would accept a relatively small diff between the two machines (but a difference since one is more powerful) and since there were approx 150 source files the hard drive reads could have caused a difference. So what I had was a biased test result. My first thoughts would be to write a tiny application that would not use the harddrive and compare again, so i made a simple program to count from 0 to a parameter x number of times, test data was to count 0-9 100 times. XP box did it in 3mins 10 secs, linux 5 mins 4 secs ( i did this numerous times and results were always v similar). Again big difference, so it definetely wasn't the harddrive's access time etc. So the only way I was going to find out was to install Linux on my xp box and dual boot it (which I did). So XP and Gentoo Linux on the same box, and xp still beats it hands down (it only improved by around 30 secs), which really dissapointed me. It makes me want to use my xp box for more stuff than previously, which i had been trying not since i like using kde over the Windows desktop Now one thing I should have mentioned earlier is that the app was written in Java (compiled using Sun's jdk1.4.2 (not gentoo's blackdown since it's too buggy)) and we are therefore also testing the platform implementation of the VM (if you don't know about java basically when code runs there is a middleman between the code running and the OS (the Java Virtual Machine). It is possible that the jdk does not work as well on Linux but this is what I use and so do millions of others, and therefore it's can be a v good benchmark. I will shortly if i get some time repeat the tests in C++ to remove this factor (it interfaces directly with the OS since it's compiled into native binaries) and if any1 does care for the result then let me know. You are testing java implementations and optimizations. If you want to have a better comparison you could try running sun's jdk on both machines, although I don't know how similar the implementation for windows and linux is. Now if you think my benchmark is totally unsafe/inaccurate as a means for benchmarking then feel free to state so, but compiling and running java code is something I do frequently, and Linux can't seem to do it as quickly as XP. Just to sum up my findings, from my testing XP seems to run simple sequential cpu tasks not much less than twice the speed of Linux. My afterthoughts about the two OS's is that if I ran multiple threads of my test (say 500) that my XP box would flake out and linux would handle it much better, perhaps when i go away for the wkend next i'll do that but right now iv got far too much work to do on them (final yr proj). Hope this helped any1, -andy _ It's fast, it's easy and it's free. Get MSN Messenger today! http://www.msn.co.uk/messenger -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] +++ This Mail Was Scanned By Mail-seCure System at the Tel-Aviv University CC. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: Linux/Windows Universal Benchmark
Just been browsing and came across this dicussion, and thought i might be able to give some input. I'm a softwaregrad (well this summer) and use both linux(gentoo 2.6.xxx cant remember specifically) and xp pro sp1 as home os's (games, dev etc) and when building a fairly small application i felt that it was taking too long to build my code using linux (approx 70 secs). So I logged into my windows and it did it in under 40secs, quite a difference. However, these two machines were of different spec xp: 2500xp(1.833mhz) barton 512cache,1gig ram, 7200 sata hdd (high spec) linux:1900xp(1.6mhz) thoroughbred 256cache, 1gig ram, 5400 ide hdd (fairly low spec) I would accept a relatively small diff between the two machines (but a difference since one is more powerful) and since there were approx150 source files the hard drive reads could have caused a difference. So what I had was a biased test result. My first thoughts would be to write a tiny application that would not use the harddrive and compare again, so i made a simple program to count from 0 to a parameter x number of times, test data was to count 0-9 100 times. XP box did it in 3mins 10 secs, linux 5 mins 4 secs ( i did this numerous times and results were always v similar). Again big difference, so it definetely wasn't the harddrive's access time etc. So the only way I was going to find out was to install Linux on my xp box and dual boot it (which I did). So XP and Gentoo Linux on the same box, and xp still beats it hands down (it only improved by around 30 secs), which really dissapointed me. It makes me want to use my xp box for more stuff than previously, which i had been trying not since i like using kde over the Windows desktop Now one thingI should have mentioned earlier is that the app was written in Java (compiled using Sun's jdk1.4.2 (not gentoo's blackdown since it's too buggy)) and we are therefore also testing the platform implementation of the VM (if you don't know about java basically when code runs there is a middleman between the code running and the OS (the Java Virtual Machine). It is possible that the jdk does not work as well on Linux but this is what I use and so do millions of others, and therefore it's can bea v good benchmark. I will shortly if i get some time repeat the tests in C++ to remove this factor (it interfaces directly with the OS since it's compiled into native binaries) and if any1 does care for the result then let me know. Now if you think my benchmark is totally unsafe/inaccurate as a means for benchmarking then feel free to state so, but compiling and running java code is something I do frequently, and Linux can't seem to do it as quickly as XP. Just to some up my findings, XP runs simple sequential cpu tasks not much less than twice the speed of Linux. My afterthoughts about the two OS's is that if I ran multiple threads of my test (say 500) that my XP box would flake out and linux would handle it much better, perhaps when i go away for the wkend next i'll do that but right now iv got far too much work to do on them (final yr proj). Hope this helped any1, -andy
Re: Linux/Windows Universal Benchmark
Andy Morris wrote: So XP and Gentoo Linux on the same box, and xp still beats it hands down (it only improved by around 30 secs), which really dissapointed me. It makes me want to use my xp box for more stuff than previously, which i had been trying not since i like using kde over the Windows desktop Now one thing I should have mentioned earlier is that the app was written in Java (compiled using Sun's jdk1.4.2 (not gentoo's blackdown since it's too buggy)) and we are therefore also testing the platform implementation of the VM (if you don't know about java basically when code runs there is a middleman between the code running and the OS (the Java Virtual Machine). It is possible that the jdk does not work as well on Linux but this is what I use and so do millions of others, and therefore it's can be a v good benchmark. I will shortly if i get some time repeat the tests in C++ to remove this factor (it interfaces directly with the OS since it's compiled into native binaries) and if any1 does care for the result then let me know. I for one am definitely interested in hearing the results. I suspect that different type functions will result in different results however. For example, something involving network access might be just the reverse. If your test app is not GUI-dependent, what happens if you shut down X and all the extraneous daemons, etc, and then run the program on Linux? In other words, there may be more processes on the Linux box that prevent your app from having as big a slice of processor time as it has on the Windows box. (I'm just wondering here, not trying to defend Linux.) -- Kent -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: Linux/Windows Universal Benchmark
On Wed, Apr 14, 2004 at 10:23:48PM +0100, Andy Morris wrote: Now one thing I should have mentioned earlier is that the app was written in Java (compiled using Sun's jdk1.4.2 (not gentoo's blackdown since it's too buggy)) and we are therefore also testing the platform implementation of the VM (if you don't know about java basically when code runs there is a middleman between the code running and the OS (the Java Virtual Machine). It is possible that the jdk does not work as well on Linux but this is what I use and so do millions of others, and therefore it's can be a v good benchmark. I will shortly if i get some time repeat the tests in C++ to remove this factor (it interfaces directly with the OS since it's compiled into native binaries) and if any1 does care for the result then let me know. I think it's very likely that at least some of the Java implementations available for Linux suck performance-wise compared to Windows. Java's quite a poor development platform on Linux; it doesn't help that the Sun JDK is non-free so people generally can't hack on it, and the free JVMs have only started to receive attention relatively recently. I'm afraid I wouldn't regard a Java benchmark as a remotely fair assessment of any difference between Linux and XP in themselves, whether or not millions of people use it. I also don't think that counting is a very interesting benchmark really. :) -- Colin Watson [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: Linux/Windows Universal Benchmark
Just been browsing and came across this dicussion, and thought i might be able to give some input. I'm a software grad (well this summer) and use both linux(gentoo 2.6.xxx cant remember specifically) and xp pro sp1 as home os's (games, dev etc) and when building a fairly small application i felt that it was taking too long to build my code using linux (approx 70 secs). So I logged into my windows and it did it in under 40 secs, quite a difference. However, these two machines were of different spec xp:2500xp(1.833mhz) barton 512cache,1gig ram, 7200 sata hdd (high spec) linux:1900xp(1.6mhz) thoroughbred 256cache, 1gig ram, 5400 ide hdd (fairly low spec) I would accept a relatively small diff between the two machines (but a difference since one is more powerful) and since there were approx 150 source files the hard drive reads could have caused a difference. So what I had was a biased test result. My first thoughts would be to write a tiny application that would not use the harddrive and compare again, so i made a simple program to count from 0 to a parameter x number of times, test data was to count 0-9 100 times. XP box did it in 3mins 10 secs, linux 5 mins 4 secs ( i did this numerous times and results were always v similar). Again big difference, so it definetely wasn't the harddrive's access time etc. So the only way I was going to find out was to install Linux on my xp box and dual boot it (which I did). So XP and Gentoo Linux on the same box, and xp still beats it hands down (it only improved by around 30 secs), which really dissapointed me. It makes me want to use my xp box for more stuff than previously, which i had been trying not since i like using kde over the Windows desktop Now one thing I should have mentioned earlier is that the app was written in Java (compiled using Sun's jdk1.4.2 (not gentoo's blackdown since it's too buggy)) and we are therefore also testing the platform implementation of the VM (if you don't know about java basically when code runs there is a middleman between the code running and the OS (the Java Virtual Machine). It is possible that the jdk does not work as well on Linux but this is what I use and so do millions of others, and therefore it's can be a v good benchmark. I will shortly if i get some time repeat the tests in C++ to remove this factor (it interfaces directly with the OS since it's compiled into native binaries) and if any1 does care for the result then let me know. Now if you think my benchmark is totally unsafe/inaccurate as a means for benchmarking then feel free to state so, but compiling and running java code is something I do frequently, and Linux can't seem to do it as quickly as XP. Just to sum up my findings, from my testing XP seems to run simple sequential cpu tasks not much less than twice the speed of Linux. My afterthoughts about the two OS's is that if I ran multiple threads of my test (say 500) that my XP box would flake out and linux would handle it much better, perhaps when i go away for the wkend next i'll do that but right now iv got far too much work to do on them (final yr proj). Hope this helped any1, -andy _ It's fast, it's easy and it's free. Get MSN Messenger today! http://www.msn.co.uk/messenger -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: Linux/Windows Universal Benchmark
On Thu, 2004-04-15 at 00:13 +0100, Andy Morris wrote: ... i made a simple program to count from 0 to a parameter x number of times, test data was to count 0-9 100 times. XP box did it in 3mins 10 secs, linux 5 mins 4 secs ( i did this numerous times and results were always v similar). Well, now you know that if you want to do something (pointless) like that as quickly as possible, then you should use windows :-) -Mark -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux/Windows Universal Benchmark
Mark Roach wrote: On Thu, 2004-04-15 at 00:13 +0100, Andy Morris wrote: ... i made a simple program to count from 0 to a parameter x number of times, test data was to count 0-9 100 times. XP box did it in 3mins 10 secs, linux 5 mins 4 secs ( i did this numerous times and results were always v similar). Well, now you know that if you want to do something (pointless) like that as quickly as possible, then you should use windows :-) This certainly may have implications for computation intensive applications. I don't think it would overly surprise anyone if it were discovered that Sun's Linux JVM performs poorer than its Window's counterpart. I'm sure such tests have already been performed in detail elsewhere. However, I agree that the test described above by itself can't really tell us anything of interest. Perhaps you should perform the test with multiple data sets (iterations) to approximate the rate at which the performance gap will increase. dircha -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux/Windows Universal Benchmark
Andy Morris wrote: Just been browsing and came across this dicussion, and thought i might be able to give some input. I'm a software grad (well this summer) and use both linux(gentoo 2.6.xxx cant remember specifically) and xp pro sp1 as home os's (games, dev etc) and when building a fairly small application i felt that it was taking too long to build my code using linux (approx 70 secs). So I logged into my windows and it did it in under 40 secs, quite a difference. However, these two machines were of different spec xp:2500xp(1.833mhz) barton 512cache,1gig ram, 7200 sata hdd (high spec) linux:1900xp(1.6mhz) thoroughbred 256cache, 1gig ram, 5400 ide hdd (fairly low spec) I would accept a relatively small diff between the two machines (but a difference since one is more powerful) and since there were approx 150 source files the hard drive reads could have caused a difference. So what I had was a biased test result. I don't know much technically, but I do have mobile racks, and in the days when I did have XP running I did my own comparisons. Explorer in XP took over two minutes to come up with a particular URL, take out the rack, slip in another one, and Mozilla takes 13 and a half seconds, on the same URL. Drives were both Maxtor 5400s, same Athlon 1GHz box. There were other tests, but that is one that sticks in my mind. I remember Linux taking slightly longer to load, but in everything else, it seemed to Excell, if you'll pardon the pun. I have absolutely no idea what the figures are. XP sits in a green box on the shelf now. Regards, David. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux/Windows Universal Benchmark
This certainly may have implications for computation intensive applications. I don't think it would overly surprise anyone if it were discovered that Sun's Linux JVM performs poorer than its Window's counterpart. I'm sure such tests have already been performed in detail elsewhere. However, I agree that the test described above by itself can't really tell us anything of interest. Perhaps you should perform the test with multiple data sets (iterations) to approximate the rate at which the performance gap will increase. dircha I use Tomcat quite a bit on Debian Testing, and would love to see some comprehensive benchmarks. It would be somewhat ironic if Suns Java VM were demonstrably quicker on Windows that Linux (and Solaris?). I'm saying this as I suspect Java's purpose is/was to get developers writing platform agnostic software, allowing other OS's to enter the market. If there's a noticable qualitative difference favoring the monopolist though... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Linux/Windows Universal Benchmark
Hi, I want to benchmark my desktop system running in Linux (Gnome, XF86 v3.x, 2.4.5 kernel) against itself running Win98se. Is there a benchmark program that will work in both enviornments to give me an accurate benchmark? No, of course not. Different systems do different things well and poorly. For example: Write a benchmark that starts and stops 10,000 processes and Linux will beat Windows hands-down. Write a benchmark that starts and stops 10,000 threads and Windows will beat Linux hands-down (if it's not still running the process benchmark...). Yes, of course, if you know what you care about (cf previous example). If you care about something simple, like floating point arithmetic or quake frames per second, it won't matter what operating system you run, because it's all done in the hardware. If you care about something complex, like a writing to a disk or using TCP sockets or building up a tearing down processes, you will get wildly varying answers depending on which question you ask. Only zealots and computer magazine publishers believe in the benchmark that will identify the One True Best Computer.
Re: Linux/Windows Universal Benchmark
I really don't need the preaching. I wanted a straight answer. It's a perfectly legitimate question to want to know if one thing performs better than another, and if so, in what areas, and by how much. Human senses are inaccurate to judge things such as this at times, especially if differences are minute, and many factors affect it. A benchmark is more precise. There IS a benchmark cross-platform between Mac and PC. So why not Windows and Linux? That was not a very helpful answer at all. It's as bad as RTFM. -- Deven Gallo In a message dated 12/1/01 1:22:47 AM Eastern Standard Time, [EMAIL PROTECTED] writes: Only zealots and computer magazine publishers believe in the benchmark that will identify the One True Best Computer.
Re: Linux/Windows Universal Benchmark
[EMAIL PROTECTED] wrote: ... That was not a very helpful answer at all. It's as bad as RTFM. RTFM is good answer, even more so when it includes info on which FM to R. sorry, I don't know any comprehensive benchmarks... depending on what you want to test quake might be useful. erik
Re: Linux/Windows Universal Benchmark
On Sat, Dec 01, 2001 at 01:41:33AM -0500, [EMAIL PROTECTED] wrote: I really don't need the preaching. I wanted a straight answer. It's a perfectly legitimate question to want to know if one thing performs better than another, and if so, in what areas, and by how much. Human senses are inaccurate to judge things such as this at times, especially if differences are minute, and many factors affect it. A benchmark is more precise. There IS a benchmark cross-platform between Mac and PC. So why not Windows and Linux? That was not a very helpful answer at all. It's as bad as RTFM. This question is off-topic. STFW or try to find a newsgroup/mailing list that discusses benchmarking. -- Nathan Norman - Staff Engineer | A good plan today is better Micromuse Ltd. | than a perfect plan tomorrow. mailto:[EMAIL PROTECTED] | -- Patton pgpcspim7gSvV.pgp Description: PGP signature
Re: Linux/Windows Universal Benchmark
I really don't need the preaching. I wanted a straight answer. Au contraire. You got a rather information-packed answer that is as straight as possible, dosed up with a bit of good-natured joshing. You now know: Process-cyclingUnix Threads manipulation Windows FP Arithmetic Depends on CPU, not OS Graphics Display Depends on graphics card, not OS And you know that to gauge other subsystems you will have to write code and see. Relax a little, would ya?! The fact that you get a cheeky answer doesn't mean you're not welcome.
Re: Linux/Windows Universal Benchmark
Jake Catfox asked ~
a benchmark cross-platform between Mac and PC. So why not Windows and Linux?
~ some say, that, the [EMAIL PROTECTED] work-unit is a great
cross-platform measure.
The SETI web pages have some interesting Stats on CPUs/ Operating Systems
~ there are Links to {if I recall} high_powered German analytical site on
cross-platform Stats.
http://setiathome.ssl.berkeley.edu/
--
If you are not the intended recipient of this e-mail please alert
someone in authority and destroy it. Reading unauthorised e-mail
during working hours is an offence. Do not condone crime in the office -
just say no. This e-mail indemnifies the sender from all
provision of festive drinks, gratuities, seasonal goodwill or other non
tax deductable expenses.
Re: Linux/Windows Universal Benchmark
quote who= Hi, I want to benchmark my desktop system running in Linux (Gnome, XF86 v3.x, 2.4.5 kernel) against itself running Win98se. Is there a benchmark program that will work in both enviornments to give me an accurate benchmark? not really. the 2 systems are so totally different it will vary very widely. you can test specific areas of performance like cpu, memory access, filesystem access. but i do not believe there is any broad range benchmark that can treat both platforms fairly. libraries, compilers, level of optimization in the code, quality of the drivers, among other things can totally screw benchmark results. even compiling the same program(like [EMAIL PROTECTED] as another poster mentioned) for both platforms can still vary widely with results. im sure even compiling the kernel for an older arch(eg 486) and running [EMAIL PROTECTED] will affect seti's performance on a p3. so no..no fair way to benchmark the 2 systems in the mannor you seem to be wanting. bout as close as you can get is use 1 system for a week, then use the other system for a week and see which feels faster. nate
Re: Linux/Windows Universal Benchmark
quote who= I really don't need the preaching. I wanted a straight answer. It's a perfectly legitimate question to want to know if one thing performs better than another, and if so, in what areas, and by how much. Human senses are inaccurate to judge things such as this at times, especially if differences are minute, and many factors affect it. A benchmark is more precise. There IS a benchmark cross-platform between Mac and PC. So why not Windows and Linux? That was not a very helpful answer at all. It's as bad as RTFM. what is the benchmark between mac and pc ? the only one i could possibly think of is comparing photoshop results which to me is a worthless benchmark(for my needs anyways). i don't read magazines anymore so maybe they have come up with something else..the ziff davis benchmark series was a joke too. i remember reading about vendors who would optimize drivers just so it would run those benchmarks faster. i don't use gnome or kde so i dont know how much of a performance hit you take with them. but on my systems the only thing i need high performance for is 3D(unreal tournament), and video encoding(Divx 4). for everday apps today's cpus are more then powerful enough(or should be) on any OS. i honestly noticed hardly any difference between going from p3-800 512MB to athlon 1.3G 768MB ram. (even in unreal tournament). and i don't believe that changing an OS could affect overall performance by that much(increasing ram and cpu speed as above). making it a non issue for me and most others im sure. i upgraded to a athlon 1.3g soley to do divx 4 video capture with the hopes i could get more frames out of it then my p3 but i was wrong ..for everyday apps i can't tell a diff between my P3-700 256MB laptop, my P3-733 512MB(ultra 160 scsi) desktop at work and my 1.3g athlon 768mb at home(performance wise). nate (my windowmanager of choice is afterstep with 32 virtual desktops on my 1600x1200 screen) only 3 more days and my desktop at works hits the 200day uptime milestone! woohoo!#! i love debian.
Re: Linux/Windows Universal Benchmark
Quoth [EMAIL PROTECTED], Hi, I want to benchmark my desktop system running in Linux (Gnome, XF86 v3.x, 2.4.5 kernel) against itself running Win98se. Is there a benchmark program that will work in both enviornments to give me an accurate benchmark? Quake 3 Arena. Of course, it might not tell you everything you're interested in, but all of the good hardware review site use it. cheers, damon -- Damon Muller :: Department of Criminology :: University of Melbourne Homicide is, no matter what else it might be, a social relationship. -- Paul Bonnana
Re: Linux/Windows Universal Benchmark
No, of course not. Different systems do different things well and poorly. For example: Write a benchmark that starts and stops 10,000 processes and Linux will beat Windows hands-down. Write a benchmark that starts and stops 10,000 threads and Windows will beat Linux hands-down (if it's not still running the process benchmark...). U out of curiosity, Why / how does windows beat Linux ? Is it technically very difficult / impossible to have an OS that does processes and threads very fast or has Linux CHOSEN to give more importance to processes than to threads and why ? Thanx Shri
Re: Linux/Windows Universal Benchmark
On Sat, Dec 01, 2001 at 02:38:51PM +, Shri Shrikumar wrote:
| No, of course not. Different systems do different things well and
| poorly. For example: Write a benchmark that starts and stops 10,000
| processes and Linux will beat Windows hands-down. Write a benchmark
| that starts and stops 10,000 threads and Windows will beat Linux
| hands-down (if it's not still running the process benchmark...).
|
| U out of curiosity, Why / how does windows beat Linux ? Is it
| technically very difficult / impossible to have an OS that does
| processes and threads very fast or has Linux CHOSEN to give more
| importance to processes than to threads and why ?
In linux native threads (as opposed to green or user threads which
have nothing to do with the kernel at all) are mapped onto processes.
Each thread has its own PID and if you run top or gtop you will see
each one listed separately (even though they all share the same
address space). For an example run xmms or galeon and look at top. I
imagine that one reason linux maps threads on to processes is to
simplify the scheduler -- all those entities appear the same so
threads don't need any special handling.
Threads are actually a controversial entity. I do quite a bit of java
programming, and threads are quite convenient, and even necessary for
maintaining responsiveness in a GUI. However threads are the cause of
more bugs than anything else. Threads are convenient because all
threads share the same address space, thus they can communicate very
simply by passing an object reference. The tradeoff is that the
programmer must worry about synchronization and timing. Periodically
the question arises in newsgroups if I am in threada and I want to
stop (kill) threadb, how do I do it?. The answer is *you can't*.
The only way is to have a shared flag, and have threadb check it
periodically and terminate itself if the flag is set. To explain this
I switch to processes for a moment :
With processes, each process has its own address space and its own set
of resources. As you are surely aware you can kill any process ('kill
-KILL') without affecting other processes. (slightly simplified, if
they are communicating via a socket, etc, there will be side effects)
This is because of the separate address space, etc. When you kill a
process the kernel takes care of releasing the memory, closing files
and sockets, and releasing other resources.
With threads, though, they share the same resources thus there is no
way for the kernel (or any other program) to reliably clean up when
you want a thread killed.
Some programmers believe that threads should not be used at all, but
only processes that communicate via some sort of IPC. Those
programmers also believe that if your OS has processes that aren't
lightweight enough for that to be feasible, then it is a bug in your
OS; threads are not the right band-aid for the problem.
I'm not totally decided on the matter. I find threads to be
convenient, though I am certainly aware of the headaches they can
cause. I haven't done any IPC programing so I have no experience
there to compare the merits.
-D
--
Failure is not an option. It is bundled with the software.
Re: Linux/Windows Universal Benchmark
On Sat, Dec 01, 2001 at 12:52:52AM -0500, [EMAIL PROTECTED] wrote: | Is there a benchmark program [...] to give me an accurate benchmark? The short answer is No. You can measure certain things, that you are interested in, but there is always the problem (as someone mentioned) that once the benchmark is known the developers optimize for the benchmark instead of real life. The other issue is the meaning of any given measurement. Why do you want to measure it? What is your goal? For something as broad as an OS, just use it. If you can't see the difference, then it doesn't matter. -D -- (E)scape (M)eta (A)lt (C)ontrol (S)hift
Re: Linux/Windows Universal Benchmark
On Sat, Dec 01, 2001 at 11:52:47AM -0500, dman wrote: Some programmers believe that threads should not be used at all, but only processes that communicate via some sort of IPC. Those programmers also believe that if your OS has processes that aren't lightweight enough for that to be feasible, then it is a bug in your OS; threads are not the right band-aid for the problem. If I'm out for performance, I tend to go for neither if I can get away with it. Alan Cox once said: A computer is a state machine. Threads are for people who can't program state machines. In my day job, I've seen just how stonkingly fast it's possible for code to go when it operates in a single process and uses non-blocking I/O. It eliminates all the context-switching rubbish and gives as much time as possible to the application, and the system stays responsive even under massive load. If you're processing lots of I/O streams, neither multiple processes nor multiple threads really scale far enough. There are downsides, of course - you have to have good libraries and a very disciplined programming style. It's certainly not the easiest model to use, and you still have to have good code in general. We're in the world of benchmarking here, though. -- Colin Watson [EMAIL PROTECTED]
Re: Linux/Windows Universal Benchmark
* [EMAIL PROTECTED] ([EMAIL PROTECTED]) spake thusly: I really don't need the preaching. I wanted a straight answer. Take a bench, a marker, and use the marker put a mark on the bench. That's a benchmark. Straight enough for you? Dima -- Q276304 - Error Message: Your Password Must Be at Least 18770 Characters and Cannot Repeat Any of Your Previous 30689 Passwords -- RISKS 21.37
Re: Linux/Windows Universal Benchmark
At 2001-12-01T05:52:52Z, [EMAIL PROTECTED] writes: Hi, I want to benchmark my desktop system running in Linux (Gnome, XF86 v3.x, 2.4.5 kernel) against itself running Win98se. Is there a benchmark program that will work in both enviornments to give me an accurate benchmark? No offense, but that's one of the more vague questions I've heard in some time. What exactly are you wanting to benchmark? File I/O? 2d graphics? OpenGL? Sound? With how many other programs running, and what kind? Most people agree that the *only* valid benchmark of a system is how well it runs the applications that *you* want to run. Everything else is rather pointless. -- Kirk Strauser
Re: Linux/Windows Universal Benchmark
* Shri Shrikumar ([EMAIL PROTECTED]) spake thusly: No, of course not. Different systems do different things well and poorly. For example: Write a benchmark that starts and stops 10,000 processes and Linux will beat Windows hands-down. Write a benchmark that starts and stops 10,000 threads and Windows will beat Linux hands-down (if it's not still running the process benchmark...). U out of curiosity, Why / how does windows beat Linux ? Is it technically very difficult / impossible to have an OS that does processes and threads very fast or has Linux CHOSEN to give more importance to processes than to threads and why ? Because... errm, do a university-level OS course. Unix is a time-sharing system, which means it is geared up for running multiple processes, and give each of these processes decent interactive performance. This way a university can give each student a terminal running vi, and none of them has to wait 10 minutes for their keystrokes to get processed. So in a sense yes, Linux chose to be that way. Or, rather, Berkeley chose to make Unix that way. Oh, and threads didn't exist back then IIRC. Dima -- Q276304 - Error Message: Your Password Must Be at Least 18770 Characters and Cannot Repeat Any of Your Previous 30689 Passwords -- RISKS 21.37
Re: Linux/Windows Universal Benchmark
On Sat, Dec 01, 2001 at 02:53:15PM -0600, Dimitri Maziuk wrote: | * [EMAIL PROTECTED] ([EMAIL PROTECTED]) spake thusly: | I really don't need the preaching. I wanted a straight answer. | | Take a bench, a marker, and use the marker put a mark on the | bench. That's a benchmark. Hehe. This gave me a laugh! -D -- A)bort, R)etry, D)o it right this time
