That's the exact opposite end of the spectrum from embarrassingly parallel problems. In embarrassingly parallel problems you have so little data dependency that tasks can run independently or nearly independently. In you friend's case, the tasks were so interdependent that all the tasks were waiting on one task to finish, so there was nearly no speedup from adding more processors.
The bottom line is that the best solution to some problems is a grid of loosely connected computers. The best solution to others is one very fast processor. In between, there is SMP, where there are many processors in one computer, and clusters, where the processors are connected through a high-speed network. There is no one best solution, as implied by the original poster. The best solution depends on the problem to be solved.
Distributed processing is fine for "embarrassingly parallel" problems where the compute nodes don't need to communicate with each other. However, many problems solved by supercomputers or large clusters need communication between the compute nodes, so aren't amenable to distributed solutions.
First of all, there are some jerks who happen to have Bugzilla accounts. I have run into several in the past year. None are Mozilla employees. I'm certain you can find links to Slashdot comments where people say they've been treated with disrespect in Bugzilla, but that doesn't mean that Mozilla developers are abusive.
Second, simply presenting a list of bugs in one browser in itself doesn't tell you whether it's stable or not. All browsers have crash bugs, memory leak bugs, and security bugs. All you're showing is that Firefox is no exception.
Third, Mozilla developers tend to focus on bug reports that have lots of detailed information. Most bugs, especially serious ones, are reported multiple times. It's simply more efficient for them to focus on the ones with lots of details. It gets the bugs fixed more quickly. Maybe you think that's giving excuses for not fixing bugs, but that is how it appears to me.
I've followed your links, and seen your comments in bug reports. I don't see where "the Mozilla team becomes abusive." I see some of your bug reports (bug 204668 and bug 222660 for example) where testers and developers (some of them Mozilla employees and others volunteers) asked you for further details so they could track down bugs. Rather than provide specific information to help track down the bugs, you instead started ranting on Slashdot.
Why would Mozilla have responsibility for bugs in extensions written by a third party? That would be like me blaming Microsoft for any buggy Windows applications I run, because they provided the OS that the applications run on. Bugs in programs and extensions are the responsibility of the developers of those programs and extensions.
If you see a quirk in Firefox, simply write up a bug report specifying in enough detail what the problem is, and it will be fixed. Whining about them on Slashdot is about the least effective thing you can do.
I just printed the two-page, multicolor Minefield Start Page on my HP inkjet, and both Mem Usage and VM Size for firefox.exe stayed under 50 MB the entire time, even with the Update Speed set to High to catch any momentary surges in memory usage. If there's a problem with memory usage during printing, it seems to either be fixed in nightly builds, or occurs with only certain inkjet drivers.
Yet another ignorant Firefox memory post. First off, you've got some basic facts wrong. Firefox never stores "rendered pages as images". That's a conflagration of two different caches. Second, you seem unwilling to recognize that the caches actually do make Firefox measurably faster.
One cache is the memory cache, which stores images uncompressed (decoded) for fast access. It stores just images, not "rendered pages as images." You can change the amount of memory used for this cache by setting browser.cache.memory.capacity. Recently a bug in nightly builds of Firefox 3 was fixed that caused only half of the memory cache to be used to store images, and fixing that bug resulted in a 4-6% performance gain. Turning off the memory cache entirely would cause performance to suffer much more. You would probably notice the slowdown, but probably not notice that Firefox uses about 20 MB less, unless you were on the verge of running out of memory.
The other cache is the back-forward cache, which stores the DOM information of recently visited pages so they don't need to be reparsed when visiting them again. You change the amount of memory used for this cache by changing browser.sessionhistory.max_total_viewers. Again, turning this cache off will result in noticeable delays when going back and forwards up to a few pages, but probably not a noticeable difference in the memory usage.
Even with both caches at their default settings, Firefox generally uses less memory than other browsers. There's really no need to mess around with the settings unless your computer has very little RAM to begin with. In that case, you should read http://kb.mozillazine.org/Reducing_memory_usage_(F irefox)
Yeah, there's not a whole lot you can do to reduce the memory usage of a browser, when you asking it to display lots of complex pages at the same time. There are some things you can do, such as not use extensions that have serious memory leaks, and use a minimal set of plugins: http://kb.mozillazine.org/Reducing_memory_usage_(F irefox)
I agree. openMosix with its fork-and-forget SSI model works well for certain tasks, but for those tasks a computer with lots of cores works better. First of all, the operating system automatically migrates processes to other cores, and second, even individual threads can migrate without experiencing horrible shared memory latency. With 8-core computers on the market and 16-core computers coming soon, there's just not as much need for openMosix.
Most clustering applications benefit some form of communication such as MPI or at least a scheduler such as PBS or SGE, so openMosix's fork-and-forget model doesn't work well. Those applications can use Rocks Cluster Distribution, the real mainstay of open-source clustering.
That last one's critical.
To add to your point, evolution has even been reproduced with simulated "organisms" on computers rather than with living organisms with remarkable results. It not only makes sense, it works when we try it ourselves. It's the furthest thing from faith I can think of.
Possibly. But remember that Percival Lowell thought Martians built canals and astrophysicists at first thought pulsars were signals from little green men. However, simpler explanations for the observed phenomena were found. I'm sure when we find actual evidence of aliens, we'll recognize it as such and not try to explain it as a natural phenomenon. There's a tendency to explain what we don't understand by an appeal to some sort of intelligence anyway.
There has been no "evolution" from creationism to intelligent design. It was merely a name change. When creationism was ruled unscientific, they changed the name to make it sound more "scientific". Even a judge could clearly see the ruse. A study of the text in Of Pandas and People showed where the authors did a simple search-and-replace of one term for the other (as explained on the Wikipedia article I linked to).
I haven't seen anyone discuss this problem at MozillaZine, so I'll respond here. I can test only on Windows XP. Starting Firefox 2.0.0.4 and opening cnn.com I see Mem Usage of 42 MB. Doing exactly the same thing in Safari for Windows I see Mem Usage of 52 MB. As far as I can see, Safari uses more memory than Firefox.
If someone still thinks they see any problem with Firefox, please report it in Bugzilla or discuss it on MozillaZine. If you do nothing to report a bug, do not act shocked that it does not get fixed. Discussing the problem on Slashdot or otherwise making vague comments such as "fix the memory leak already" doesn't count, as you need to give specific details of the problem to the appropriate place.
The poster of the article already stated he was taking his business elsewhere in the summary. So a post telling him to take his business elsewhere really is redundant!
If Opera isn't mentioned, it's not because of any kind of conspiracy against proprietary software. The truth is, it's simply not very popular. It's had less than 1% usage for years. At best, it's the #4 browser with a fraction of Safari's usage. If Opera's such a superior browser, why don't people use it more?
Just one thing remains to be asked: Why the heck would anybody use Firefox if there are better alternatives everywhere?
Exactly. So many people use Firefox simply because the alternatives are not better. Build a better browser, and users will come use it. Now it's time for Safari and Opera to improve their products if they want more users.
doesn't it seem kind of strange to be saying that you are happy that apple is releasing the broswer on the one hand and complaining that they might be trying to take out your market share on the other hand?
He's only complaining that it's Apple's apparent intention to completely wipe out all competition other than Internet Explorer. That means fewer browsers to choose from, and therefore a less open web. That's not good for the web. Having more competition in the form of more browsers to choose from, including Safari, Firefox, and Opera, is good for the web. Get it now?
Please post both problems on the Firefox Bugs forum at MozillaZine. I cannot test because I don't have access to a Mac, nor can I access your custom app. You'll need to write up a testcase for others to see that page loading problem for someone to write up a bug report about it. Thanks for helping!
When I try doing the same thing in Firefox as in other browsers, Firefox usually ends up using less memory than other browsers. Safari for Windows usually ends up using more memory than other browsers. I'm not the only one who sees Safari using more memory than Firefox.
Is there a way we could all see Firefox having a large memory footprint compared to other browsers? Then someone could write a bug report and get the problem fixed. If you have any hints on how to see the problem, such as a site that when loaded causes Firefox to use much more memory than another browser, please don't keep them such a secret!
Slashdot Mirror
Idiocracy, here we come!
That's the exact opposite end of the spectrum from embarrassingly parallel problems. In embarrassingly parallel problems you have so little data dependency that tasks can run independently or nearly independently. In you friend's case, the tasks were so interdependent that all the tasks were waiting on one task to finish, so there was nearly no speedup from adding more processors.
The bottom line is that the best solution to some problems is a grid of loosely connected computers. The best solution to others is one very fast processor. In between, there is SMP, where there are many processors in one computer, and clusters, where the processors are connected through a high-speed network. There is no one best solution, as implied by the original poster. The best solution depends on the problem to be solved.
Distributed processing is fine for "embarrassingly parallel" problems where the compute nodes don't need to communicate with each other. However, many problems solved by supercomputers or large clusters need communication between the compute nodes, so aren't amenable to distributed solutions.
First of all, there are some jerks who happen to have Bugzilla accounts. I have run into several in the past year. None are Mozilla employees. I'm certain you can find links to Slashdot comments where people say they've been treated with disrespect in Bugzilla, but that doesn't mean that Mozilla developers are abusive.
Second, simply presenting a list of bugs in one browser in itself doesn't tell you whether it's stable or not. All browsers have crash bugs, memory leak bugs, and security bugs. All you're showing is that Firefox is no exception.
Third, Mozilla developers tend to focus on bug reports that have lots of detailed information. Most bugs, especially serious ones, are reported multiple times. It's simply more efficient for them to focus on the ones with lots of details. It gets the bugs fixed more quickly. Maybe you think that's giving excuses for not fixing bugs, but that is how it appears to me.
Or maybe browsers are simply large, complicated programs that will always have thousands of bugs.
I've followed your links, and seen your comments in bug reports. I don't see where "the Mozilla team becomes abusive." I see some of your bug reports (bug 204668 and bug 222660 for example) where testers and developers (some of them Mozilla employees and others volunteers) asked you for further details so they could track down bugs. Rather than provide specific information to help track down the bugs, you instead started ranting on Slashdot.
Why would Mozilla have responsibility for bugs in extensions written by a third party? That would be like me blaming Microsoft for any buggy Windows applications I run, because they provided the OS that the applications run on. Bugs in programs and extensions are the responsibility of the developers of those programs and extensions.
You mean the one where If you open a lot of windows and tabs in Firefox on a laptop, and put the laptop in and out of standby, you will eventually notice that the laptop fan is running all the time, even when there is no activity. That's the CPU bug, and it can potentially shorten the life of your laptop? It looks like it's fixed. As for a "memory gobbling bug", you'll have to describe in much more detail what you mean. Firefox seems to use less memory than other browsers, and in addition, about 100 memory leak bugs have been fixed in the past year.
If you see a quirk in Firefox, simply write up a bug report specifying in enough detail what the problem is, and it will be fixed. Whining about them on Slashdot is about the least effective thing you can do.
I just printed the two-page, multicolor Minefield Start Page on my HP inkjet, and both Mem Usage and VM Size for firefox.exe stayed under 50 MB the entire time, even with the Update Speed set to High to catch any momentary surges in memory usage. If there's a problem with memory usage during printing, it seems to either be fixed in nightly builds, or occurs with only certain inkjet drivers.
Yet another ignorant Firefox memory post. First off, you've got some basic facts wrong. Firefox never stores "rendered pages as images". That's a conflagration of two different caches. Second, you seem unwilling to recognize that the caches actually do make Firefox measurably faster.
One cache is the memory cache, which stores images uncompressed (decoded) for fast access. It stores just images, not "rendered pages as images." You can change the amount of memory used for this cache by setting browser.cache.memory.capacity. Recently a bug in nightly builds of Firefox 3 was fixed that caused only half of the memory cache to be used to store images, and fixing that bug resulted in a 4-6% performance gain. Turning off the memory cache entirely would cause performance to suffer much more. You would probably notice the slowdown, but probably not notice that Firefox uses about 20 MB less, unless you were on the verge of running out of memory.
The other cache is the back-forward cache, which stores the DOM information of recently visited pages so they don't need to be reparsed when visiting them again. You change the amount of memory used for this cache by changing browser.sessionhistory.max_total_viewers. Again, turning this cache off will result in noticeable delays when going back and forwards up to a few pages, but probably not a noticeable difference in the memory usage.
Even with both caches at their default settings, Firefox generally uses less memory than other browsers. There's really no need to mess around with the settings unless your computer has very little RAM to begin with. In that case, you should read http://kb.mozillazine.org/Reducing_memory_usage_(F irefox)
Yeah, there's not a whole lot you can do to reduce the memory usage of a browser, when you asking it to display lots of complex pages at the same time. There are some things you can do, such as not use extensions that have serious memory leaks, and use a minimal set of plugins: http://kb.mozillazine.org/Reducing_memory_usage_(F irefox)
I agree. openMosix with its fork-and-forget SSI model works well for certain tasks, but for those tasks a computer with lots of cores works better. First of all, the operating system automatically migrates processes to other cores, and second, even individual threads can migrate without experiencing horrible shared memory latency. With 8-core computers on the market and 16-core computers coming soon, there's just not as much need for openMosix.
Most clustering applications benefit some form of communication such as MPI or at least a scheduler such as PBS or SGE, so openMosix's fork-and-forget model doesn't work well. Those applications can use Rocks Cluster Distribution, the real mainstay of open-source clustering.
Firefox crashes for you? Read the MozillaZine Knowledge Base article about Firefox crashes and you can probably fix your problem.
4. Random mutations occur.
That last one's critical. To add to your point, evolution has even been reproduced with simulated "organisms" on computers rather than with living organisms with remarkable results. It not only makes sense, it works when we try it ourselves. It's the furthest thing from faith I can think of.
Possibly. But remember that Percival Lowell thought Martians built canals and astrophysicists at first thought pulsars were signals from little green men. However, simpler explanations for the observed phenomena were found. I'm sure when we find actual evidence of aliens, we'll recognize it as such and not try to explain it as a natural phenomenon. There's a tendency to explain what we don't understand by an appeal to some sort of intelligence anyway.
There has been no "evolution" from creationism to intelligent design. It was merely a name change. When creationism was ruled unscientific, they changed the name to make it sound more "scientific". Even a judge could clearly see the ruse. A study of the text in Of Pandas and People showed where the authors did a simple search-and-replace of one term for the other (as explained on the Wikipedia article I linked to).
I haven't seen anyone discuss this problem at MozillaZine, so I'll respond here. I can test only on Windows XP. Starting Firefox 2.0.0.4 and opening cnn.com I see Mem Usage of 42 MB. Doing exactly the same thing in Safari for Windows I see Mem Usage of 52 MB. As far as I can see, Safari uses more memory than Firefox.
If someone still thinks they see any problem with Firefox, please report it in Bugzilla or discuss it on MozillaZine. If you do nothing to report a bug, do not act shocked that it does not get fixed. Discussing the problem on Slashdot or otherwise making vague comments such as "fix the memory leak already" doesn't count, as you need to give specific details of the problem to the appropriate place.
The poster of the article already stated he was taking his business elsewhere in the summary. So a post telling him to take his business elsewhere really is redundant!
If Opera isn't mentioned, it's not because of any kind of conspiracy against proprietary software. The truth is, it's simply not very popular. It's had less than 1% usage for years. At best, it's the #4 browser with a fraction of Safari's usage. If Opera's such a superior browser, why don't people use it more?
Please post both problems on the Firefox Bugs forum at MozillaZine. I cannot test because I don't have access to a Mac, nor can I access your custom app. You'll need to write up a testcase for others to see that page loading problem for someone to write up a bug report about it. Thanks for helping!
When I try doing the same thing in Firefox as in other browsers, Firefox usually ends up using less memory than other browsers. Safari for Windows usually ends up using more memory than other browsers. I'm not the only one who sees Safari using more memory than Firefox.
Is there a way we could all see Firefox having a large memory footprint compared to other browsers? Then someone could write a bug report and get the problem fixed. If you have any hints on how to see the problem, such as a site that when loaded causes Firefox to use much more memory than another browser, please don't keep them such a secret!