Debating the Linux Process Scheduler

An anonymous reader writes "The Linux 2.6.23 kernel is expected around the end of the month, and will be the first to include Ingo Molnar's much debated rewrite of the process scheduler called the Completely Fair Scheduler. In another Linux kernel mailing list thread one more developer is complaining about Molnar and his new code. However, according to KernelTrap a number of other Linux developers have stood up to defend Molnar and call into question the motives of the complaints. It will be interesting to see how the new processor really performs when the 2.6.23 kernel is released."

Comments

Can someone provide some insight?

[Opportunist]

Is someone who does understand the differences able to explain, in non-kernel-developer terms, what the big differences will be for the average user, developer or administrator? I mean, I'd love to discuss it, but first of all I'd want to know what we're discussing.

Re:Can someone provide some insight?

[Jeff DeMaagd]

I'd love to discuss it, but first of all I'd want to know what we're discussing.

That's not the Slashdot way. We're supposed to have an unfounded opinion based on insufficient facts and preexisting prejudices.

Re:Can someone provide some insight?

[Aladrin]

Average? Probably nothing. But for devs/admins that are worried about certain processes taking more time than others, it -should- be more fair and keep things running smoother.

It's possible for programs right now to exploit how the current schedule dishes out time. As far as I know, they currently only do so out of ignorance, rather than malice. The new scheduler just corrects the problem.

It's not something a user can really see unless they know exactly what they are looking for, and unless a dev/admin has a program that's behaving unfairly, it's not really going to matter to them, either.

There is another invisible effect as well... Kolivas apparently publicly announced his decision to stop working on the kernel, which would include the current scheduler. That means finding another maintainer for his code, should any problems surface. If you've got 2 pieces of code that test the same in speed (as they do according to some), and 1 has a dev that's willing to keep working on it, and the other doesn't... Which would you pick?

The new code also has the added advantage of being a really really neat idea, which encourages people to work on it as well.

Re:Can someone provide some insight?

[Zephiris]

Essentially, the difference is how well processor resources are divided up, how evenly, and how big the pieces are each process or task gets. Most anyone who has used Linux has had the dreaded moment where you're trying to multitask a bit, and are compiling a program while listening to music, or waching a video, and then...that's terrific, video frames are dropped, or the audio skips. Even if intermitant, it's quite annoying, at the very least. The 'Completely Fair Scheduler' is an attempt to have more fair, sane, and generally less complex scheduling. This also happens to reduces the worst case latencies, averaging from (at least on the tests with my computer) 120+ms on vanilla 2.6.22 scheduler, to ~2.6ms with CFS.

It's largely a drastic improvement over the old scheduling mechanisms that Linux has relied on, although other OSes have largely worked through such problems some time ago.

While it's not exactly THE most scientific, I had a few rounds of testing over which did better on load vs. things still behaving exactly the way they should. I ran all of them with audio playing through KDE artsd, video player, glxgears, etc, loaded, plus inducing a CPU load via 'stress'. Linux, even with CFS, it's still fairly easy to 'upset' it by just producing a fairly large (2-4) amount of load. Solaris did notably better. While it seemed to have a few quirks with scheduling in general, it could sustain a load of around 8-12 without producing video/audio frame drops. FreeBSD, with the experimental SCHED_ULE 2.0 scheduler (as of March 2007) could sustain a load of over 80 with no problems, frame drops, or even glxgears slowing down to a complete crawl (although you wouldn't want to especially use OpenGL at such, it was still getting the speed of software glxgears), and even at 120+ load, the mouse wouldn't respond, while everything else kept going fine. This seems purely useless, but it really comes in handy if trying to do one or more KDE compiles while watching video, on Linux, this tends to be prevented. For the uninitiated, load averages like that are basically a multiplier vs. how much actual work your computer can do in real time. Eg, a 0.5 load would mean you're doing 50% of what you could in realtime. A 2.0 load means you're trying to handle twice what you can do in realtime, it is weighted against how many processors you have (I have one), but other things like disk access can also contribute to the load average, depending on OS.

So, longer story short, a superior CPU scheduler can make a world of difference in how things behave when your system's something else with the CPU(s) at the same time.

Re:Can someone provide some insight?

[Opportunist]

That sounds sensible. With the increase of Linux boxes run by "average" people (who will not directly notice a difference), the threat of malware for Linux is going to be on the rise, too. And those people usually know how to exploit even the smallest flaws in a system.

Re:Can someone provide some insight?

[dpilot]

Do you want your media to play without skips or drops while you're compiling your new kernel?

There have long been tricks like "interactive priority boost" or "nice -10 X" that attempt to make the desktop more responsive, and media play smoothly. But others believe those are just tricks, bound to misbehave in corner cases, and that a good scheduler and well implemented priority scheme will do just as well without the drawbacks. That's where CFS is trying to be. In particular most desktop responsiveness is of the sort, "I need a little CPU, and I need it NOW!" while compiles and such are "I need lots of CPU, and I'll take it whenever I can get it." The CFS keeps track of not just who's using a timeslice, but how much time they're using. That way, those short bursts of CPU keep their priority intact, while more CPU-intensive processes tend to get some priority degradation.

This goes back a little farther than Ingo Molnar's current involvement. A while back, Con Kolivas began putting in a bunch of work on the scheduler trying to get desktop response to work right, essentially he wanted his media, and his compiles, too. He did a lot of work and attracted a lot of users and fanbois along the way. More recently, Ingo Molnar get interested too, and came up with the "Completely Fair Scheduler." When it came time to pick one, Linus saw the CFS doing pretty well, still under heavy and active development. CK's scheduler was also pretty good, but the fanbois poisoned the waters, insisting that it was perfect as it was, and didn't need fixing. Linux chose an active development model over "perfection." Unfortunately Con Kolivas felt slighted in the process, and left. IIRC, he may have been absent during the decision window, and his fanbois did him in.

Add to all of this the fact that the kernel can now run tickless, so that laptops can really scale back their power in between keystrokes or while you're reading the screen. There has been quite a bit of interesting work on scheduling, lately.

Re:Can someone provide some insight?

[ciroknight]

Kolivas apparently publicly announced his decision to stop working on the kernel, which would include the current scheduler. That means finding another maintainer for his code, should any problems surface. If you've got 2 pieces of code that test the same in speed (as they do according to some), and 1 has a dev that's willing to keep working on it, and the other doesn't... Which would you pick?

Wow, not even a full year has past and we're already getting revisionist historians trying to change the situation.

Kolivas quit because of the scheduler debacle, because nobody would listen to Kolivas but were apt to follow Linus and his cronie Ingo around when they drum up more-or-less the exact same thing. Instead of critically listening to Kolivas' points, Linus and Ingo attacked Kolivas' merits. Under that kind of personal attack, I couldn't say I wouldn't have quit just to shut them up. Not all of us are stubborn mules and jackasses.

Re:Can someone provide some insight?

[mamer]

Another question: I've been waiting for OpenMP support in gcc, it seems to be coming soon. In the meantime, I tried to parallelize my code by hard coding my threads using pthreads. I tested it by running several matrix-matrix multiplications "in parallel" on a multi-core CPU, and what I've got was all threads running on the same processor. Only after increasing my number of threads to a couple of dozens, I get some of them to run on the second processor. So basically, I am not getting any performance gain. I asked a number of people an they tell me "this is an old problem, basically the Linux kernel scheduler is stupid and nobody has bothered to fix it". Now, is Ingo's new scheduler fixing this? if gcc-openmp relies on the kernel scheduler, should we expect that open-mp will basically work-but-not-really-work on shared memory multi processor machines? I think this is an important issue to address, especially in an era where high-performace computing has become the driving force behind the hardware. BTW, how are otehr commercial compilers overcoming this scheduling problem?

Re:Can someone provide some insight?

[TheRaven64]

Well damn. That's just... embarassing It's more embarrassing than the grandparent states, actually. ULE 2 has been improved a lot in the last few months, and the newer ULE 3 performs a lot better, particularly on SMP systems. From what I've seen, the new Linux scheduler has roughly the same shape (and size) performance curves on 1-8 CPU systems as the old 4BSD scheduler that ULE replaces.

Why's everyone using linux if it sucks so much? Because Linux sounds cool, while BSD sounds geeky. I was recently reminded of how badly Linux sucks when I went over some old code I'd written to get the CPU name and speed. The FreeBSD and OpenBSD implementations of this code each called a single sysctl for each result. The Linux version had to read /proc/cpuinfo and parse it. Actually, it had to parse it in two different ways, because it turns out the format of cpuinfo is different on x86 and all other platforms. Reading the battery life was even worse. On FreeBSD it's just a matter of reading the hw.acpi.battery.life sysctl (one line of code). For Linux, it involves parsing some messy procfs stuff with a format that has a habit of changing between releases. I don't understand how a developer could prefer Linux to any other UNIX.

Re:Can someone provide some insight?

[peragrin]

What goes around comes around.

Revisionist history is working both ways I see. Whenever Linux or another kernel developer would bring up a point of failure in Kolvias's scheduler instead of Fixing the problem Kolvias would lash out and say it wasn't broken.

CFS won not because it was a better scheduler at the time, but because Inglo worked with the developers to make it better, instead of fighting everyone who questioned anything about it. FOSS projects are about helping everyone, and listening to new Ideas. Something Kolvias was having a hard time doing.

That is at least how i read the whole debate.

Re:Can someone provide some insight?

[SEAL]

Whenever Linux or another kernel developer would bring up a point of failure in Kolvias's scheduler instead of Fixing the problem Kolvias would lash out and say it wasn't broken.

CFS won not because it was a better scheduler at the time, but because Inglo worked with the developers to make it better, instead of fighting everyone who questioned anything about it.

I believe Linus was the claiming Ingo worked better with the developers (or at least I saw him write to that effect on the LKML). By contrast, Kolivas had many individuals helping out with his branch who were quite pleased with his progress.

The problem is that Kolivas was working to help desktop, and particularly 3D game users. He'd say it wasn't broken if it was optimizing for those particular platforms at a cost of a fraction of a percent Oracle performance. Most of the kernel devs don't care 2 cents about 3D or desktop users, and many are employed by large enterprise businesses, so you can see who won.

Personally I think Kolivas should've been given full access to merge his code. Many of us found his work very useful and it's sad to see him driven away by a few people who are oblivious to what everyday users need. Even if his scheduler was not the default one, it still would've been nice to have as a mainline kernel option.

- SEAL

Re:Can someone provide some insight?

[recoiledsnake]

CFS won not because it was a better scheduler at the time, but because Inglo worked with the developers to make it better, instead of fighting everyone who questioned anything about it.

That was certainly not how I read it when it happened. I came off with the impression that Kolivas was ripped off of his ideas by Inglo and that Linus was wrong in brushing Inglo off and going with Inglo.

The LKML email linked reinforces my suspicion. Habits and attitudes don't die easily. We see ANOTHER guy complaining that his patch was half bakedly ripped off with no explanations whatsoever and a tangential acknowledgement.

FOSS projects are about helping everyone, and listening to new Ideas. Something Kolvias was having a hard time doing.

Did you even read the linked email? Inglo seems to be talking in patches and with no discussion. If something were to happen to him or if he quits due to any reason, there would be no one who would understand the code and the features behind it enough to maintain it.

Re:Can someone provide some insight?

[GooberToo]

Average? Probably nothing.

I don't think that's really fair. Average these days actually has a large spread. Average web user? Average game player? Average video watcher? Average musician? Averaging what? And that's the point of the CFS. Each of those users have different expectations and each reflect a different type of scheduler workload. The old scheduler has lots and lots of code to handle various performance problems for corner cases for each type of "average user" depicted above. This in turn means the code is hard to read, hard to understand, and even harder to maintain. Worse, some code which help some corner cases actually make things worse for others. This in turn means more special case coding and even more complex testing.

The CFS is designed to simplify most everything the O(1) schedule does without tons of complex heuristics and special code to address various corner cases. In other words, in stead of trying to guess what you really want, the CFS simply tries to be as fair as possible for everyone, thusly ensuring many categories of corner cases are simply no longer an issue with the CFS.

This does not mean CFS is always better than O(1), but early results indicate they are traveling down a good road. In fact, the last round of patches I read about, actually establish CFS ~12% faster then the old O(1) schedule. Of course, it's yet to be seen how well it will be received and how quickly it will prove it self with a large user base. The devil is in the details with schedulers and sometimes real world user workloads don't reflect anything which has been tested/validating during development.

Nonetheless, development on CFS continues and it certainly is providing hope it will be smaller, faster, provide lower latency, be easier to maintain, and address a wider range of workloads, including many corner cases, without requiring complex code to track and analyze heuristics.

Long story short, yes, if all goes well, even the average user may notice an improvement depending on the particulars of their workload. Wouldn't you notice a more responsive desktop? ;)

Re:Can someone provide some insight?

[m50d]

Well damn. That's just... embarassing. Why's everyone using linux if it sucks so much?

The same reason as windows: hardware support.

Re:Can someone provide some insight?

[rbanffy]

"The problem is that Kolivas was working to help desktop, and particularly 3D game users."

I don't know for sure, but I think it should be trivial to have a kernel switch activated on boot to set the preferred scheduler. This way, 3D gamers would be happy to set the -ks (I fail to remember its correct name) while the rest of us would be happy to leave it alone and get the CFS. Maybe by having a modular scheduler architecture would allow to have kernels with the -openvz or -oraclesbest or whatever other schedulers one may want (and could get support from the kernel developers).

A slightly more clever approach would be to let one switch schedulers on the fly, but I think it's asking too much.

Re: Can someone provide some insight?

[Dolda2000]

I was recently reminded of how badly Linux sucks when I went over some old code I'd written to get the CPU name and speed. The FreeBSD and OpenBSD implementations of this code each called a single sysctl for each result. The Linux version had to read /proc/cpuinfo and parse it. While I agree with you that Linux seems almost pathetic compared to BSD when under heavy load, I have to object to that reason for calling the *BSDs superior to Linux. In fact, the sysctl interface is the single most bothering thing about the *BSDs to me. Isn't the treatment of everything as a file the very core of the Unix philosophy? It may be true that files like /proc/cpuinfo could very well be made easier to parse, but I find nothing to complain about with e.g. the /proc/sys/ tree, or, for that matter, sysfs. Requiring a special syscall just to fetch kernel data that could just as well be made available through the file system is, in my mind, just ugly for a Unix system, and I can only imagine that it is done that way for historical reasons.

In almost any other way, however, I would prefer the *BSDs over Linux. I still run Linux, though, but that's mainly because I want to keep running Kerberized NFS. :) When FreeBSD implements that, I may well switch.

Re:Can someone provide some insight?

[AnyoneEB]

You are right. It would be easy. In fact, someone wrote a patch for it called plugsched. It was not accepted into the kernel due to the fact that it would supposedly discourage the idea of simply making a scheduler which worked well for everyone.

Re:Can someone provide some insight?

[ciroknight]

You shouldn't say "someone did", it was Kolivas himself that first offered the pluggable scheduler patch so that his patch could be used along side any new future schedulers and offer a concrete way to benchmark the changes caused by scheduling. And this was done years ago, circa 2004: http://ck.kolivas.org/patches/plugsched/

Of course, Linus and Ingo rejected those patches as well.

In a blind taste test....

[budword]

I doubt any of us could tell the difference. Storm in a tea cup.

Re:In a blind taste test....

[MyLongNickName]

As a windows user who has very little experience with Windows: This is one of the strengths of open source. if you have a large enough base of contributors, these "little" details are brought out into the open, and you can really understand how things work. I've read a bit on the subject, and it is interesting to see the different approaches that can be taken to something that most of us do not even think about.

With Windows, how does this work? I will never know for sure. if MS doesn't choose to make it known, it isn't known. If they choose to make it known, then I just have to trust they are telling the truth (Windows Update anyone).

With a project like this, you are much more likely to get the best approach to the situation.

Re:Snooze

[Reason58]

Seriously has Slashdot got such a boner for Linux that they'll post absolutely every fucking little boring thing they find? You must be new here.

Esoteric Discussions

[Archangel Michael]

More importantly, if there are more than one Scheduler, and if someone could tell the difference, why isn't s/he using the ALTERNATE Scheduler and compiling their own custom, tweaked and totally tuned kernel?

Seriously, most people aren't going to notice, and those that do notice, ought to be able to compile their own kernel, and ought to do exactly that. This is nothing short of an esoteric discussion and shouldn't extend beyond kernel developers. Most people don't know, and don't care which scheduler is implemented.

I'm one of those somewhere between caring and not. I only care about the supposed differences in approach to scheduling, and quite frankly, from what little I understand, the various schemes to scheduling have their advantages and disadvantages. I seriously doubt that ONE is better in all circumstances compared to all the others.

Ugh bring back 2.7 please

[maelstrom]

I know they've changed the model of development for the kernel, but how many new schedulers have we gone through between 2.4 and 2.6 now? Maybe it is just me, but the scheduler seems like a pretty important piece of the kernel.... Ripping it out every 6 months and calling it "stable" seems a bit off to me.

Oh well. I guess I'm just getting cranky in my "old" age.

Re:Ugh bring back 2.7 please

[luciofm]

There was the 2.4 schuduler, the old O(1) 2.6 scheduler and now the new 2.6 CFS scheduler...
This doenst seem to me to be ripped every 6 months, unless the 2.6 tree is just about 6 months older...

Re:more important things

[Hatta]

There isn't much more important work for a kernel than improving performance under load. They could probably do better by focusing on I/O scheduling than CPU scheduling. My CPU spends an inordinate amount of time waiting for IO. But kernel performance is of the utmost importance to kernel developers. What does it matter if it runs on your hardware if it doesn't run well?

Re:Snooze

[TheRaven64]

Maybe I missed them, but where were all of the Slashdot articles about the ULE 2 and ULE 3 FreeBSD schedulers? From all the benchmarks I've seen, they make the Linux scheduler look embarrassingly antiquated (performance characteristics matching the 4BSD scheduler that ULE was originally designed to replace).

Scheduler Question for the /. Experts

[Cassini2]

One way to cause dramatic performance problems on a Windows machine is to simply write a program that accesses lots of files. Performing a network backup with the Windows Networking API is a good example of this. Windows responds by fetching the files from disk and using system memory as a cache. In the process, the working set of programs running on the computer is paged out. The result is that low-priority activities can dramatically slow down potentially important activities on the computer. A good example of this is doing a network backup or a background virus scan on a Windows computer while trying to do any foreground activity (like browsing the web or using Microsoft Word).

So far, in my experience, Linux seems pretty immune to these priority inversions. Will the new scheduling algorithms allow low-priority processes to cause priority inversions by abusing non-processor resources like the network or disk drives?

Gross mis-characterizations of Molnar

[l4m3z0r]

As I started reading the comments on here I noticed that many were quick to down Ingo for his transgressions and its quite obvious from the comments that no one has bothered to read the exchange on LKML in order to become familiar with what is going on. I have read it, I have 0 bias for either Zippel or Molnar and I can say without any reservation that Zippel is a wank and Molnar is borderline saintly.

A recap of what I have read and understood about the entire situation:

• Zippel shows up with some patches out of the blue and makes some claims about them.
• Molnar takes time out of his day to look at and understand the patches and points out a few things such as lack of comments and missing functionality, also points out that some of the things Zippel wants to do have been done since the last time Zippel looked at the code and that he should look into the latest they've done.
• At this point Zippel decides to try to get into a pissing contest with Molnar and makes more claims about the superiority of his math and how the patches aren't the point.
• They argue back and forth all the while Molnar very seriously considering and understanding Zippels concerns. Molnar asks Zippel to split up his giant patch into smaller pieces that are commented and cleaned in order to bring the useful ones into the kernel more easily.
• Zippel essentially whines and repeats the same arguments over and over for a bunch of threads.
• Molnar basically pleads for him to cooperate so CFS can get the benefits Zippel has to offer.
• Zippel refuses, acts like a child.
• Molnar reworks a few of Zippel's ideas into smaller patches and incorporated them into CFS and attributes this to Zippel.
• Zippel is very rude about this and accuses Molnar of stealing from Zippel.
• Molnar again asks him to cooperate.
• Other people get annoyed start criticizing Zippel.

Ultimately I think Zippel is purposefully trying to provoke Molnar throughout all of this. His wild accusations are nothing more than games that he is playing, the guy has a chip on his shoulder and if Linux was my toy, I would have blocked him from the mailing lists.

Re:Snooze

[oojah]

I guess articles on the FreeBSD schedulers are either not being submitted or are being rejected. It's a shame either way. I don't use BSD (for no particular reason) but I'd still very much like to hear about what's going on there.