Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Hello, Erik. You wrote 7 ноября 2012 г., 19:11:03: EC That thread starts here: EC http://lists.freebsd.org/pipermail/freebsd-arch/2010-April/010143.html Year 2010! And we still limited by MAXPHYS (128K) transfers :( -- // Black Lion AKA Lev Serebryakov l...@freebsd.org ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
EC That thread starts here: EC http://lists.freebsd.org/pipermail/freebsd-arch/2010-April/010143.html Year 2010! And we still limited by MAXPHYS (128K) transfers :( put options MAXPHYS=2097152 in your kernel config. EVERYTHING works in all production machines for over a year the only exception is my laptop with OCZ petrol SSD that hangs on any transfer 1MB, i've set it to 0.5MB here. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Nov 8, 2012, at 12:56 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: EC That thread starts here: EC http://lists.freebsd.org/pipermail/freebsd-arch/2010-April/010143.html Year 2010! And we still limited by MAXPHYS (128K) transfers :( put options MAXPHYS=2097152 in your kernel config. EVERYTHING works in all production machines for over a year the only exception is my laptop with OCZ petrol SSD that hangs on any transfer 1MB, i've set it to 0.5MB here. Have you measured the performance increase? I'm also interested in bigger MAXBSIZE as this is what the NFS server uses as maximum transfer size. Linux and Solaris can do up to 1MB. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
But working on software to make it better in some kind of synthetic benchmark is common in commercial software world. (We have more performance per buck than company X) Synthetic benchmarks as you put it shouldn't be the ultimate basis for a decision, but instead allow users to gauge whether or not a certain software or hardware configuration is suitable for their given workload. No more, no less. only when OS is not tuned for benchmarks. You see that given OS is great for some database test doing repetitively few operations, then you run in for YOUR workload for which OS isn't tuned and it's bad. Even if it is still database only workload. Even worse that on the same machine you do other things. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
actually FreeBSD defaults are actually good for COMMON usage. and can be tuned. default MAXBSIZE is one exception. Common usage is vague. While FreeBSD might do ok for some applications (dev box, simple workstation/laptop, etc), there are other areas that require additional tuning to get better perf that arguably shouldn't as much (or there should be templates for doing so): 10GbE and mbuf and network tuning; file server and file descriptor, network tuning, etc; low latency desktop and scheduler tweaking; etc. still any idea why MAXBSIZE is 128kB by default. for modern hard disk it is a disaster. 2 or even 4 megabyte is OK. Not to say that freebsd is entirely at fault, but because it's more of a commodity OS that Linux, more tweaking is required... actually IMHO much more tweaking is needed with linux, at least from what i know from other people. And they are not newbies ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Tue, 6 Nov 2012, Yuri wrote: On 11/06/2012 11:10, Samuel J. Greear wrote: Single and multi-socket hardware are not really directly comparable in PostgreSQL tests. So if the CPUs are split between sockets, would such system generally perform better or worse with PostgeSQL vs. non-split situation? multiple sockets means more interconnect delay=lower performance. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
In my experience, in modern world, most of computers are not true-multiuser. It is dedicated servers (DB, front-end, middle layer with something like RoR or node.js) or personal (mobile) workstations. If hardware is shared between different tasks, it is shared via hypervisor and multiple OS instances... which is completely strange and inefficient. But yes i am aware that people do that. But that's not about performance at all ;) ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Nov 7, 2012, at 4:48 PM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: actually FreeBSD defaults are actually good for COMMON usage. and can be tuned. default MAXBSIZE is one exception. Common usage is vague. While FreeBSD might do ok for some applications (dev box, simple workstation/laptop, etc), there are other areas that require additional tuning to get better perf that arguably shouldn't as much (or there should be templates for doing so): 10GbE and mbuf and network tuning; file server and file descriptor, network tuning, etc; low latency desktop and scheduler tweaking; etc. still any idea why MAXBSIZE is 128kB by default. for modern hard disk it is a disaster. 2 or even 4 megabyte is OK. Not to say that freebsd is entirely at fault, but because it's more of a commodity OS that Linux, more tweaking is required... actually IMHO much more tweaking is needed with linux, at least from what i know from other people. And they are not newbies ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org Actually MAXBSIZE is 64k, MAXPHYS is 128k. There was a thread about NFS performance where it was mentioned that bigger MAXBSIZE leads to KVA fragmentation. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Den 07/11/2012 kl. 15.35 skrev Nikolay Denev nde...@gmail.com: Actually MAXBSIZE is 64k, MAXPHYS is 128k. There was a thread about NFS performance where it was mentioned that bigger MAXBSIZE leads to KVA fragmentation. That thread starts here: http://lists.freebsd.org/pipermail/freebsd-arch/2010-April/010143.html Erik ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org Actually MAXBSIZE is 64k, MAXPHYS is 128k. sorry MAXPHYS There was a thread about NFS performance where it was mentioned that bigger MAXBSIZE leads to KVA fragmentation. NFS is never fast. but EVERYTHING is slow when operating on big files. modern SATA disk can read/write over 1MB within single seek time ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
defaults) is sysctl/tunable variables set in the *BSD OSes (on DFly, FreeBSD, and NetBSD). Unfortunately (based on my experience) FreeBSD could be a lot better when it comes to defaults, and more tuning is required to actually FreeBSD defaults are actually good for COMMON usage. and can be tuned. default MAXBSIZE is one exception. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
some serious system issue. It looks like the DragonflyBSD folks made a goal to do well on pgbench and got to the level of ~88% of linux with 80 clients. It's just bad that anyone judge and (even worse) modify/tune operating system to do well in SINGLE benchmark running basically single program doing few repetitive things. Linux is tuned to win in benchmark and it does, while having disastrous performance in normal unix style usage - multiple different programs doing multiple different things for multiple different users - in the same time. This is a case with at least 99% of users. The less than 1% that have so heavy load that needs separete machine dedicated to single program doing one thing - could use linux (if it REALLY will be better in production workload ) or even better - use some dedicated hardware just for this, if it exist. Does machine that is dedicated to run single program need OS at all? In such benchmark FreeBSD with UFS wins hands down and that's the reason i use it. Still it is interesting WHY FreeBSD is slower in that special case, and if improvements on general behaviour can be found then it's nice to do them. I tried dragonflybsd some time ago and it's performance on normal usage is disastrous. Seems like Matthew Dillion years after splitting from FreeBSD because the algorithms used in FreeBSD were plain wrong - cannot do this better but still waste time and still at all cost want to prove he can. Tuning operating system for single benchmark is an example of that childish behaviour. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
2012/11/6 Wojciech Puchar woj...@wojtek.tensor.gdynia.pl: Tuning operating system for single benchmark is an example of that childish behaviour. LOL. That's what we did several years ago : http://people.freebsd.org/~kris/scaling/dfly.html I won't blame dflybsd for benchmarking something specific. Maybe there is something we can learn from what they did for the 3.2 release. -- Olivier Smedts _ ASCII ribbon campaign ( ) e-mail: oliv...@gid0.org- against HTML email vCards X www: http://www.gid0.org- against proprietary attachments / \ Il y a seulement 10 sortes de gens dans le monde : ceux qui comprennent le binaire, et ceux qui ne le comprennent pas. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Your entire email is conjecture, the performance of DragonFly 3.2 is improved across the board vs 3.0. Not just batch performance, interactive performance (especially under X11) is also greatly improved. i must try. i checked 3.0 and earlier versions and it was a disaster. Relative to FreeBSD of course. Sam On Tue, Nov 6, 2012 at 2:25 AM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: some serious system issue. It looks like the DragonflyBSD folks made a goal to do well on pgbench and got to the level of ~88% of linux with 80 clients. It's just bad that anyone judge and (even worse) modify/tune operating system to do well in SINGLE benchmark running basically single program doing few repetitive things. Linux is tuned to win in benchmark and it does, while having disastrous performance in normal unix style usage - multiple different programs doing multiple different things for multiple different users - in the same time. This is a case with at least 99% of users. The less than 1% that have so heavy load that needs separete machine dedicated to single program doing one thing - could use linux (if it REALLY will be better in production workload ) or even better - use some dedicated hardware just for this, if it exist. Does machine that is dedicated to run single program need OS at all? In such benchmark FreeBSD with UFS wins hands down and that's the reason i use it. Still it is interesting WHY FreeBSD is slower in that special case, and if improvements on general behaviour can be found then it's nice to do them. I tried dragonflybsd some time ago and it's performance on normal usage is disastrous. Seems like Matthew Dillion years after splitting from FreeBSD because the algorithms used in FreeBSD were plain wrong - cannot do this better but still waste time and still at all cost want to prove he can. Tuning operating system for single benchmark is an example of that childish behaviour. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Tuning operating system for single benchmark is an example of that childish behaviour. LOL. That's what we did several years ago : http://people.freebsd.org/~kris/scaling/dfly.html i've seen that page some time ago but i don't really care of it. i just wasn't interested. Still - DOING such benchmark is good, as it can show general problems in used algorithms. But working on software to make it better in some kind of synthetic benchmark is common in commercial software world. (We have more performance per buck than company X) ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Your entire email is conjecture, the performance of DragonFly 3.2 is improved across the board vs 3.0. Not just batch performance, interactive performance (especially under X11) is also greatly improved. Sam On Tue, Nov 6, 2012 at 2:25 AM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: some serious system issue. It looks like the DragonflyBSD folks made a goal to do well on pgbench and got to the level of ~88% of linux with 80 clients. It's just bad that anyone judge and (even worse) modify/tune operating system to do well in SINGLE benchmark running basically single program doing few repetitive things. Linux is tuned to win in benchmark and it does, while having disastrous performance in normal unix style usage - multiple different programs doing multiple different things for multiple different users - in the same time. This is a case with at least 99% of users. The less than 1% that have so heavy load that needs separete machine dedicated to single program doing one thing - could use linux (if it REALLY will be better in production workload ) or even better - use some dedicated hardware just for this, if it exist. Does machine that is dedicated to run single program need OS at all? In such benchmark FreeBSD with UFS wins hands down and that's the reason i use it. Still it is interesting WHY FreeBSD is slower in that special case, and if improvements on general behaviour can be found then it's nice to do them. I tried dragonflybsd some time ago and it's performance on normal usage is disastrous. Seems like Matthew Dillion years after splitting from FreeBSD because the algorithms used in FreeBSD were plain wrong - cannot do this better but still waste time and still at all cost want to prove he can. Tuning operating system for single benchmark is an example of that childish behaviour. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Nov 6, 2012, at 8:55 AM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: Tuning operating system for single benchmark is an example of that childish behaviour. LOL. That's what we did several years ago : http://people.freebsd.org/~kris/scaling/dfly.html i've seen that page some time ago but i don't really care of it. i just wasn't interested. Still - DOING such benchmark is good, as it can show general problems in used algorithms. But working on software to make it better in some kind of synthetic benchmark is common in commercial software world. (We have more performance per buck than company X) Synthetic benchmarks as you put it shouldn't be the ultimate basis for a decision, but instead allow users to gauge whether or not a certain software or hardware configuration is suitable for their given workload. No more, no less. The fact that they're being used in this manner is a bit like a salesman selling snake oil as the results aren't necessarily the result of a best configuration for all competing platforms, but instead an unknown configuration in this case. A similar statement about the importance of micro benchmarks can be made... Thanks, -Garrett ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Nov 6, 2012, at 1:26 AM, Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: defaults) is sysctl/tunable variables set in the *BSD OSes (on DFly, FreeBSD, and NetBSD). Unfortunately (based on my experience) FreeBSD could be a lot better when it comes to defaults, and more tuning is required to actually FreeBSD defaults are actually good for COMMON usage. and can be tuned. default MAXBSIZE is one exception. Common usage is vague. While FreeBSD might do ok for some applications (dev box, simple workstation/laptop, etc), there are other areas that require additional tuning to get better perf that arguably shouldn't as much (or there should be templates for doing so): 10GbE and mbuf and network tuning; file server and file descriptor, network tuning, etc; low latency desktop and scheduler tweaking; etc. Not to say that freebsd is entirely at fault, but because it's more of a commodity OS that Linux, more tweaking is required... Thanks, -Garrett ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On 11/05/2012 12:52, Garrett Cooper wrote: FWIW, I think that the last time scheduler benchmarks from anyone at @FreeBSD.org (was kris@ the last one, or has flo@ run benchmarks since I myself ran the similar test on i7 920 (4 cores 8 threads) @ 2.67 24GB with 9.1-RC3 with all the same params except shmem size was 4GB, not 6GB: http://i.imgur.com/mfnqr.png In DragonflyBSD tests FreeBSD peaked at 96k tps. And my machine, with roughly 3X lesser power, peaked at 44.5k tps. So in my test BSD performed relatively better. And graph shape is more resembling linux/DragonflyBSD ones. It looks like in their test FreeBSD behaved in somewhat impaired way. Any ideas what can I try to tune? Yuri ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Tue, Nov 6, 2012 at 10:58 AM, Yuri y...@rawbw.com wrote: On 11/05/2012 12:52, Garrett Cooper wrote: FWIW, I think that the last time scheduler benchmarks from anyone at @FreeBSD.org (was kris@ the last one, or has flo@ run benchmarks since I myself ran the similar test on i7 920 (4 cores 8 threads) @ 2.67 24GB with 9.1-RC3 with all the same params except shmem size was 4GB, not 6GB: http://i.imgur.com/mfnqr.png In DragonflyBSD tests FreeBSD peaked at 96k tps. And my machine, with roughly 3X lesser power, peaked at 44.5k tps. So in my test BSD performed relatively better. And graph shape is more resembling linux/DragonflyBSD ones. It looks like in their test FreeBSD behaved in somewhat impaired way. Any ideas what can I try to tune? Single and multi-socket hardware are not really directly comparable in PostgreSQL tests. Sam Yuri ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On 11/06/2012 11:10, Samuel J. Greear wrote: Single and multi-socket hardware are not really directly comparable in PostgreSQL tests. So if the CPUs are split between sockets, would such system generally perform better or worse with PostgeSQL vs. non-split situation? Yuri ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Tue, Nov 6, 2012 at 2:01 PM, Yuri y...@rawbw.com wrote: On 11/06/2012 11:10, Samuel J. Greear wrote: Single and multi-socket hardware are not really directly comparable in PostgreSQL tests. So if the CPUs are split between sockets, would such system generally perform better or worse with PostgeSQL vs. non-split situation? Yuri Unless the algorithms you are testing are able to operate entirely out of the processors caches (and PostgreSQL does not fall into that category) performance will be generally lower as you add sockets. FreeBSD's ULE scheduler is aware of this and takes it into account, but the performance ULE is able to maintain across sockets (this applies to other OS's schedulers too) is more damage control than anything else. Sam ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
Hello, Wojciech. You wrote 6 ноября 2012 г., 13:25:05: WP performance in normal unix style usage - multiple different programs doing WP multiple different things for multiple different users - in the same time. WP This is a case with at least 99% of users. The less than 1% that have so WP heavy load that needs separete machine dedicated to single program doing WP one thing In my experience, in modern world, most of computers are not true-multiuser. It is dedicated servers (DB, front-end, middle layer with something like RoR or node.js) or personal (mobile) workstations. If hardware is shared between different tasks, it is shared via hypervisor and multiple OS instances... -- // Black Lion AKA Lev Serebryakov l...@freebsd.org ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
pgbench performance is lagging compared to Linux and DragonflyBSD?
There is the post by DragonflyBSD folks that claims that Linux and DragonflyBSD are quite ahead of FreeBSD on pgbench test on 12 Core 2x Xeon X5650 with 24 threads. Here are their results with graphs: http://lists.dragonflybsd.org/pipermail/users/attachments/20121010/7996ff88/attachment-0002.pdf And here is their original post: http://lists.dragonflybsd.org/pipermail/users/2012-October/017536.html I am not sure if this is the problem of some sysctl or kernel parameters or some serious system issue. It looks like the DragonflyBSD folks made a goal to do well on pgbench and got to the level of ~88% of linux with 80 clients. Yuri ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
On Mon, Nov 5, 2012 at 10:52 AM, Yuri y...@rawbw.com wrote: There is the post by DragonflyBSD folks that claims that Linux and DragonflyBSD are quite ahead of FreeBSD on pgbench test on 12 Core 2x Xeon X5650 with 24 threads. Here are their results with graphs: http://lists.dragonflybsd.org/** pipermail/users/attachments/**20121010/7996ff88/attachment-**0002.pdfhttp://lists.dragonflybsd.org/pipermail/users/attachments/20121010/7996ff88/attachment-0002.pdf And here is their original post: http://lists.dragonflybsd.org/** pipermail/users/2012-October/**017536.htmlhttp://lists.dragonflybsd.org/pipermail/users/2012-October/017536.html I am not sure if this is the problem of some sysctl or kernel parameters or some serious system issue. It looks like the DragonflyBSD folks made a goal to do well on pgbench and got to the level of ~88% of linux with 80 clients. The important item that has been left out (or is just implied as OS level defaults) is sysctl/tunable variables set in the *BSD OSes (on DFly, FreeBSD, and NetBSD). Unfortunately (based on my experience) FreeBSD could be a lot better when it comes to defaults, and more tuning is required to get better performance. So if they're working with the OS defaults, this might not be a fair equivalent to the best performance that FreeBSD can yield, but it's probably fair to do this for the sake of repeatability and to prove what these OSes can do out of the box. This is in addition to the [lock] contention issues that jeffr@ and a few others are working on alleviating. FWIW, I think that the last time scheduler benchmarks from anyone at @FreeBSD.org (was kris@ the last one, or has flo@ run benchmarks since then? My Googling is a bit inconclusive) was run was several years ago as well, so if Linux has improved I'm not at all surprised. However, please also take into consideration that the hardware then and the hardware now are grossly different. So the interactions between the hardware then and the hardware now might differ greatly. In short, more inspection needs to be done to figure out whether or not the findings are true [with caveats] or false. Thanks, -Garrett ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: pgbench performance is lagging compared to Linux and DragonflyBSD?
The important item that has been left out (or is just implied as OS level defaults) is sysctl/tunable variables set in the *BSD OSes (on DFly, FreeBSD, and NetBSD). Unfortunately (based on my experience) FreeBSD could be a lot better when it comes to defaults, and more tuning is required to get better performance. So if they're working with the OS defaults, this might not be a fair equivalent to the best performance that FreeBSD can yield, but it's probably fair to do this for the sake of repeatability and to prove what these OSes can do out of the box. This is in addition to the [lock] contention issues that jeffr@ and a few others are working on alleviating. has someone wrote a howto for how to tune pgsql 9+ in FreeBSD? im mostly asking here to get information posted here for future reference, as google will pick this thread up and it will help others. -- Sam Fourman Jr. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
