Hi list, I am evaluating the scheduler behavior in xen. I am using Xen 3.3.0 Dom0 and Dom1,2,3 and 4 all are opensuse 11. I have one CPU Intensive *TEST* which has no. of arithmatic instruction in an infinite while() loop. i am pinging domain1 with an external machine. and noting the RTT values. i have below experiments *time (s) domain state* 0 dom0,1,2,3,4 all idle 50 dom2 *TEST *started 100 dom3 *TEST *started 150 dom4 *TEST *started 200 dom0 *TEST *started 250 dom2 *TEST* stopped 300 dom3 *TEST* stopped 350 dom4 *TEST* stopped 400 dom0 *TEST* stopped For these 400 seconds time, i have performed experiments with Credit and SEDF sceduler. the configuration is Credit configuration - weight 256, cap 0 *Domain VCPU * 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. SEDF configuration - Period 10ms slice - 1.9ms *Domain VCPU * 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. the results of RTT values are attached herewith. the performance of Credit is very bad in comparison to SEDF in this scenario. Please provide me some thought on it. Thanks and Regards Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, INDIA http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
A couple of comments: * Why did you pin the vcpus to pcpus? AIUI, pings will always be handled by vcpu 0. So if you pin the vcpus to pcpu0, and pcpu0 is busy, it can''t migrate over to pcpu1 if it''s not busy. Try unpinning the cpus and see if that changes anything. * The Credit scheduler is known to have some issues with latency-sensitive workloads. Workloads like pass-though video are becoming more important, so there''s been a lot of discussion about this subject. I''m working on a new scheduler, credit2, that will hopefully address a lot of these issues. * "Ping" is not an application that people find it important to virtualize. :-) Remember that end-to-end application performance and fairness are the high-level goals, so although "ping" may be a useful test to isolate certain aspects of a scheduler, it should never be used to evaluate the "goodness" of one scheduler over another. * That said, it''s not clear to me (given what I know of sched_credit and ping) why you''d see these results. -George On Tue, May 5, 2009 at 10:25 AM, gaurav somani <onlineengineer@gmail.com> wrote:> Hi list, > > I am evaluating the scheduler behavior in xen. > > I am using Xen 3.3.0 > Dom0 and Dom1,2,3 and 4 all are opensuse 11. > I have one CPU Intensive TEST which has no. of arithmatic instruction in an > infinite while() loop. > i am pinging domain1 with an external machine. and noting the RTT values. > > i have below experiments > time (s) domain state > 0 dom0,1,2,3,4 all idle > 50 dom2 TEST started > 100 dom3 TEST started > 150 dom4 TEST started > 200 dom0 TEST started > 250 dom2 TEST stopped > 300 dom3 TEST stopped > 350 dom4 TEST stopped > 400 dom0 TEST stopped > > For these 400 seconds time, i have performed experiments with Credit and > SEDF sceduler. > the configuration is > > > Credit configuration - weight 256, cap 0 > Domain VCPU > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > > SEDF configuration - Period 10ms slice - 1.9ms > Domain VCPU > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > the results of RTT values are attached herewith. the performance of Credit > is very bad in comparison to SEDF in this scenario. > Please provide me some thought on it. > > > Thanks and Regards > > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > INDIA > > http://dcomp.daiict.ac.in/~gaurav > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Hi, Your experiment results are actually “expected”. There are several reasons make the results happen. Firstly, Credit has long time slice (i.e., 30 millisec by default), and this make the observed close-to 30ms RTT to happen, actually, during such long RTT (the spike), the domain in test (domain 1 of your experiment) is just scheduled-out (actually, I guess the spikes illustrated in your figure 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus (may executing the TEST program, which is cpu-hogs), and they will not give up their 30-ms time slice if they are not preempted. Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, it will not be BOOSTed even it do have some I/O requests to handle. This explains why your figure1, which has little spikes in the first 150 second, and crowed spikes follow. Since vcpu of domain 1 simply used out their credits, and can not preempt the other running vcpus of other domains, even they do have I/O operations to accomplish (i.e., PING). Lastly, compared with Credit, SEDF generally has short time slices, and it has bigger chances to preempt current running vcpu when another vcpu found it has some I/O operations to be accomplished. (actually, I donot understand SEDF very well, but conclude these features by reading the trace data) Actually, you can use BCredit patch to eliminate the spikes, I can guarantee that it will work very well! :-) And I agree with George that you should not use PING to justify whether a scheduler is good or not. Best, Zhiyuan _____ 发件人: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] 代表 gaurav somani 发送时间: 2009年5月5日 17:26 收件人: xen-devel@lists.xensource.com 主题: [Xen-devel] Credit scheduler vs SEDF scheduler Hi list, I am evaluating the scheduler behavior in xen. I am using Xen 3.3.0 Dom0 and Dom1,2,3 and 4 all are opensuse 11. I have one CPU Intensive TEST which has no. of arithmatic instruction in an infinite while() loop. i am pinging domain1 with an external machine. and noting the RTT values. i have below experiments time (s) domain state 0 dom0,1,2,3,4 all idle 50 dom2 TEST started 100 dom3 TEST started 150 dom4 TEST started 200 dom0 TEST started 250 dom2 TEST stopped 300 dom3 TEST stopped 350 dom4 TEST stopped 400 dom0 TEST stopped For these 400 seconds time, i have performed experiments with Credit and SEDF sceduler. the configuration is Credit configuration - weight 256, cap 0 Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. SEDF configuration - Period 10ms slice - 1.9ms Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. the results of RTT values are attached herewith. the performance of Credit is very bad in comparison to SEDF in this scenario. Please provide me some thought on it. Thanks and Regards Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, INDIA http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
gaurav somani
2009-May-06 09:09 UTC
Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler
Hi Zhiyuan and George, Thanks for the comments. you are right i was getting spikes with more than 100ms and some times 400ms. I will be trying with the patch u suggetsed. one more aspect, can we classify schedulers according to applications they run.? Like compute intensive on Credit and Latency intensive on sEDF or some trade off between them (In a cluster of servers). Thanks Gaurav 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com>> Hi, > > > > Your experiment results are actually “expected”. There are several reasons > make the results happen. > > > > Firstly, Credit has long time slice (i.e., 30 millisec by default), and > this make the observed close-to 30ms RTT to happen, actually, during such > long RTT (the spike), the domain in test (domain 1 of your experiment) is > just scheduled-out (actually, I guess the spikes illustrated in your figure > 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus > (may executing the TEST program, which is cpu-hogs), and they will not give > up their 30-ms time slice if they are not preempted. > > > > Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, > it will not be BOOSTed even it do have some I/O requests to handle. This > explains why your figure1, which has little spikes in the first 150 second, > and crowed spikes follow. Since vcpu of domain 1 simply used out their > credits, and can not preempt the other running vcpus of other domains, even > they do have I/O operations to accomplish (i.e., PING). > > > > Lastly, compared with Credit, SEDF generally has short time slices, and it > has bigger chances to preempt current running vcpu when another vcpu found > it has some I/O operations to be accomplished. (actually, I donot understand > SEDF very well, but conclude these features by reading the trace data) > > > > Actually, you can use BCredit patch to eliminate the spikes, I can > guarantee that it will work very well! J And I agree with George that you > should not use PING to justify whether a scheduler is good or not. > > > > Best, > > Zhiyuan > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月5日 17:26 > *收件人:* xen-devel@lists.xensource.com > *主题:* [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > I am evaluating the scheduler behavior in xen. > > I am using Xen 3.3.0 > Dom0 and Dom1,2,3 and 4 all are opensuse 11. > I have one CPU Intensive *TEST* which has no. of arithmatic instruction in > an infinite while() loop. > i am pinging domain1 with an external machine. and noting the RTT values. > > i have below experiments > *time (s) domain state* > 0 dom0,1,2,3,4 all idle > 50 dom2 *TEST *started > 100 dom3 *TEST *started > 150 dom4 *TEST *started > 200 dom0 *TEST *started > 250 dom2 *TEST* stopped > 300 dom3 *TEST* stopped > 350 dom4 *TEST* stopped > 400 dom0 *TEST* stopped > > For these 400 seconds time, i have performed experiments with Credit and > SEDF sceduler. > the configuration is > > > Credit configuration - weight 256, cap 0 > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > > SEDF configuration - Period 10ms slice - 1.9ms > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > the results of RTT values are attached herewith. the performance of Credit > is very bad in comparison to SEDF in this scenario. > Please provide me some thought on it. > > > Thanks and Regards > > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > INDIA > > http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> > > > >-- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav http://www.linkedin.com/in/gauravsomani _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Believe your can use BCredit for this job. However, you can only set the attributes of a domain statically, which means the scheduler can not adjust its behavior according to the types of the work loads. Remember there is some publication in proceedings of VEE 2009, which had addressed your problem. The solution continuously guests the performance characteristics of the application, and adjust the preemption decisions of the scheduler itself. You can take a look at that, but unfortunately, they did not provide the patch behind the solution. Besides, I am not sure that whether such solution is really helpful to solve real problems on real hardware, since it is based its decisions on possibilities. Zhiyuan _____ 发件人: gaurav somani [mailto:onlineengineer@gmail.com] 发送时间: 2009年5月6日 17:09 收件人: Zhiyuan Shao 抄送: xen-devel@lists.xensource.com 主题: Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler Hi Zhiyuan and George, Thanks for the comments. you are right i was getting spikes with more than 100ms and some times 400ms. I will be trying with the patch u suggetsed. one more aspect, can we classify schedulers according to applications they run.? Like compute intensive on Credit and Latency intensive on sEDF or some trade off between them (In a cluster of servers). Thanks Gaurav 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com> Hi, Your experiment results are actually “expected”. There are several reasons make the results happen. Firstly, Credit has long time slice (i.e., 30 millisec by default), and this make the observed close-to 30ms RTT to happen, actually, during such long RTT (the spike), the domain in test (domain 1 of your experiment) is just scheduled-out (actually, I guess the spikes illustrated in your figure 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus (may executing the TEST program, which is cpu-hogs), and they will not give up their 30-ms time slice if they are not preempted. Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, it will not be BOOSTed even it do have some I/O requests to handle. This explains why your figure1, which has little spikes in the first 150 second, and crowed spikes follow. Since vcpu of domain 1 simply used out their credits, and can not preempt the other running vcpus of other domains, even they do have I/O operations to accomplish (i.e., PING). Lastly, compared with Credit, SEDF generally has short time slices, and it has bigger chances to preempt current running vcpu when another vcpu found it has some I/O operations to be accomplished. (actually, I donot understand SEDF very well, but conclude these features by reading the trace data) Actually, you can use BCredit patch to eliminate the spikes, I can guarantee that it will work very well! :-) And I agree with George that you should not use PING to justify whether a scheduler is good or not. Best, Zhiyuan _____ 发件人: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] 代表 gaurav somani 发送时间: 2009年5月5日 17:26 收件人: xen-devel@lists.xensource.com 主题: [Xen-devel] Credit scheduler vs SEDF scheduler Hi list, I am evaluating the scheduler behavior in xen. I am using Xen 3.3.0 Dom0 and Dom1,2,3 and 4 all are opensuse 11. I have one CPU Intensive TEST which has no. of arithmatic instruction in an infinite while() loop. i am pinging domain1 with an external machine. and noting the RTT values. i have below experiments time (s) domain state 0 dom0,1,2,3,4 all idle 50 dom2 TEST started 100 dom3 TEST started 150 dom4 TEST started 200 dom0 TEST started 250 dom2 TEST stopped 300 dom3 TEST stopped 350 dom4 TEST stopped 400 dom0 TEST stopped For these 400 seconds time, i have performed experiments with Credit and SEDF sceduler. the configuration is Credit configuration - weight 256, cap 0 Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. SEDF configuration - Period 10ms slice - 1.9ms Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. the results of RTT values are attached herewith. the performance of Credit is very bad in comparison to SEDF in this scenario. Please provide me some thought on it. Thanks and Regards Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, INDIA http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> -- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav http://www.linkedin.com/in/gauravsomani _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
gaurav somani
2009-May-10 12:47 UTC
Re: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler
Hi list, Some doubts. (1) How the guest kernel scheduling strategies (linux kernel scheduling like DEADLINE and IO schedulers in guests) affect the VM Scheduling? (2) I want to know what types of events are ocurring during scheduling- which domain is scheduled and what types of tasks it is running. I am using xenmon- some probs. (1) it does not show me --iocount option correctly. it always shows iocount0, although i am running dd utility in one of the guset domain. (2) when using xentop - it does not show any N/W data for domain 0. am i missing something in configuration. 2009/5/7 Zhiyuan Shao <zyshao.maillist@gmail.com>> Believe your can use BCredit for this job. However, you can only set the > attributes of a domain statically, which means the scheduler can not adjust > its behavior according to the types of the work loads. > > > > Remember there is some publication in proceedings of VEE 2009, which had > addressed your problem. The solution continuously guests the performance > characteristics of the application, and adjust the preemption decisions of > the scheduler itself. You can take a look at that, but unfortunately, they > did not provide the patch behind the solution. Besides, I am not sure that > whether such solution is really helpful to solve real problems on real > hardware, since it is based its decisions on possibilities. > > > > Zhiyuan > > > ------------------------------ > > *发件人:* gaurav somani [mailto:onlineengineer@gmail.com] > *发送时间:* 2009年5月6日 17:09 > *收件人:* Zhiyuan Shao > *抄送:* xen-devel@lists.xensource.com > *主题:* Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi Zhiyuan and George, > > Thanks for the comments. > > you are right i was getting spikes with more than 100ms and some times > 400ms. > > I will be trying with the patch u suggetsed. > > one more aspect, can we classify schedulers according to applications they > run.? > > Like compute intensive on Credit and Latency intensive on sEDF or some > trade off between them (In a cluster of servers). > > > Thanks > > Gaurav > > 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com> > > Hi, > > > > Your experiment results are actually “expected”. There are several reasons > make the results happen. > > > > Firstly, Credit has long time slice (i.e., 30 millisec by default), and > this make the observed close-to 30ms RTT to happen, actually, during such > long RTT (the spike), the domain in test (domain 1 of your experiment) is > just scheduled-out (actually, I guess the spikes illustrated in your figure > 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus > (may executing the TEST program, which is cpu-hogs), and they will not give > up their 30-ms time slice if they are not preempted. > > > > Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, > it will not be BOOSTed even it do have some I/O requests to handle. This > explains why your figure1, which has little spikes in the first 150 second, > and crowed spikes follow. Since vcpu of domain 1 simply used out their > credits, and can not preempt the other running vcpus of other domains, even > they do have I/O operations to accomplish (i.e., PING). > > > > Lastly, compared with Credit, SEDF generally has short time slices, and it > has bigger chances to preempt current running vcpu when another vcpu found > it has some I/O operations to be accomplished. (actually, I donot understand > SEDF very well, but conclude these features by reading the trace data) > > > > Actually, you can use BCredit patch to eliminate the spikes, I can > guarantee that it will work very well! J And I agree with George that you > should not use PING to justify whether a scheduler is good or not. > > > > Best, > > Zhiyuan > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月5日 17:26 > *收件人:* xen-devel@lists.xensource.com > *主题:* [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > I am evaluating the scheduler behavior in xen. > > I am using Xen 3.3.0 > Dom0 and Dom1,2,3 and 4 all are opensuse 11. > I have one CPU Intensive *TEST* which has no. of arithmatic instruction in > an infinite while() loop. > i am pinging domain1 with an external machine. and noting the RTT values. > > i have below experiments > *time (s) domain state* > 0 dom0,1,2,3,4 all idle > 50 dom2 *TEST *started > 100 dom3 *TEST *started > 150 dom4 *TEST *started > 200 dom0 *TEST *started > 250 dom2 *TEST* stopped > 300 dom3 *TEST* stopped > 350 dom4 *TEST* stopped > 400 dom0 *TEST* stopped > > For these 400 seconds time, i have performed experiments with Credit and > SEDF sceduler. > the configuration is > > > Credit configuration - weight 256, cap 0 > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > > SEDF configuration - Period 10ms slice - 1.9ms > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > the results of RTT values are attached herewith. the performance of Credit > is very bad in comparison to SEDF in this scenario. > Please provide me some thought on it. > > > Thanks and Regards > > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > INDIA > > http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> > > > > > > > -- > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > Gandhinagar > Gujarat, INDIA > > onlineengineer@ieee.org > gaurav_somani@daiict.ac.in > http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> > http://www.linkedin.com/in/gauravsomani >-- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav http://www.linkedin.com/in/gauravsomani _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Zhiyuan Shao
2009-May-11 09:54 UTC
答复: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler
Some points I think that may help to your doubts: 1) how the scheduler of guest OS and VMM cooperate together. Frankly, I also think this is a rather interesting topic to research, and currently working on that. However, before putting hands on this topic, we need some answers on several other tightly related problems. First problem is time keeping mechanism of Xen VMM. Not yet reading all the related codes of Xen, I guess for PV guests (do not consider the HVM case, since it is more complicated than PV), each time that a VCPU is scheduled in, one of the members of VCPU structure that stores current time will be updated (see struct vcpu_time_info), while the time precedes when the VCPU is actually is scheduled-out is simply “lost”. This will definitely cause problems by confusing the scheduler of the guest OS, however, hardly find documents that elaborate this problem or analyze what will happen in deep. The second problem is load-balancing. Unfortunately, this is a more complicated problem. The guest OS (for the SMP guest cases) has its own load balancing strategy, and I think the bottom line is that it will not waste processing time. Taking an example here, running a single thread process (such as “dd if=/dev/zero of=./test.file bs=64 count=3200K”, it writes a 200MB file to virtual disk) inside a two-vcpu PV guest. The process originally running on top of single VCPU, but after a while, the VCPU is scheduled out (blocked and waiting for the data actually write to the virtual disk file), and another VCPU is scheduled in, the process will migrate to the other VCPU naturally. This case will result in extra overhead, and I got performance data to prove this. Besides understanding the load balancing strategy of guest OS, load balancing for scheduler of the VMM deserves also careful consideration and design. Current default scheduler of Credit actually do not doing very well on this point. The PCPU will steal VCPU from its neighbors once it found it is idle. Smarter solutions are needed in later schedulers (as George had mentioned). 2) what types of events are ocurring during scheduling I think it is good for you to check the xentrace data for this problem. 3) xenmon problems. I had not used that and can not give you any suggestions, sorry for that. 4) Xentop Do not know what you mean for N/W? Best, Zhiyuan _____ 发件人: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] 代表 gaurav somani 发送时间: 2009年5月10日 20:48 收件人: xen-devel@lists.xensource.com 主题: Re: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler Hi list, Some doubts. (1) How the guest kernel scheduling strategies (linux kernel scheduling like DEADLINE and IO schedulers in guests) affect the VM Scheduling? (2) I want to know what types of events are ocurring during scheduling- which domain is scheduled and what types of tasks it is running. I am using xenmon- some probs. (1) it does not show me --iocount option correctly. it always shows iocount0, although i am running dd utility in one of the guset domain. (2) when using xentop - it does not show any N/W data for domain 0. am i missing something in configuration. 2009/5/7 Zhiyuan Shao <zyshao.maillist@gmail.com> Believe your can use BCredit for this job. However, you can only set the attributes of a domain statically, which means the scheduler can not adjust its behavior according to the types of the work loads. Remember there is some publication in proceedings of VEE 2009, which had addressed your problem. The solution continuously guests the performance characteristics of the application, and adjust the preemption decisions of the scheduler itself. You can take a look at that, but unfortunately, they did not provide the patch behind the solution. Besides, I am not sure that whether such solution is really helpful to solve real problems on real hardware, since it is based its decisions on possibilities. Zhiyuan _____ 发件人: gaurav somani [mailto:onlineengineer@gmail.com] 发送时间: 2009年5月6日 17:09 收件人: Zhiyuan Shao 抄送: xen-devel@lists.xensource.com 主题: Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler Hi Zhiyuan and George, Thanks for the comments. you are right i was getting spikes with more than 100ms and some times 400ms. I will be trying with the patch u suggetsed. one more aspect, can we classify schedulers according to applications they run.? Like compute intensive on Credit and Latency intensive on sEDF or some trade off between them (In a cluster of servers). Thanks Gaurav 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com> Hi, Your experiment results are actually “expected”. There are several reasons make the results happen. Firstly, Credit has long time slice (i.e., 30 millisec by default), and this make the observed close-to 30ms RTT to happen, actually, during such long RTT (the spike), the domain in test (domain 1 of your experiment) is just scheduled-out (actually, I guess the spikes illustrated in your figure 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus (may executing the TEST program, which is cpu-hogs), and they will not give up their 30-ms time slice if they are not preempted. Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, it will not be BOOSTed even it do have some I/O requests to handle. This explains why your figure1, which has little spikes in the first 150 second, and crowed spikes follow. Since vcpu of domain 1 simply used out their credits, and can not preempt the other running vcpus of other domains, even they do have I/O operations to accomplish (i.e., PING). Lastly, compared with Credit, SEDF generally has short time slices, and it has bigger chances to preempt current running vcpu when another vcpu found it has some I/O operations to be accomplished. (actually, I donot understand SEDF very well, but conclude these features by reading the trace data) Actually, you can use BCredit patch to eliminate the spikes, I can guarantee that it will work very well! :-) And I agree with George that you should not use PING to justify whether a scheduler is good or not. Best, Zhiyuan _____ 发件人: xen-devel-bounces@lists.xensource.com [mailto:xen-devel-bounces@lists.xensource.com] 代表 gaurav somani 发送时间: 2009年5月5日 17:26 收件人: xen-devel@lists.xensource.com 主题: [Xen-devel] Credit scheduler vs SEDF scheduler Hi list, I am evaluating the scheduler behavior in xen. I am using Xen 3.3.0 Dom0 and Dom1,2,3 and 4 all are opensuse 11. I have one CPU Intensive TEST which has no. of arithmatic instruction in an infinite while() loop. i am pinging domain1 with an external machine. and noting the RTT values. i have below experiments time (s) domain state 0 dom0,1,2,3,4 all idle 50 dom2 TEST started 100 dom3 TEST started 150 dom4 TEST started 200 dom0 TEST started 250 dom2 TEST stopped 300 dom3 TEST stopped 350 dom4 TEST stopped 400 dom0 TEST stopped For these 400 seconds time, i have performed experiments with Credit and SEDF sceduler. the configuration is Credit configuration - weight 256, cap 0 Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. SEDF configuration - Period 10ms slice - 1.9ms Domain VCPU 0 2 1 2 2 2 3 2 4 2 all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. the results of RTT values are attached herewith. the performance of Credit is very bad in comparison to SEDF in this scenario. Please provide me some thought on it. Thanks and Regards Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, INDIA http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> -- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> http://www.linkedin.com/in/gauravsomani -- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav http://www.linkedin.com/in/gauravsomani _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
gaurav somani
2009-May-30 16:33 UTC
Re: 答复: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler
Hi all, I mean N/W for network send and receive statistics in xentop. I also wish to know the actual credit status of each domain with time granularity like after each 30 ms. Is there anything helpful available. Thanks Gaurav 2009/5/11 Zhiyuan Shao <zyshao.maillist@gmail.com>> Some points I think that may help to your doubts: > > 1) how the scheduler of guest OS and VMM cooperate together. > > Frankly, I also think this is a rather interesting topic to research, and > currently working on that. However, before putting hands on this topic, we > need some answers on several other tightly related problems. First problem > is time keeping mechanism of Xen VMM. Not yet reading all the related codes > of Xen, I guess for PV guests (do not consider the HVM case, since it is > more complicated than PV), each time that a VCPU is scheduled in, one of the > members of VCPU structure that stores current time will be updated (see > struct vcpu_time_info), while the time precedes when the VCPU is actually is > scheduled-out is simply “lost”. This will definitely cause problems by > confusing the scheduler of the guest OS, however, hardly find documents that > elaborate this problem or analyze what will happen in deep. > > The second problem is load-balancing. Unfortunately, this is a more > complicated problem. The guest OS (for the SMP guest cases) has its own load > balancing strategy, and I think the bottom line is that it will not waste > processing time. Taking an example here, running a single thread process > (such as “dd if=/dev/zero of=./test.file bs=64 count=3200K”, it writes a > 200MB file to virtual disk) inside a two-vcpu PV guest. The process > originally running on top of single VCPU, but after a while, the VCPU is > scheduled out (blocked and waiting for the data actually write to the > virtual disk file), and another VCPU is scheduled in, the process will > migrate to the other VCPU naturally. This case will result in extra > overhead, and I got performance data to prove this. Besides understanding > the load balancing strategy of guest OS, load balancing for scheduler of the > VMM deserves also careful consideration and design. Current default > scheduler of Credit actually do not doing very well on this point. The PCPU > will steal VCPU from its neighbors once it found it is idle. Smarter > solutions are needed in later schedulers (as George had mentioned). > > 2) what types of events are ocurring during scheduling > > I think it is good for you to check the xentrace data for this problem. > > 3) xenmon problems. > > I had not used that and can not give you any suggestions, sorry for that. > > 4) Xentop > > Do not know what you mean for N/W? > > > > Best, > > Zhiyuan > > > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月10日 20:48 > *收件人:* xen-devel@lists.xensource.com > *主题:* Re: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > Some doubts. > (1) How the guest kernel scheduling strategies (linux kernel scheduling > like DEADLINE and IO schedulers in guests) affect the VM Scheduling? > > (2) I want to know what types of events are ocurring during scheduling- > which domain is scheduled and what types of tasks it is running. > > I am using xenmon- some probs. > > (1) it does not show me --iocount option correctly. it always shows > iocount= 0, although i am running dd utility in one of the guset domain. > (2) when using xentop - it does not show any N/W data for domain 0. > > am i missing something in configuration. > > > 2009/5/7 Zhiyuan Shao <zyshao.maillist@gmail.com> > > Believe your can use BCredit for this job. However, you can only set the > attributes of a domain statically, which means the scheduler can not adjust > its behavior according to the types of the work loads. > > > > Remember there is some publication in proceedings of VEE 2009, which had > addressed your problem. The solution continuously guests the performance > characteristics of the application, and adjust the preemption decisions of > the scheduler itself. You can take a look at that, but unfortunately, they > did not provide the patch behind the solution. Besides, I am not sure that > whether such solution is really helpful to solve real problems on real > hardware, since it is based its decisions on possibilities. > > > > Zhiyuan > > > ------------------------------ > > *发件人:* gaurav somani [mailto:onlineengineer@gmail.com] > *发送时间:* 2009年5月6日 17:09 > *收件人:* Zhiyuan Shao > *抄送:* xen-devel@lists.xensource.com > *主题:* Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi Zhiyuan and George, > > Thanks for the comments. > > you are right i was getting spikes with more than 100ms and some times > 400ms. > > I will be trying with the patch u suggetsed. > > one more aspect, can we classify schedulers according to applications they > run.? > > Like compute intensive on Credit and Latency intensive on sEDF or some > trade off between them (In a cluster of servers). > > > Thanks > > Gaurav > > 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com> > > Hi, > > > > Your experiment results are actually “expected”. There are several reasons > make the results happen. > > > > Firstly, Credit has long time slice (i.e., 30 millisec by default), and > this make the observed close-to 30ms RTT to happen, actually, during such > long RTT (the spike), the domain in test (domain 1 of your experiment) is > just scheduled-out (actually, I guess the spikes illustrated in your figure > 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus > (may executing the TEST program, which is cpu-hogs), and they will not give > up their 30-ms time slice if they are not preempted. > > > > Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, > it will not be BOOSTed even it do have some I/O requests to handle. This > explains why your figure1, which has little spikes in the first 150 second, > and crowed spikes follow. Since vcpu of domain 1 simply used out their > credits, and can not preempt the other running vcpus of other domains, even > they do have I/O operations to accomplish (i.e., PING). > > > > Lastly, compared with Credit, SEDF generally has short time slices, and it > has bigger chances to preempt current running vcpu when another vcpu found > it has some I/O operations to be accomplished. (actually, I donot understand > SEDF very well, but conclude these features by reading the trace data) > > > > Actually, you can use BCredit patch to eliminate the spikes, I can > guarantee that it will work very well! J And I agree with George that you > should not use PING to justify whether a scheduler is good or not. > > > > Best, > > Zhiyuan > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月5日 17:26 > *收件人:* xen-devel@lists.xensource.com > *主题:* [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > I am evaluating the scheduler behavior in xen. > > I am using Xen 3.3.0 > Dom0 and Dom1,2,3 and 4 all are opensuse 11. > I have one CPU Intensive *TEST* which has no. of arithmatic instruction in > an infinite while() loop. > i am pinging domain1 with an external machine. and noting the RTT values. > > i have below experiments > *time (s) domain state* > 0 dom0,1,2,3,4 all idle > 50 dom2 *TEST *started > 100 dom3 *TEST *started > 150 dom4 *TEST *started > 200 dom0 *TEST *started > 250 dom2 *TEST* stopped > 300 dom3 *TEST* stopped > 350 dom4 *TEST* stopped > 400 dom0 *TEST* stopped > > For these 400 seconds time, i have performed experiments with Credit and > SEDF sceduler. > the configuration is > > > Credit configuration - weight 256, cap 0 > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > > SEDF configuration - Period 10ms slice - 1.9ms > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > the results of RTT values are attached herewith. the performance of Credit > is very bad in comparison to SEDF in this scenario. > Please provide me some thought on it. > > > Thanks and Regards > > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > INDIA > > http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> > > > > > >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Sorry for unreadable subject in earlier mail ~~~~~~~~~~~~~~~~~~~~~~~~ Hi all, I mean N/W for network send and receive statistics in xentop. I also wish to know the actual credit status of each domain with time granularity like after each 30 ms. Is there anything helpful available. Thanks Gaurav 2009/5/11 Zhiyuan Shao <zyshao.maillist@gmail.com> Some points I think that may help to your doubts:> > 1) how the scheduler of guest OS and VMM cooperate together. > > Frankly, I also think this is a rather interesting topic to research, and > currently working on that. However, before putting hands on this topic, we > need some answers on several other tightly related problems. First problem > is time keeping mechanism of Xen VMM. Not yet reading all the related codes > of Xen, I guess for PV guests (do not consider the HVM case, since it is > more complicated than PV), each time that a VCPU is scheduled in, one of the > members of VCPU structure that stores current time will be updated (see > struct vcpu_time_info), while the time precedes when the VCPU is actually is > scheduled-out is simply “lost”. This will definitely cause problems by > confusing the scheduler of the guest OS, however, hardly find documents that > elaborate this problem or analyze what will happen in deep. > > The second problem is load-balancing. Unfortunately, this is a more > complicated problem. The guest OS (for the SMP guest cases) has its own load > balancing strategy, and I think the bottom line is that it will not waste > processing time. Taking an example here, running a single thread process > (such as “dd if=/dev/zero of=./test.file bs=64 count=3200K”, it writes a > 200MB file to virtual disk) inside a two-vcpu PV guest. The process > originally running on top of single VCPU, but after a while, the VCPU is > scheduled out (blocked and waiting for the data actually write to the > virtual disk file), and another VCPU is scheduled in, the process will > migrate to the other VCPU naturally. This case will result in extra > overhead, and I got performance data to prove this. Besides understanding > the load balancing strategy of guest OS, load balancing for scheduler of the > VMM deserves also careful consideration and design. Current default > scheduler of Credit actually do not doing very well on this point. The PCPU > will steal VCPU from its neighbors once it found it is idle. Smarter > solutions are needed in later schedulers (as George had mentioned). > > 2) what types of events are ocurring during scheduling > > I think it is good for you to check the xentrace data for this problem. > > 3) xenmon problems. > > I had not used that and can not give you any suggestions, sorry for that. > > 4) Xentop > > Do not know what you mean for N/W? > > > > Best, > > Zhiyuan > > > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月10日 20:48 > *收件人:* xen-devel@lists.xensource.com > *主题:* Re: 答复: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > Some doubts. > (1) How the guest kernel scheduling strategies (linux kernel scheduling > like DEADLINE and IO schedulers in guests) affect the VM Scheduling? > > (2) I want to know what types of events are ocurring during scheduling- > which domain is scheduled and what types of tasks it is running. > > I am using xenmon- some probs. > > (1) it does not show me --iocount option correctly. it always shows > iocount= 0, although i am running dd utility in one of the guset domain. > (2) when using xentop - it does not show any N/W data for domain 0. > > am i missing something in configuration. > > > 2009/5/7 Zhiyuan Shao <zyshao.maillist@gmail.com> > > Believe your can use BCredit for this job. However, you can only set the > attributes of a domain statically, which means the scheduler can not adjust > its behavior according to the types of the work loads. > > > > Remember there is some publication in proceedings of VEE 2009, which had > addressed your problem. The solution continuously guests the performance > characteristics of the application, and adjust the preemption decisions of > the scheduler itself. You can take a look at that, but unfortunately, they > did not provide the patch behind the solution. Besides, I am not sure that > whether such solution is really helpful to solve real problems on real > hardware, since it is based its decisions on possibilities. > > > > Zhiyuan > > > ------------------------------ > > *发件人:* gaurav somani [mailto:onlineengineer@gmail.com] > *发送时间:* 2009年5月6日 17:09 > *收件人:* Zhiyuan Shao > *抄送:* xen-devel@lists.xensource.com > *主题:* Re: 答复: [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi Zhiyuan and George, > > Thanks for the comments. > > you are right i was getting spikes with more than 100ms and some times > 400ms. > > I will be trying with the patch u suggetsed. > > one more aspect, can we classify schedulers according to applications they > run.? > > Like compute intensive on Credit and Latency intensive on sEDF or some > trade off between them (In a cluster of servers). > > > Thanks > > Gaurav > > 2009/5/6 Zhiyuan Shao <zyshao.maillist@gmail.com> > > Hi, > > > > Your experiment results are actually “expected”. There are several reasons > make the results happen. > > > > Firstly, Credit has long time slice (i.e., 30 millisec by default), and > this make the observed close-to 30ms RTT to happen, actually, during such > long RTT (the spike), the domain in test (domain 1 of your experiment) is > just scheduled-out (actually, I guess the spikes illustrated in your figure > 1 should read higher than 30ms!). At the same time, the scheduled-in vcpus > (may executing the TEST program, which is cpu-hogs), and they will not give > up their 30-ms time slice if they are not preempted. > > > > Secondly, as regulated by Credit scheduler, if a VCPU has no more credits, > it will not be BOOSTed even it do have some I/O requests to handle. This > explains why your figure1, which has little spikes in the first 150 second, > and crowed spikes follow. Since vcpu of domain 1 simply used out their > credits, and can not preempt the other running vcpus of other domains, even > they do have I/O operations to accomplish (i.e., PING). > > > > Lastly, compared with Credit, SEDF generally has short time slices, and it > has bigger chances to preempt current running vcpu when another vcpu found > it has some I/O operations to be accomplished. (actually, I donot understand > SEDF very well, but conclude these features by reading the trace data) > > > > Actually, you can use BCredit patch to eliminate the spikes, I can > guarantee that it will work very well! J And I agree with George that you > should not use PING to justify whether a scheduler is good or not. > > > > Best, > > Zhiyuan > > > > > ------------------------------ > > *发件人:* xen-devel-bounces@lists.xensource.com [mailto: > xen-devel-bounces@lists.xensource.com] *代表 *gaurav somani > *发送时间:* 2009年5月5日 17:26 > *收件人:* xen-devel@lists.xensource.com > *主题:* [Xen-devel] Credit scheduler vs SEDF scheduler > > > > Hi list, > > I am evaluating the scheduler behavior in xen. > > I am using Xen 3.3.0 > Dom0 and Dom1,2,3 and 4 all are opensuse 11. > I have one CPU Intensive *TEST* which has no. of arithmatic instruction in > an infinite while() loop. > i am pinging domain1 with an external machine. and noting the RTT values. > > i have below experiments > *time (s) domain state* > 0 dom0,1,2,3,4 all idle > 50 dom2 *TEST *started > 100 dom3 *TEST *started > 150 dom4 *TEST *started > 200 dom0 *TEST *started > 250 dom2 *TEST* stopped > 300 dom3 *TEST* stopped > 350 dom4 *TEST* stopped > 400 dom0 *TEST* stopped > > For these 400 seconds time, i have performed experiments with Credit and > SEDF sceduler. > the configuration is > > > Credit configuration - weight 256, cap 0 > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > > SEDF configuration - Period 10ms slice - 1.9ms > *Domain VCPU * > 0 2 > 1 2 > 2 2 > 3 2 > 4 2 > all vcpu0s are pinned to pcpu0 and vcpu1s to pcpu1. > > the results of RTT values are attached herewith. the performance of Credit > is very bad in comparison to SEDF in this scenario. > Please provide me some thought on it. > > > Thanks and Regards > > Gaurav somani > M.Tech (ICT) > Dhribhai Ambani Institute of ICT, > INDIA > > http://dcomp.daiict.ac.in/~gaurav <http://dcomp.daiict.ac.in/%7Egaurav> > > > > > >-- Gaurav somani M.Tech (ICT) Dhribhai Ambani Institute of ICT, Gandhinagar Gujarat, INDIA onlineengineer@ieee.org gaurav_somani@daiict.ac.in http://dcomp.daiict.ac.in/~gaurav http://www.linkedin.com/in/gauravsomani _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel