2010/4/23 Takuya Yoshikawa <yoshikawa.takuya@xxxxxxxxxxxxx>: > (2010/04/23 19:28), Avi Kivity wrote: > >>> >>> OK, I will do in the next version. In this RFC, I would be happy if I can >>> know the overall design is right or not. >>> >> >> Everything looks reasonable to me. >> > > Thank you! > >> Do you have performance numbers? I'm interested in both measurements of >> KVM_SWITCH_DIRTY_LOG under various conditions and macro benchmarks (for >> example, total guest throughput improvement under Kemari). >> > > Currently, I'm just checking the performance visually. > - on laptops, we can feel the speed directly: my favorite is installing > ubuntu or debian by alternate installers > > - live migration with heavy work load > > Now that I've got the overall design, I want to measure the performance by > numbers. > > > About performance under Kemari: we(oss.ntt.co.jp staffs: me and Fernando) > are > now concentrating on improving the basic live-migration infrastructures and > they(lab.ntt.co.jp staffs) are working hard for building Kemari itself. > > - We are also interested in using live-migration with HA software and this > needs > light, stable live-migration: same as Kemari! > > > So measuring Kemari's performance with our patch-set will need some more > time, > ** How about Tamura-san? ** You can measure your patch quite easy. 1. Run a simple program that malloc huge size (2GB for example), then go into infinite loop that touches each in a really bad manner, like only touch odd page numbers. 2. Start live migration which usually won't finish. Similar with Kemari. 3. Measure the response time of your ioctl(). I used 1 and 2 to measure dirty physmap speed up in QEMU. Thanks, Yoshi > So at this stage, I'll show you performance improvement(I hope to improve) > by > another test cases. > > If possible next week! > > Thanks, > Takuya > > -- > To unsubscribe from this list: send the line "unsubscribe kvm" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- To unsubscribe from this list: send the line "unsubscribe kvm-ia64" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
