Re: [PATCH v3] migration/throttle: Add cpu-throttle-tailslow migration parameter

2020-05-07 Thread Dr. David Alan Gilbert
* Keqian Zhu (zhukeqi...@huawei.com) wrote: > At the tail stage of throttling, the Guest is very sensitive to > CPU percentage while the @cpu-throttle-increment is excessive > usually at tail stage. > > If this parameter is true, we will compute the ideal CPU percentage > used by the Guest, which

Re: [PATCH v3] migration/throttle: Add cpu-throttle-tailslow migration parameter

2020-05-07 Thread zhukeqian
Hi Dr.David, Sorry for the reply delay, just come back from holiday. On 2020/4/30 22:12, Dr. David Alan Gilbert wrote: > * Keqian Zhu (zhukeqi...@huawei.com) wrote: >> At the tail stage of throttling, the Guest is very sensitive to >> CPU percentage while the @cpu-throttle-increment is excessive

Re: [PATCH v3] migration/throttle: Add cpu-throttle-tailslow migration parameter

2020-04-30 Thread Dr. David Alan Gilbert
* Keqian Zhu (zhukeqi...@huawei.com) wrote: > At the tail stage of throttling, the Guest is very sensitive to > CPU percentage while the @cpu-throttle-increment is excessive > usually at tail stage. > > If this parameter is true, we will compute the ideal CPU percentage > used by the Guest, which

[PATCH v3] migration/throttle: Add cpu-throttle-tailslow migration parameter

2020-04-13 Thread Keqian Zhu
At the tail stage of throttling, the Guest is very sensitive to CPU percentage while the @cpu-throttle-increment is excessive usually at tail stage. If this parameter is true, we will compute the ideal CPU percentage used by the Guest, which may exactly make the dirty rate match the dirty rate