Hi Yinghao,
Here are some detail information provided by Yihao.
1. Implement the lastest version of CPU SLI feature (Sevice Level Index)
for rich container;
this new version show great advantages in performance and accuracy.
The CPU SLI features isolate CPU runtime metrics for rich container
users. It allows users
to use commands like `top` or read pseduo files like /proc/loadavg
in a container, and get
the expected information, just as if they do in a traditional
virtual machine.
CPU SLI features have evolved several versions. The latest version
overcame left bugs, and
it achieved both great stability and accuracy.
For now it covers these metrics.
* CPU usage (usr%, nice%, sys%, idle%, steal%, iowait%, irq%,
softirq%, guest%)
* 1/5/15 load accounting
* 1/5/15 running load accounting (meaning uninterruptible excluded)
* nr_running, nr_uninterruptible, nr_migration
* sched latency histogram
2. Design and implement cpu isolation solution for function computing
running on Daishu, which
enable 2000 microvm running on one machine which acceptable overhead
(< 1%).
Dynamic load tracking contributes most of the overhead, especially
considering that enormous
2000 microvms expand a both wide and deep group scheduling tree.
Yihao managed to minimalize
both cgroup hierarchy and dynamic load tracking, to limit the
overhead of scheduling those
microvms to less than 1%.
3. The core develper in plugsched project, which aims to upgrade linux
kernel scheduler subsystem online.
more details please visit project repo:
https://gitee.com/anolis/plugsched
4. Backport cpuidle-haltpoll feature to ANCK and optimize performance
for virtual machine.
Guest haltpoll is a relatively novel technique (at that time Yihao
backports it). And he made it
enabled widely both on Aliyun ECS and in Alibaba Corp.
Also on Aliyun ECS, he oberserved and tackled a poorly performed
locking problem, that qspinlock
was not enabled at all. This problem dates back long. At that time
he solved this problem. The same
problem still persists on Amazon EC2.
5. Proficient in problem analysis, be skilled using gdb, crash, pyGDB,
qemu ...
https://lore.kernel.org/all/20220428103726.108597-1-wuyihao@linux.alibaba.c…
-- csp
在 2022/5/18 上午11:38, 谢英豪 写道:
congrats .expecting more informative materaial from
you。
------------------------------------------------------------------------
谢英豪
18512581128
苏宁易购集团股份有限公司
IT总部苏宁云BU基础云研发中心Iaas研发中心计算及内核研发部
本邮件含有受到法律保护的不可公开内容。如您并非此邮件的指定收件人,请立刻将之及所有复本从系统中删除,严禁复制、传送、公开、使用、分发或者披露其所载的全部或部份资料。发件人对资料传送的任何错误或遗漏概不负责。
------------------------------------------------------------------------
----邮件原文----
发件人:Xunlei Pang
发件时间:2022-05-18 10:29:30
收件人:Shanpei Chen <shanpeic(a)linux.alibaba.com>,cloud-kernel
<cloud-kernel(a)lists.openanolis.cn>,Michael Wang
<yun.wang(a)linux.alibaba.com>,Artie Ding
<artie.ding(a)linux.alibaba.com>,Gang Deng
<gavin.dg(a)linux.alibaba.com>,Yihao Wu <wuyihao(a)linux.alibaba.com>
主题 :[ck] Re: Add Yihao Wu as scheduler's maintainer
在 2022/5/16 下午6:08, Shanpei Chen 写道:
Hi, all
I'm glad to invide Yihao Wu as maintainer
of ANCK scheduler.
Yihao is working on linux kernel scheduler for 5 years, and show great
interest in kernel technologies.
Here are some highlight contributions from Yihao:
1. Implement the lastest version of CPU SLI feature (Sevice Level
Index)
for rich container;
this new version show great advantages in performance and accuracy.
2. Design and implement cpu isolation solution for function computing
running on Daishu, which
enable 2000 microvm running on one machine which acceptable overhead
(< 1%).
3. The core develper in plugsched project, which aims to upgrade linux
kernel scheduler subsystem online.
more details please visit project repo:
https://gitee.com/anolis/plugsched
4. Backport cpuidle-haltpoll feature to ANCK and optimize performance
for virtual machine.
5. Proficient in problem analysis, be skilled using gdb, crash, pyGDB,
qemu ...
https://lore.kernel.org/all/20220428103726.108597-1-wuyihao@linux.alibaba.c…
Working with Yihao is efficient and have a sense of achievement. Let us
hope it happens soon.
-- csp
Yihao has been working on scheduler for many years, he strikes me as a
very careful and skilled kernel geek, I believe he will do even better
under this new responsiblity:
Acked-by: Xunlei Pang
_______________________________________________
Cloud Kernel mailing list -- cloud-kernel(a)lists.openanolis.cn
To unsubscribe send an email to cloud-kernel-leave(a)lists.openanolis.cn