找回密码
 新注册用户
搜索
查看: 5029|回复: 8

[新闻] Temporary server Things are getting better!(新增了7台服务器)

[复制链接]
发表于 2020-3-15 07:04:13 | 显示全部楼层 |阅读模式
本帖最后由 金鹏 于 2020-3-31 18:47 编辑

服务器在原来的 140.163.4.241   、 140.163.4.231  、  155.247.164.213  、 128.252.203.10四组基础上,再新增 13.90.152.57   ,40.114.52.201        ,37.187.12.48  和  128.252.203.2 及 13.82.98.119 ,128.252.203.4  ,52.224.109.74  等七组服务器
Things are getting better!
by bruce » Wed Mar 18, 2020 4:33 am
[size=1.3em]For most of the past few days, I've had conciderable idle resources ... trying to get new assignments. I know my system is configured correctly and it's just the unprecedented loads being put on FAH's servers by the sudden spike in the number of Donors (people) trying to help.

After being AFK for 24 hrs for this machine, I found it had crashed so I restarted it.

This machine:
Two GPUs plus two CPU slots.

Status after the restart:
CPU slots resumed processing WUs from previous checkpoints.
GPU slots started trying do download new assignments.
... followed by repeated messages like these with various values of xx.
18:42:11:WU01:FS02:Connecting to xx.xx.xx.xx:8080
18:42:12:WARNING:WU01:FS02:Failed to get assignment from 'xx.xx.xx.xx:8080': No WUs available for this configuration
but after 15 to 20 minutes, workable values for xx were found and both GPU slots started folding.

yes, 15 to 20 minutes is longer than folks will be happy with, but it's a lot better than the 15 to 20 hours i have seen over the past few days.

A big attaboy and Congratulations to all you FAH.org guys & gals who have been frantically working to get server capacity on-line and WUs that can be distributed by them.

===================================================

由于包括GPU云的大户和大量新增的散户造成算力暴增昨晚一夜之间就消化掉了服务器上各类WU任务包

斯坦福会尽快解决供不应求问题,目前正在制造足够多的WU任务包,任务包上架前但必须先使服务器脱机状态,所以静候佳音耐心等待

最新更新:新冠包已经恢复供应,新冠包已经恢复供应,新冠包已经恢复供应



Re: Temporary server outages
by toTOW » Mon Mar 16, 2020 8:28 am
[size=1.3em]The servers have WUs to distribute, but you guys are so enthusiastic that they can keep up with the pace ...

Please be patient.
Re: Temporary server outages
by toTOW » Sun Mar 15, 2020 9:40 am
[size=1.3em]50k WUs have been added to 140.163.4.231 / plfah1-1.mskcc.org and 60k more are being generated on 140.163.4.241 / plfah2-1.mskcc.org.

Work generation has higher priority than WUs distribution, so you'll see the new work units flowing from these servers as soon as generation is finished.
Temporary server outages
by bruce » Sun Mar 15, 2020 3:05 am
[size=1.3em]We've been overwhelmed with the support we're getting from new Donors.

Several servers ran out of WU's overnight. The issue is being addressed as fast as the project's owner can get to it. Unfortunately the server(s) may have to be taken off-line while the new WUs are being generated.



查看服务器状态   https://apps.foldingathome.org/serverstats


回复

使用道具 举报

 楼主| 发表于 2020-3-15 13:22:40 | 显示全部楼层
本帖最后由 金鹏 于 2020-3-15 13:35 编辑

貌似另外两个项目组也会上线基于core22内核的新冠包(P11753-11764),期待能缓解供需矛盾

服务器在原来的 140.163.4.241    和  140.163.4.231  基础上新增了  155.247.164.213   和  128.252.203.10两组

捕获.PNG
回复

使用道具 举报

发表于 2020-3-15 19:42:34 | 显示全部楼层
没用,还是一下子就吃完了……
回复

使用道具 举报

发表于 2020-3-15 23:49:57 | 显示全部楼层
看了foldingforum的帖子,是说平时每小时签发4000个包,爆发的那天上升到了每小时签发27000个包。现在已经达到了15.4万个包了。
我觉得这个可以考虑上云了。
发包频率 154K per hour.png

评分

参与人数 1基本分 +20 收起 理由
金鹏 + 20 项目组等着捐赠改善

查看全部评分

回复

使用道具 举报

发表于 2020-3-16 02:57:51 | 显示全部楼层
下午跑了两个包,然后又没了...估计别人也不缺我这点微小的算力吧
回复

使用道具 举报

发表于 2020-3-17 15:29:56 | 显示全部楼层
顶顶顶,新鲜血液加入,哈哈

评分

参与人数 2基本分 +28 收起 理由
Keyco + 8 赞一个!
金鹏 + 20 赞一个!

查看全部评分

回复

使用道具 举报

发表于 2020-3-18 09:13:24 | 显示全部楼层
我970也跑了一阵子了,希望能贡献一份力量

评分

参与人数 1基本分 +20 收起 理由
金鹏 + 20 赞一个!

查看全部评分

回复

使用道具 举报

您需要登录后才可以回帖 登录 | 新注册用户

本版积分规则

论坛官方淘宝店开业啦~

Archiver|手机版|小黑屋|中国分布式计算总站 ( 沪ICP备05042587号 )

GMT+8, 2024-4-18 13:35

Powered by Discuz! X3.5

© 2001-2024 Discuz! Team.

快速回复 返回顶部 返回列表