Posts by wenjing wu
log in
1) Message boards : News : cas@home project status (Message 18699)
Posted 601 days ago by wenjing wu
The main application running on CAS@home is the TreeThreader which predicts protein structure. After finishing a big campaign in Oct 2015, there have been sporadic TreeThreader jobs running on CAS@home which is around 200 sequences to predict every month. Currently, the TreeThreader application provides a public service FALCON to allow worldwide biologists submitting their protein sequences for structure prediction, and CAS@home is the backend computing platform for this public service. This service has just been open to the public, and it expects more biologists to use it in the near future with a recent publication. Apart from the jobs from FALCON, there will also be another campaign of TreeThreader between Feb and May 2016 to prepare for the CASP event . As usual, CAS@home appreciates all the support from its volunteers!
2) Message boards : News : LHC@home Test4Theory passed 2 Trillion events (Message 18627)
Posted 818 days ago by wenjing wu
The Test4Theory application of LHC@home just announced that they have reached the total of 2 Trillion events simulated since 2011 by volunteers for the CERN PH-TH project MCPlots.

The original news can be found here
3) Message boards : News : CAS@home maintenance period (Message 18572)
Posted 1157 days ago by wenjing wu
IHEP Computer Center enters the annual maintenance period starting from 25th July, and finishes at 30th July. The CAS@home server will be shutdown during this period!
Thanks for your understanding!
4) Message boards : News : CAS@home computing schedule (Message 1223)
Posted 1320 days ago by wenjing wu
Thanks to all the generous contribution from our volunteers, TreeThreader has finished its first computing campaign of 50K protein sequences. Very important scientific conclusion has been drawn from the results and a new paper is in the process of writing. Official results will be published later on CAS@home news.A new campaign of TreeThreader will be launched soon with a task of 80K sequences to be predicted!

感谢各位志愿者的热心帮助,TreeThreader已经成功完成了第一批的计算任务,成功破解了5万条新蛋白质序列的结构。基于这些计算结果,生物学家在利用海藻提炼石油的研究方面已经得出重要的科学结论,目前正在撰写相关论文。CAS@home后续会继续公布正式的科学结论。TreeThreader后续将有更大的计算任务(约8万条蛋白质序列),目前计算任务正在部署中。
5) Message boards : News : TreeThreader Job Submission Error (Message 1165)
Posted 1521 days ago by wenjing wu
Due to network problems between ICT and IHEP, the TreeThreader job submission service keeps failing. In order to diagnose this problem, the job submission service will be stopped, hence no jobs will be generated during this period. We will send notifications once the service is back! Sorry for the inconvenience!

Updates:
ICT has started to slowly submit jobs from last week to test the stability! (28th Aug, 2013)


由于网络故障问题,TreeThreader 的作业提交系统不断出现失败,为了诊断故障,我们决定暂时中止TreeThreader的作业提交服务,因此这段时间内CAS@home上将没有可领取的作业。我们会在解决问题后,及时通知!带来不便,敬请谅解!

更新:
计算所的前端作业服务器开始小规模地提交作业,测试系统的稳定性,一切还在恢复过程中!(2013-08-28)
6) Message boards : News : Deadline for TreeThreader (Message 1141)
Posted 1539 days ago by wenjing wu
这些需要在BOINC的客户端支持。目前,CAS维护团队没有支持BOINC客户端方面的开发,都是志愿者从BOINC官方下载的客户端。所以,对于此类建议,可以直接跟BOINC开发人员联系。我们也乐意在BOINC的workshop上转提这些建议!
7) Message boards : Number crunching : 到底是ThreeThreader还是TreeThreader啊? (Message 1138)
Posted 1542 days ago by wenjing wu
Thanks! I will fix that!
8) Message boards : News : Deadline for TreeThreader (Message 1132)
Posted 1553 days ago by wenjing wu
这个建议很好,目前boinc的scheduler并没有支持这么细的调度策略。我可以向开发人员反应!谢谢!
此外,CAS没有发布任何Mac上的应用程序!
我倒是觉得可以根据主机每天的平均计算能力来分配不同大小的任务包。不是每个志愿者都每天长时间开机全力运算的,很多人应该都是实际自己需要使用计算机时分配一定的计算资源给BONIC,这个计算资源还会被其他项目分享,所以实际上算力应该是不怎么强的,大任务包很多人可能都没有办法按时完成。像我的E350处理器上网本,根本就不会考虑现在的CAS任务。所以建议做一个程序,统计一下每台机器在一个月内真实提供了多少计算能力,根据这个时间去调整任务包大小和时限,应该会对志愿者的机器利用得充分一些。另外,OSX系统好像收不到任务包呢?

9) Message boards : News : TreeThreader Job Submission server glitch (Message 1111)
Posted 1585 days ago by wenjing wu
TT的数据是缓存在客户端的,我们对调度器进行了优化,主机只认领需要已经缓存数据的任务,因此在计算过程中的网络流量并不大。一般每个主机只需要一次性下载100-200MB左右的数据,就可以持续不断进行TreeThreader的计算任务。
支持CAS。 半夜还要忙乎。

网络服务器的压力不是一般的大。鸭梨山大。

40MB一个任务,1000个任务也就要约40GB的数据量。

要是一个任务包能优化一下,一个包10MB以内能计算1~2个小时,任务多了也不怕。呵呵

10) Message boards : News : TreeThreader Job Submission server glitch (Message 1110)
Posted 1585 days ago by wenjing wu
TT的数据是缓存在客户端的,我们对调度器进行了优化,主机只认领需要已经缓存数据的任务,因此在计算过程中的网络流量并不大。一般每个主机只需要一次性下载100-200MB左右的数据,就可以持续不断进行TreeThreader的计算任务。
支持CAS。 半夜还要忙乎。

网络服务器的压力不是一般的大。鸭梨山大。

40MB一个任务,1000个任务也就要约40GB的数据量。

要是一个任务包能优化一下,一个包10MB以内能计算1~2个小时,任务多了也不怕。呵呵

11) Message boards : News : TreeThreader Job Submission server glitch (Message 1104)
Posted 1586 days ago by wenjing wu
There is a network glitch on the TreeThreader Job Submission server, so no new jobs will be submitted to CAS@home before this is fixed! We will keep you updated of the situation from here! Sorry for the inconvenience!

Update: The Job Submission Service is back!

TreeThreader的作业发送服务器出现网络故障,目前正在抢修。在此之前,新作业的发送将被停止,我们将再此提供最新的更新。带来不便,请见谅!

更新: TreeThreader作业提交系统已经恢复正常!
12) Message boards : News : TreeThreader Calls for more Computing Power (Message 1101)
Posted 1586 days ago by wenjing wu
ICT在考虑TreeThreader程序第三版的开发,会把这个建议转发给开发人员!
能否开发GPU能用的程序呢,毕竟GPU算蛋白质折叠比CPU快太多了。

同意!!!

13) Message boards : Number crunching : OpenMM (Message 1091)
Posted 1595 days ago by wenjing wu
No, Current applications on CAS@home do not support it!
CAS@Home support OpenMM ?

https://simtk.org/xml/index.xml

@+

*_*

14) Message boards : News : TreeThreader Calls for more Computing Power (Message 1085)
Posted 1601 days ago by wenjing wu
这个是不同平台上的结果有差别,当时的validator认为结果是无效的,我们已经解决这个问题了,并且对结果进行了重新验证与积分授予!
最近好像无效结果比较多啊,192个结果,7个无效,有时候算了很久的,结果是无效结果,看着很郁闷啊。不知道是怎么回事。

15) Message boards : Number crunching : Ridiculous validation (Message 1078)
Posted 1602 days ago by wenjing wu
The inconsistency problem on different platform is fixed! And the credit is re-granted to the mistreated Linux hosts!
16) Message boards : Number crunching : Work unit pegged at 100% CPU, 64 bit Debian (Message 1076)
Posted 1602 days ago by wenjing wu
Another thought: Maybe your host downloaded more jobs than the number of your CPU cores, so when you suspend one job, anther waiting job kicks in and use the same core and keeps the core occupied.
On the server side, we allow clients download more jobs than the number of the CPU cores as more users prefer to download more and run them while the hosts get offline.
If you want to free the CPU cores, either you can specify less CPU percentage for BOINC to use from the GUI, or suspend the whole project.

I detached the machine from the project, and reattached. Three more tasks downloaded, and each used 100% of a core.

I've suspended the project and killed the tasks.

What should I do?

17) Message boards : Number crunching : Work unit pegged at 100% CPU, 64 bit Debian (Message 1075)
Posted 1602 days ago by wenjing wu
This used to be known as a bug in the BOINC wrapper which controls the running of application, we fixed this bug a long time ago. I tried on a Red Hat Linux machine with "suspend" and "resume", and it works well on it. What I can suggest is to use the most recent BOINC client to see how it works.


I detached the machine from the project, and reattached. Three more tasks downloaded, and each used 100% of a core.

I've suspended the project and killed the tasks.

What should I do?

18) Message boards : Number crunching : Open checkpoint file error? (Message 1074)
Posted 1602 days ago by wenjing wu
I checked the error jobs on your host. Actually the "checkpoint file open" error is an ignorable error which happens to every job. I think something else on your computer terminated these jobs.

I've been getting a few of these [url=http://casathome.ihep.ac.cn/results.php?
userid=1173&offset=0&show_names=0&state=6&appid=]errors[/url] recently on my host 26742 running the ICT tasks. Here's a typical stderr output:

<core_client_version>7.0.64</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code 195 (0xc3)
</message>
<stderr_txt>
wrapper: starting
15:10:11 (5928): wrapper: running run-treeThreader.exe ()
Warning: Open checkpoint file error.
app exit status: 0x1
16:27:23 (5928): called boinc_finish

</stderr_txt>
]]>

19) Message boards : Number crunching : Job running when it shouldn't be. (Message 1073)
Posted 1602 days ago by wenjing wu
This used to be a bug in the wrapper, but we fixed it a long time ago.. please update your BOINC core client and Manager to the most recent one!

This wu is on my system, (Windows XP), at the moment. BOINC Manager shows it as "Waiting to run", yet Windows App Manager shows it running. Suspending the job does not stop it either.

20) Message boards : Number crunching : Ridiculous validation (Message 1072)
Posted 1602 days ago by wenjing wu
It is true, the inconsistence is from results being generated on different platforms.
I will do something to fix it and re-grant the credit if possible!

Yeah, same here.
A closer inspection of your and my results gives the conclusion that Linux isn't validating against Windows - very bad...



Next 20