[BOINC] [生命科学类及其他] World Community Grid

vmzy

Filesystem Error
Oct 13, 2010 8:22:30 PM
We just experienced an error with the BOINC filesystem. As a result, the operating system re-mounted it in read-only mounted. We are now in the process of taking it offline and running an fsck on the filesystem. While that runs BOINC will be offline and not able to send or receive files.

Oct 13, 2010 10:21:58 PM
The fsck is likely to take many hours to run.

Oct 13, 2010 11:47:12 PM
We have modified the server to return a 503 error with a one hour retry for all attempts to download files.

Oct 14, 2010 12:21:52 AM
The fsck is likely to take a long while (many hours). We are still in phase 1 of fsck (out of 5).

Oct 14, 2010 11:21:45 AM
We have brought the filesystem online but we aren't sure what condition it is in. Instead of making it available, we have made it read-only and we are copying the data to our new filesystem that we were in the process of creating. The copy is now occurring but it will still be a number of hours to be available.

Oct 14, 2010 11:32:35 AM
Copy is 14% complete. I will update this count periodically.

Oct 14, 2010 12:20:20 PM
21.8% copied

Oct 14, 2010 1:02:50 PM
32.3% copied.

Oct 14, 2010 1:02:50 PM
We have two sets of copies running. One for the workunit input files (download files) and one for the result files (upload). The upload is running faster than the download. As a result, we might be able to bring the uploads online before the downloads are available.

Oct 14, 2010 2:13:18 PM
Overall 50.5% copied
Upload 81.5% copied

Oct 14, 2010 3:00:44 PM
Uploads are back online and handling the surge easily.

Overall copying is at 61.2%. We are watching to see how fast this progresses now that the upload copies are done.

Oct 14, 2010 4:23:17 PM
The new filesystem is supporting the uploads just fine so that looks good.

We are at 77.2% complete for the overall copy.

When the copy finishes we will need to check a few things out and then we will open up the scheduler and allow new work to be issues and file downloads to occur.

Oct 14, 2010 5:58:02 PM
We are at 94.8% complete.

Oct 14, 2010 6:32:47 PM
We are at 101.2% complete.

The denominator in my % complete estimate doesn't include the additional data that has been uploaded to the servers since we re-enabled the uploads. So we are close - but I don't know exactly how close we are.

Oct 14, 2010 7:06:29 PM
Sorry folks - it is going to be a bit longer. The block size is different between the two filesystems so the amount of data copied was being shown differently. As a result, my % complete computations were incorrect. We still have some distance to go. I'll work on a better way to inform you of the % complete.

Oct 14, 2010 8:11:12 PM
Ok - copies are done. We have a bit of work to do to reconfigure things and we have stopped uploads again while we do this work.

Once we are done with the reconfiguration we should be able to bring everything back online.

Oct 14, 2010 8:54:08 PM
We have finished configuring things on the backend. Work is flowing to the members.

At the moment the servers are being hit hard and you might get some connection errors. They will recover over the next 30 minutes or so.

However, we know that some number of files have been corrupted. BOINC will download these files and verify them with the signature created before the filesystem was corrupted. If the file does not match the signature, it will report the result as an error. We expect to see some of these over the next week or two. We ask your patience as we work through these errors and get the workunits reloaded with fixed files.

大意：
BOINC服务器的文件系统挂了，正在恢复中，估计需要很多小时。

vmzy

Nov 6, 2010 1:05:04 PM
World Community Grid volunteers donate over 400,000 years to humanitarian research!

We are very pleased to announce that World Community Grid volunteers have donated more than 400,000 years of computer run time.

To see the statistics, please click here. To see how that amount was distributed among the research projects, please click here.

The amount of run time provided continues to accelerate. Here are the run time milestones since World Community Grid was launched in November of 2004:
* 100,000 years took 32 months and was achieved in July of 2007,
* 200,000 years took 17 months and was achieved in November of 2008,
* 300,000 years took 13 months and was achieved in December of 2009,
* 400,000 years took 11 months and was achieved in November of 2010.

We look forward to reaching the 500,000 milestone prior to November of 2011.

A hardy THANK YOU to all the members who have helped achieve this milestone!
大意：
项目里程碑，总计算时间达到40万年。

vmzy

Nov 8, 2010 6:41:27 PM
Launch of the Windows and Mac Versions of The Clean Energy Project - Phase 2

World Community Grid is pleased to announce the launch of the Windows and Mac version of The Clean Energy Project - Phase 2 (CEP2). The Linux version was released in June, 2010. This project is provided by the Department of Chemistry and Chemical Biology, Harvard University in Cambridge, Massachusetts, USA and hopes to find promising new materials for organic solar cells.

For additional information on the release of the Windows and Mac version of this project, please click here .
大意：
cep2，发布 Windows 和Mac版计算内核。Linux 版内核已于6月发布了。

vmzy

Nov 16, 2010 2:12:41 PM
World Community Grid is 6 years old!
World Community Grid is 6 years old today! Many thanks go to the members for their continued participation and contribution.

There have been many accomplishments during the past 12 months. Please read about some of them in the special News & Updates article here .

Thank you for your past and future contribution to help accelerate scientific research to help humanity!
大意：
今日是WCG 6周年，感谢大家为全人类做出的贡献。

vmzy

Reduced HFCC project priority
Jan 25, 2011 7:04:01 PM
Due to an outage of one of the Chiba server's we've reduced the feeder priority for HFCC to prevent a backlog of results. Users with only HFCC selected should be able to continue to get HFCC work units, but users with multiple projects selected will see a reduction in the number of HFCC work units downloaded.

Seippel

大意：
因为Chiba 的服务器挂了一台。为了防止任务结果囤积。我们调低了HFCC的项目发包优先级。参加多子项目的用户可能会接到很少的HFCC了。

Feb 1, 2011 3:16:51 AM
HFCC has now been restored to it's previous feeder priority.
大意：
HFCC的发包优先级恢复正常。

vmzy

Mar 31, 2011 6:45:30 PM
Prepping for version 6.42 of Help Conquer Cancer
Greetings all,

We are going to be updating the application version of Help Conquer Cancer to version 6.42. This will require us to run the project down to no results before releasing the new application. What this means is that over the next week, members will notice that work for this application will decrease until all the active batches have complete. Within the next 36-48 hours there should be new work for all to catch. After this only resends will be sent out. Once this happens will will start to over schedule to help clear the work units faster. In about 1 week we will be able to start the new application up and have work units available to all members. We will try to keep members posted on the progress of this change.

Please note that your computer may go idle if you only have Help Conquer Cancer selected as your only project.

(Please note that all times are estimates and may vary a bit.)

If you have any questions, please feel free to ask :)

Thanks,
-Uplinger
大意：
准备发布Help Conquer Cancer新版6.42程序，为此我们要先把当前任务全算完。
故即将停止新任务发放，以便尽快把老任务算完。所以下周HCC可能会断粮。
等老任务全算完，我们会发布新程序和任务。
Apr 8, 2011 4:53:08 PM
Updated Windows Help Conquer Cancer Application (v 6.42)
Greetings all,

We have successfully completed the testing and prepping of the system for the updated version of Help Conquer Cancer. This application update will decrease the runtime required for work units to complete. The change is only for Windows and will show up as version 6.42.

Thanks,
-Uplinger
大意：
HCC6.42windows版计算程序发布，这个版本减少了计算时间。

zglloo

2011年五月18 日

Lay Person Abstract:

The postprocessing of the information computed using World Community Grid about the structures of 62,000 rice proteins requires a computationally intensive process, called clustering. This clustering computation determines the most probable structure for a given protein. 100,000 structure predictions were computed for each protein. The clustering algorithm finds the most likely structure for the protein based on which and how many of the 100,000 predicted structures most closely match each other. To dramatically speed up this postprocessing step, the researchers developed a way to take advantage of Graphical Processing Units (GPU's), used in many new computers to speed up processing required to render images for games and other entertainment. This paper describes the approach they used to speed up the postprocessing of the results of the Nutritious Rice for the World project. The scientists were able to increase the speed of postprocessing by over a factor of 260, which will help them accelerate the next steps of their research.

Technical Abstract:

Calculation of the root mean square deviation (RMSD) between the atomic coordinates of two optimally superposed structures is a basic component of structural comparison techniques. We describe a quaternion based method, GPU-Q-J, that is stable with single precision calculations and suitable for graphics processor units (GPUs). The application was implemented on an ATI 4770 graphics card in C/C++ and Brook+ in Linux. The Nutritious Rice for the World Project (NRW) on World Community Grid predicted de novo, the structures of over 62,000 small proteins and protein domains returning a total of 10 billion candidate structures. Clustering ensembles of structures on this scale requires calculation of large similarity matrices consisting of RMSDs between each pair of structures in the set. As a real-world test, we calculated the matrices for 6 different ensembles from NRW. The GPU method was 260 times faster that the fastest existing CPU based method and over 500 times faster than the method that had been previously used. GPU-Q-J is a significant advance over previous CPU methods. It relieves a major bottleneck in the clustering of large numbers of structures for NRW. It also has applications in structure comparison methods that involve multiple superposition and RMSD determination steps, particularly when such methods are applied on a proteome and genome wide scale.

大意是：
水稻项目开始使用 GPU 计算了，不过只是在服务端进行验证，用来完成对数据的后期处理，可以比原来快 260 倍。

vmzy

Jul 11, 2011 7:33:04 PM
Reduced CMD2 project priority
Due to an increase in the backlog of results for CMD2, we've reduced the feeder priority for this project to allow the backend processing for results to catch up. Users with only CMD2 selected should be able to continue to get CMD2 work units, but users with multiple projects selected will see a reduction in the number of CMD2 work units downloaded.

Seippel
大意：
由于任务结果处理不过来，所以下调HCMD2任务的发包优先级。

panda7456

回复 173# vmzy

这个项目的新闻在哪看？

vmzy

Jul 18, 2011 7:35:55 PM
Adding Additional Servers to Support Growth
We are adding three new servers to our infrastructure that will support our continued growth. For the majority of the World Community Grid volunteers, this change will be transparent and no action is required.

As a result of this change, the World Community Grid software will communicate with these three new IP addresses. These new IP addresses are listed below. The firewall on some volunteer's computers will prompt you to allow the World Community Grid software to communicate with these new IP addresses. Please allow the software to do so.

The new IP addresses are:

170.224.194.69
170.224.160.205
170.225.97.195

We thank you for your continued participation and support.
大意：
为了提高服务器负载能力，我们新加了几台服务器，如果您装有防火墙请把新服务器ip加入白名单。

vmzy

Jul 22, 2011 10:20:26 PM
New Beta Test
We have released new work units for beta test. We are sending out 556 work units. These will be sent out with a quorum of two. So 1112 results will be sent out.

The memory requirements will be around 400MB for this test. Disk usage is very small, less than 10 MB. Uploads are about 100KB and downloads are about 300KB.

Please let us know in the other forums if you encounter any issues.

Thanks,
Seippel
大意：
发放新测试任务，资源需求：内存400M，硬盘10M，下载100K，上传300K。
注：官方又要加新项目了？

vmzy

Jul 25, 2011 7:29:55 PM
CMD2 project currently on hold
Due to an outage on one of the CMD2 researchers' servers, we have temporarily disabled this project and no additional work units are currently being sent out.
大意：
因为CMD2服务器出现故障，我们暂停了HCMD2项目。

zglloo

CERN Press Release "CERN supports European Year of Volunteering through Citizen Cyberscience Centre"
2011 八月 22

Researchers from the Citizen Cyberspace Centre (CCC) at CERN -- the European Organization for Nuclear Research -- in association with Tsinghua University in Beijing, China, are utilizing IBM's World Community Grid to conduct the Computing for Clean Water project. If successful, the nano tube water filtration project could result in new and improved methods for providing cost-efficient water purification systems throughout the developing world. According to some sources, nearly 885 million people worldwide lack access to safe water supplies.

World Community Grid is featured in a press release about CERN and grid computing. To read the press release, please click here.

Also, please check out the accompanying article on the Citizen IBM Blog. You can read that article here.

vmzy

Aug 24, 2011 11:54:41 AM
World Community Grid volunteers donate over 500,000 years to humanitarian research!
We are very pleased to announce that World Community Grid volunteers have donated more than 500,000 years of computer run time.

To see the statistics, please click here. To see how that amount was distributed among the research projects, please click here.

The amount of run time provided continues to accelerate. Here are the run time milestones since World Community Grid was launched in November of 2004:
* 100,000 years took 32 months and was achieved in July of 2007,
* 200,000 years took 17 months and was achieved in November of 2008,
* 300,000 years took 13 months and was achieved in December of 2009,
* 400,000 years took 11 months and was achieved in November of 2010.
* 500,000 years took 9 months and was achieved in August of 2011.

We look forward to reaching the 600,000 milestone during 2012.

A very big THANK YOU to all the members who have helped achieve this milestone!
大意：
里程碑：计算时间达到50万年。

vmzy

Sep 5, 2011 3:21:37 AM
Large Number of Errors for Drug Search for Leishmaniasis
We experienced a large number of result errors due to an error in one of the input data files. We have cancelled all of the impact workunits. The next time a computer with one of the affected workunits contacts our servers, the jobs on their machines that are impacted will be aborted.

We did not find this particular error in the targets 6-8 (target 5 experienced the error). In order to quickly determine if targets 6-8 have these errors, we are releasing 100 workunits from each of these three targets and having them sent out to see if they run properly.

While we wait for those results to be processed we are running the project slowly for the time the being.

Target 6 appears to be having troubles as well. To be cautious we are completely stopping the project until we get the results back for the 100 test workunits for targets 7 and 8.

Target 7 and 8 look like they are running correctly. Target 6 does have errors so we have cancelled those workunits.

We are starting a 100 workunit test with targets 9-15. Those have been sent out.

We have now resumed distributing work for this project for targets 7 and 8. We are will run the project at a reduced paced until we have a some process improvements in place to avoid this type of issue in the future.
大意：
由于某个原始文件错误，导致大量任务出错，我们现在暂停了Drug Search for Leishmaniasis 子项目，待错误修正后重新开放，当前未返回任务全部作废。
目标7和8貌似是正确的，我们恢复发包，不过优先级不高。开始测试9-15目标的正确性。