[BOINC] [天文类] MilkyWay@home

Youth

刚看到帖子，要编辑啥？

cuihao

只是想多翻译一句，

cuihao

Project News: Large WU Sizes
2009年11月4日
I've started some new searches with larger sized workunits, so hopefully these will help the server strain. Let us know how they're running.--Travis
我们开始了新的研究，WU比原先大，希望能缓解服务器压力。---特拉维斯

Project News: Website Slowness
2009年11月4日
We've been looking into the website performance and it looks like we're going to be ordering some more hardware in the next couple weeks which should improve the performance. Until then you're probably just going to have to bear with the website slowness. I'm currently trying to finish up my phd thesis (I defend in less than 2 weeks), so I probably won't be frequenting the forums very much until then.--Travis
我们正在调查网站的性能，貌似下几周需要添置一些新硬件改善网站速度。在此之前，你只能容忍一下网站缓慢的速度了。目前我正忙于写博士论文，近两周可能无法经常访问论坛。---特拉维斯

zglloo

News：Large WU Sizes
I've started some new searches with larger sized workunits, so hopefully these will help the server strain. Let us know how they're running.
--Travis

我已经开始进行较大规模的wu，所以希望这将有助于一些新的服务器适应新的应变。可以我们知道他们是如何运行的。——特拉维斯

BiscuiT

November 7, 2009
Just noticed a problem with the assimilator crashing. It should be back up and running and work should start flowing again.

November 7, 2009
There's been a lot of questions about upgrading the server and putting out server side ATI applications so here's an update about what's going on on our side:
We do have versions of the ATI applications available, however here's some changes that the astronomers need to test to put in these applications so we're partially working on getting those all ready before we do a big update and put everything out for everyone.
To make a long story short, the model we're crunching now (while valuable) has some problems in describing the background distribution of stars in the milky way galaxy.

What the application does is tries to separate stars which were ripped apart from other galaxies that came close to the milkyway (like the saggitarius stream, which is what our current focus is on) as well as other clusters of stars from stars that were more 'originally' in the milky way. This will let us figure out the current shape of the milky way and give us interesting information about how galaxies interact and things of that nature. So right now we've found out that how astronomers have describing the 'background' stars of the milky way really isn't very correct. I'm pretty sure Heidi and her students are working on some kind of publication dealing with this issue right now.

So to deal with that issue, they've been testing different models which should help with this in getting us even better models that deal with this problem. So currently you guys have helped us find a problem in astronomies current view of the milky way, and hopefully will help us really understand what the milky way looks like.

So while we may not be very fast in upgrading hardware, we're at least doing some astronomy here :P Computer science too - we've just submitted a paper to this year's PPAM (parallel processing and applied mathematics) conference describing the GPU work, which I should be making a link available to everyone as soon as it's accepted.

I'm sorry that the server issues go unattended for so long, but we don't have anyone really doing the networking. It's just me in my spare time (which i have none of right now while i'm finishing writing my phd thesis). Before we had Dave to work on that but he's graduated and we haven't found another undergraduate student to work on this yet. Hopefully next semester we'll have another one. We're in the process of ordering new hardware which should improve the performance of the server, but it will be a couple weeks before it gets here, and probably another week or two before we've updated all the server side code to work on multiple CPUs.
--Travis

cuihao

看来这个Travis最近写论文比较忙，milkyway的事无暇顾及。

BG4ACE

那个优化的网页全英文的  看不懂  6.10.17  该怎么用GPU优化

cuihao

November 10, 2009
We also really want to apologize for all the recent server issues and lost credit. Hopefully you'll all still be around when we get the server back up and work flowing again. I'll post more as soon as I know about hardware orders and what's going on.
--Travis

November 10, 2009
In order to save you guys more lost credits, I don't think we'll be starting up new work until we have replacement hard drives. What I've gotten from labstaff is that the drives are running in degraded modes and hurting really bad. They're telling us the reason for the problems has been the construction around campus at RPI which has caused a lot of vibration in the computer labs which has wrecked quite a few hard drives. It seems we're not the only ones having similar issues. Hopefully we can have new hardrives in a day or two and get things back up and running.
--Travis

November 10, 2009
We've restored the server from the last backup (which was this morning) so hopefully not too much credit has been lost. I still have to purge the database of all the workunits, unfortunately. We also need to order new hard drives for the server, so I'm not sure how stable things will be for the next week or so until we get them installed. But at least hopefully that explains the issues we've been having lately.
--Travis

November 10, 2009
It looks like I'm going to have to remove all the workunits and results from the database. So if you have any running, feel free to cancel them.

November 10, 2009
We have a backup from this morning, but it may have been taken after all the corruption. We're going to try it out and see if it helps anything.
--Travis

November 10, 2009
It looks like some serious problems happened. Right now I've turned off all the BOINC daemons until we can get the database restored to a previous backup (which should hopefully bring back a bunch of credit).
--Travis

November 10, 2009
We're looking into why the server went down. Something about unrecoverable disk errors. Hopefully people haven't lost too much credit or anything like that.
--Travis

November 7, 2009
Just noticed a problem with the assimilator crashing. It should be back up and running and work should start flowing again.

似乎服务器发生了某些严重错误，项目官方作出道歉。

注意：项目清空了任务单元服务器，请终止当前计算中的所有任务[/red]

cuihao

写错了，用的chrome，编辑界面的话一塌糊涂。

就干脆不改了。

cuihao

November 15, 2009
Just letting everyone know we ordered new hard drives for the server last week, and hopefully they will be here soon. We're hoping to have everything back up and running within the week.
--Travis
订购了新的硬件，希望这周内能够搞定。

cuihao

November 19, 2009
I've also linked to the slides from my PhD defense if anyone would like to see them here: [ppt][ppt] [keynote]

November 19, 2009
Another update for your reading pleasure. We finished an invited paper for the 2009 Parallel Processing and Applied Mathematics (PPAM) conference about the GPU work here at MilkyWay. We'd like to thank Andreas Przystawik and Dave Anderson for their collaboration on this. Here's a link to the paper: Accelerating the MilkyWay@Home volunteer computing project with GPUs.

November 19, 2009
Just wanted everyone to know that I successfulled completed my defense today, so I'll have some time to get the server upgraded and all that other great stuff. Hopefully things should be back up and running in the next couple days.
--Travis

cuihao

November 22, 2009
Looks like the new hard drive is ready to go. I've started up the daemons and work should be available.

November 22, 2009
I've been on campus today trying to get to labstaff, but they haven't been around. We're upgrading our issue to get the BOINC software upgraded and the new hard drive installed, so I'm hoping we'll have everything finished and back up and running tomorrow after I meet them.
--Travis

已经恢复运行！

cuihao

November 25, 2009
The machine crashed with the new drive running. We took it out of the machine to run some diagnostics on it, and it seems to be ok so we're worried we might have a problem with the controller. We're going to try and send out more WUs and see how things go, but we may have to order a new controller if the machine crashes again.
--Travis

cuihao

November 26, 2009
I'm going to try and generate some work and see if the few changes I made server-side improve the problems we've been having. Hopefully nothing crashes.

November 26, 2009
I think I fixed the team issue, it should update on the webpage in a few minutes. Let me know if there are any other issues with it.
--Travis

November 26, 2009
Looking at the logs, its seems that the teams table in the database crashed so we're going to have to restore it to an earlier version. Chances are this is because of the controller issue, so I'm not going to generate more work until that's fixed, because I don't want you guys losing credit due to hardware issues.
--Travis

November 26, 2009
I'm making a few changes of my own which I think may help the server performance, it'll be down for a little bit.
--Travis

cuihao

November 30, 2009
Good News
The server seems to have been running fine for the last few days, and I've heard from labstaff that the controller is fine. I think I might have fixed the problem that was causing the random crashes. Sadly, the server is still extremely overloaded, so for the time being I'm increasing the workunit size (about 4x what they were before). Hopefully this should make the webpage more responsive and work more easily available. I'm really hoping to have all the server code upgraded this week which I'm hoping will improve things even more.

服务器最近运行情况不错，但是依然超载运行，为了减轻服务器负担，WU的大小扩大了4倍。