[BOINC] [天文类] Orbit@Home

Youth

Successful Testing, Meeting, and Notes

Submitted by tricaric on Sat, 05/03/2008 - 02:27.

The past month was a good one for orbit@home. We have deployed and tested the SurveySimulator application on all major platforms, and all the bugs discovered so far seem to have been fixed. Thank you for reporting problems.

We're now working on the next version of the application, that differs mainly for the larger flexibility of scenarios that can be studied. This new version, likely version 1.20, is almost ready and we should start testing it by mid-May.

上个月的成绩很不错，在所有主要的平台上部署并测试了项目的计算程序，所以发现的错误都已得到解决，感谢大家！

我们现正在准备下一个版本的计算程序，主要是希望有更好的适用性，新版本应该会在五月中旬开始测试。

Youth

Getting ready for the ACM 2008 Meeting

Submitted by tricaric on Wed, 05/07/2008 - 16:04.

We received very good news today. Our abstract to the ACM 2008 meeting on "Numerical Simulations of Efficient Search Strategies for Near Earth Objects Surveys" has been accepted in the NEO surveys and hazards section. This means that your computers will be very busy in the next two months, as soon as version 1.2x of SurveySimulator is deployed and tested.

论文被ACM接收。这意味着等1.2x版本计算程序部署并测试完成后，大家的计算机在接下来两个月将有足够的任务包可供计算。

Youth

Testing Phase Completed

测试阶段完成！

Submitted by tricaric on Wed, 05/28/2008 - 23:51.

With SurveySimulator version 1.31, we believe to have reached a fairly good point in terms of stability and flexibility. Our tests so far show very high success rates on all platforms. So we're going to release production WUs soon. These WUs will require longer times to complete, in the range between several hours to a few days on the average host. The completion time estimates have also been adjusted for better accuracy, often within a factor two from the observed value.

我们相信1.31版本计算程序的稳定性和适用性都相当不错，到目前为止的测试结果成功率非常高。因此我们将很快转入正式运行。正式运行的任务包将需要更长的计算时间，在主流的机器上从几小时到几天不等。任务包的完成时间估算已经进行过修正，应该准确度更高了。

Youth

orbit@home public beta

The orbit@home project is now open for public beta. After several
months spent testing our main application, called SurveySimulator,
we're now moving into a public beta phase, characterized by mainly
production WUs and infrequent application updates to fix minor bugs or
to add missing features. We believe the project is now mature enough
to accept BOINC users with modest experience, and we're looking into
expanding our users base from the actual ~3K to ~10K.

About orbit@home

Our project focuses on celestial mechanics, with applications to the
dynamics of asteroids. Our first scientific application,
SurveySimulator, is designed to accurately simulate the performance of
astronomical surveys searching for Near Earth Objects (NEOs),
asteroids and comets orbiting close to the Earth. By joining
orbit@home, you help us understand and optimize the search patterns
followed by the surveys, thus improving their performance and enabling
higher NEOs discovery rates.

The SurveySimulator application is available on all major platforms,
including Windows/x86, MacOSX/x86, and Linux/x86.

To join orbit@home, please visit: http://orbit.psi.edu/oah/

For project news and results, visit: http://orbit.psi.edu/

Sincerely,

Pasquale Tricarico [orbit@home scientist and developer]
Planetary Science Institute

universebreaker

大概指頂目進入開放式beta階段
users base(是不是指用戶資料?)的空間會由大約3k加大至約10k
第一個項目程式:survey simulator
是設計來進行近地物體(neo)搜索項目的精確模擬
模擬地球附近小行星和彗星的軌道
這項目是用來改良它的搜索模式~

Youth

嗯，是项目方希望用户数能增长到10k左右。

universebreaker

但可惜不是直接搜索,而是改良它的搜索模式...

Youth

不是也说了“thus improving their performance and enabling
higher NEOs discovery rates”吗？

Youth

Submitted by tricaric on Sat, 06/14/2008 - 02:10.

The first batch of production WUs completed rather quickly, and we spent the past week developing the tools to synthesize and analyze the results. We're pretty excited about the numbers we're getting so far, even if still preliminary: it looks like the optimized NEO search strategy we're developing has the potential to improve NEO detection rates beyond our most optimistic expectations. This of course calls for more tests and verifications, so we're generating more work. Expect a few waves of about 1 to 2 thousand WUs this coming week. Thanks for supporting this project, and making this research possible. We're going to post more details about the results we're getting after the ACM conference, in late July.

第一批正式的任务包很快就完成了，上个星期我们都在开发一些工具对这些结果进行合成分析。目前的结果还不错，超出了我们最乐观的预期。因此我们准备生成更多的任务包。感谢大家对项目的支持。我们将在七月下旬的ACM会议之后发布更多关于计算结果的细节。

//刚下了一个任务包，预计完成时间>150小时。。。

涼宮ハルヒ

我也下到一个任务包
要89小时。。。

Mr.Cancel

如果这个项目能有屏保就好了，分析地外轨道不明天体，屏保应该很拉风的吧
刚接到包161个小时

[ 本帖最后由 chenbang 于 2008-6-21 10:08 编辑 ]

Mr.Cancel

这个项目的计算程序很诡异哦，我到目前为止已经算了60+个小时，但是预计完成时间不减反增，居然还是160个小时。

[ 本帖最后由 chenbang 于 2008-7-8 15:39 编辑 ]

Youth

me2，我那个已经200多个小时了。。。进度40%

Julian_Yuen

这不赶上小号CPDN了...

coffee

这个项目比较感兴趣，不知道国内有没有类似的项目开发？