vmzy 发表于 2005-1-13 22:20:15

[BOINC] [物理类] Einstein@Home

Jan 4, 2005
In anticipation of 'going public' Einstein@Home is now running on the new server hardware. We need to complete and test (hopefully) one final round of application revisions before this happens. Hopefully this will eliminate the remaining problems with SIGABORT failures in the Linux application and similar problems (and a useless docked window) in the Macintosh application.
2005年1月4日
为了能尽早'正式开始',Einstein@Home现在在新服务器硬件上运转。在这之前,我们需要完成并测试(希望如此)修正版客户端。希望能解决剩下的Linux应用程序的SIGABORT故障和Macintosh应用程序的类似问题(和一个无效窗口问题)。

Jan 12, 2005
In the next several days we are going to shift from using the current BOINC scheduler to a new BOINC 'locality' scheduler. This will have some important benefits. In particular, it will ensure that host computers get lots of work for the (large) data files that they have already downloaded, instead of being required to download a new large data file for each new item of work. It will also remove these large data files from the host computers when they are no longer needed. Unfortunately the new scheduler is not compatible with the test workunits currently in progress on many hosts. Testers, please note that for this reason, we will be forced to cancel a number of workunits currently in progress, when we shift to the new scheduler.
2005年1月12日
在这几天里我们将要把现在使用的BOINC调度程序更换为新的BOINC'本地'调度程序。这将有一些重要的好处。 尤其,它将保证主机不再为已经下载了的(大)数据文件所累,按需下载一个新的大数据文件中的文件,而不再下载全部文件。当他们不再被需要时,它也将从主机删除这些大的数据文件。令人遗憾新调度程序与当前在很多主机上进行中的测试workunits不兼容。因此,内测人员请注意,当我们转到新调度程序时,我们将被迫删除许多当前进行中的workunits。

Jan 13, 2005
IMPORTANT NEWS: David Anderson, the creator of BOINC, has found and fixed a serious bug in the BOINC 'core client'. This causes failure of a job if the download of any of its input files fails (the correct behavior is to retry and complete the download, then run the job). Moreover, any other job that used the same files would also fail. A fixed core client should be available soon. This should make Einstein@Home work much better and eliminate the main source of job failure
2005年1月13日
重要消息:David Anderson,BOINC的创造者,已经发现并修正了一个BOINC '核心程序'里的严重漏洞。如果它的任何输入文件的下载失败,这将导致一个任务的失败(正确的行为是重试下载并完成下载,然后,才开始工作)。而且,使用相同的文件的任何其他任务也将失败。不久应该会发布一个修正版的核心程序。这应该使Einstein@Home更好地工作并且消除任务失败的主导因素。

翻译 BY VMZY

[ Last edited by Youth on 2006-6-17 at 11:11 ]

vmzy 发表于 2005-1-15 12:05:16

Jan 14, 2005
To prepare for the upgrade of our scheduler, we are not providing new work for the next day or two. We are taking this measure to try and minimize the amount of lost credit and to make the transition to the new scheduler as smooth as possible.
2005年1月14日
为我们的调度程序的升级作准备,接下来的几天我们不提供新任务。采用这种方法我们可以尽量把丢失的积分数量减到最小并使新调度程序的转换尽可能顺利。

翻译 BY VMZY

vmzy 发表于 2005-1-19 00:59:45

Jan 17, 2005
As explained in the news item from Jan 13, below, the BOINC 4.13 core client contained a serious bug that has now been fixed. Einstein@home users: please upgrade to the BOINC 4.14 core client, which can be found halfway down this page.
2005年1月17日
如1月13日新闻里所述,BOINC 4.13核心程序包含一个严重的漏洞,现在它已经被修正了。Einstein@home用户:请将BOINC核心程序升级到4.14版,这里是下载页面。

Jan 18. 2005
We have cancelled the remaining "old scheduler" workunits, and started testing the new locality scheduler. This should send work to hosts that already have the correct data input files, greatly reducing the download demands on host computers and the Einstein@home server.
2005年1月18日
我们注销了现有的"旧调度程序"的workunits,并且开始测试新的本地调度程序。它会把任务发给已经有校正后数据输入文件的客户端,大大减少客户端和Einstein@home服务器的下载带宽需求。

翻译 BY VMZY

vmzy 发表于 2005-1-19 22:57:08

Jan 18. 2005
The BOINC developers have released a new version of the client (4.15) that fixes a bug that caused some lost work for CPDN users when they were upgrading the client. Please upgrade to 4.15 as soon as possible.
2005年1月18日
BOINC开发员发布新版客户端(4.15)修正了一个BUG,它将导CPDN用户在升级客户端时,会丢失一些任务。请尽快升级至4.15。

vmzy 发表于 2005-2-22 11:58:34

项目正式开始,2月19日开放注册!

airwolfliu 发表于 2005-3-5 21:05:59

已经注册,开始算了

Youth 发表于 2005-3-6 14:03:48

March 5, 2005
In the two weeks since our public launch, Einstein@Home has gone from about 6,000 users (with credit) to over 20,000, and is currently adding users at the rate of about 1,000/day. Apart from one 30-minute crash, the project server has been operating stably; we estimate that it has the capacity to handle about ten times the current number of users, hosts and workunits.

2005年3月5日
项目正式开始后两个星期内,E@H的用户已经从大约约6000名增加到超过20000名,目前保持大约每天1000名的增长速度。在一次为时半小时的故障之后,项目的服务器已经正常运行,我们估计目前的服务器配置可以容纳十倍于现在的用户,主机,计算包数量。

vmzy 发表于 2005-3-9 21:56:33

March 8, 2005
We had a short outage today that resulted from the kernel running out of memory and killing the database.
You may have noticed problems connecting to the scheduler or the forums.
2005年3月8日
今天我们当一会儿机,导致内存溢出并且数据库也被破坏了。
当你连接调度程序或者论坛时你就会注意到这个问题。

[ Last edited by vmzy on 2005-3-10 at 22:06 ]

慢舟 发表于 2005-3-10 08:24:45

怎么到处停电?

Youth 发表于 2005-3-10 18:34:05

好像这里的outage应该是停止服务,而不是停电

vmzy 发表于 2005-3-10 22:07:00

已修正

vmzy 发表于 2005-3-15 22:23:35

March 14, 2005
A data download mirror is now being hosted by the PSU LSC group. Einstein@Home
users in the US Eastern Time Zone will now try to get data from this mirror first.
2005年3月14日
PSU LSC小组现在开了一个数据下载镜像服务器。在美国东部时区的Einstein@Home用户现在将首先尝试从这个镜像服务器下载数据。

Youth 发表于 2005-3-19 21:53:12

March 18, 2005

A data download mirror is now being hosted by the MIT LSC group. Einstein@Home users in the US Eastern Time Zone will also use this mirror.

基本同上,在MIT又开了个数据下载镜像服务器。

vmzy 发表于 2005-4-7 21:51:55

April 6, 2005
A new version of the validator (Teviet Creighton) is now in use. This should reduce the numbers of results which are incorrectly marked as 'invalid'.
2005年3月6日
启用一个新的检验程序(Teviet Creighton)。这应该会降低被误标为'无效'的结果的数量。


April 6, 2005
The Einstein@Home server will be shutdown on Fri Apr 8 19:00 UTC for approximately one hour for maintenance.
2005年3月6日
Einstein@Home服务器将于4月8 星期五19:00 UTC关闭大约一小时以进行维护。

Youth 发表于 2005-4-8 09:29:22

April 7, 2005
During the next few days we will finish the first round of S3 analysis and begin a second round. As we finish up the first round of S3 analysis, users may find that they are getting a new data file with each workunit; users with modem connections may want to suspend Einstein@Home for a few days to avoid multiple long downloads. Note that there may also be some instability in a few days as we shift to new workunits.
在接下来的几天,我们将完成S3分析的第一阶段,开始下一阶段。完成第一阶段期间,用户会发现计算每个wu都需要下载一个新的数据文件,窄带上网用户可以选择暂停运行Einstein@Home项目几天以避免经常性的长时间下载。另外,在接下来的几天,系统可能会不太稳定。

[ Last edited by Youth on 2005-8-16 at 13:46 ]
页: [1] 2 3 4 5 6 7 8 9 10
查看完整版本: [BOINC] [物理类] Einstein@Home

论坛官方淘宝店开业啦~
欢迎大家多多支持基金会~