Message boards :
Number crunching :
Core i7 and Optimized App
Message board moderation
Author | Message |
---|---|
Send message Joined: 16 Jan 09 Posts: 9 Credit: 167,133 RAC: 0 |
I started running Milky Way on my new Core i7 machine today and was already impressed with the speed. But, I just downloaded and installed the optimized app for SSE 4.2 and all I can say is 'WOW!' The tasks that have run completely on the new version are finishing in about 6 minutes each, and I am running 8 at a time simultaneously (4 cores, 2 threads each). And, that's while doing other things on my machine too. The only problem I'm seeing now is they're cranking out so fast, I can't download new ones fast enough. I keep getting messages saying requesting new tasks, but got 0. Anyone know how to get the project to feed me more? |
Send message Joined: 16 Jan 09 Posts: 9 Credit: 167,133 RAC: 0 |
Ok, I've watched it for a while, and here's what it seems to be doing... It will download and start 8 tasks and upload the results one by one as they finish. However, the project won't give me anymore new tasks until usually at least half of them have completed. Is my machine just going faster than they can get processed on the other end? I just ask because that means more than half the time it's running less than the full 8 that it's capable. But, at least I am getting more tasks eventually. |
Send message Joined: 16 Jan 09 Posts: 9 Credit: 167,133 RAC: 0 |
You know... I think it's ok now... I think it was just a problem of 8 tasks starting and finishing in unison. Now that they've sort of become staggared, things are running smoothly and my CPU is keeping busy. Times look to be between 6 and 9 minutes per WU now. Cool. I'll let it run and see how it does... |
Send message Joined: 16 Feb 09 Posts: 109 Credit: 11,089,510 RAC: 0 |
The server status page currently only shows a few hundred work units "ready to send". You probably are just hitting the servers when there's not a full pool of work to draw from. |
Send message Joined: 29 Aug 07 Posts: 327 Credit: 116,463,193 RAC: 0 |
Are you using BOINC v6.x.x? The scheduler in those versions really do not like MW and its 12 WU/core at a time. Calm Chaos Forum...Join Calm Chaos Now |
Send message Joined: 16 Jan 09 Posts: 9 Credit: 167,133 RAC: 0 |
I'm running BOINC 6.6.7 (64bit) Also, more info: Core i7 920 6Gb DDR3 Triple Channel RAM Asus P6T6 WS Revolution motherboard Vista Home Premium 64bit Also, I don't overclock anything. All stock/default settings. I'm not seeing any messages to indicate any problems. Looks like about 1 task completing every 60 seconds or so. |
Send message Joined: 8 Mar 08 Posts: 17 Credit: 4,411,459 RAC: 0 |
6mb DDR3 Triple Channel RAM lol hope you have enough to run Milkyway alongside windows :-P |
Send message Joined: 16 Jan 09 Posts: 9 Credit: 167,133 RAC: 0 |
oops! I corrected that to 6Gb. Thanks. Long day... |
Send message Joined: 8 Mar 08 Posts: 17 Credit: 4,411,459 RAC: 0 |
No problem. Nice rig! I'd like to build an i7 but with the mobo and special ram that you had to get as well it just gets very expensive, so i have opted to build another Quadcore instead :) |
Send message Joined: 8 Dec 07 Posts: 60 Credit: 67,028,931 RAC: 0 |
You know... I think it's ok now... I think it was just a problem of 8 tasks starting and finishing in unison. Now that they've sort of become staggared, things are running smoothly and my CPU is keeping busy. Times look to be between 6 and 9 minutes per WU now. I think you're just hitting the server when there was little to send. I'm running an i7 and it usually ranges to 1 or 2 that haven't started to a full house of 96. and if you have an ati 4800 graphics card on your machine, then you might want to give the gpu app a whirl. then you can crunch 16 @ a time in seconds. |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
The server status page currently only shows a few hundred work units "ready to send". You probably are just hitting the servers when there's not a full pool of work to draw from. Whenever the available workunits drops below 400 our assimilator automatically generates 500 new ones. I could tweak these numbers if it's making some of your work requests not go through. |
Send message Joined: 27 Aug 07 Posts: 915 Credit: 1,503,319 RAC: 0 |
The server status page currently only shows a few hundred work units "ready to send". You probably are just hitting the servers when there's not a full pool of work to draw from. Add a zero to those numbers. me@rescam.org |
Send message Joined: 26 Dec 07 Posts: 41 Credit: 2,582,082 RAC: 0 |
Whenever the available workunits drops below 400 our assimilator automatically generates 500 new ones. I could tweak these numbers if it's making some of your work requests not go through. I'd suggest you tweak away. It's not unusual to have to ask more than once to get a task sent down the pipe. As it only causes a one or two minute delay at the client end it isn't a problem for us but the repeated requests must be giving the server more to do than is necessary. Mon 23 Feb 10:33:01 |Sending scheduler request: To fetch work. Requesting 62 seconds of work, reporting 1 Mon 23 Feb 10:33:06 |Scheduler request succeeded: got 0 new tasks Mon 23 Feb 10:34:07 |Sending scheduler request: To fetch work. Requesting 224 seconds of work, reporting 0 Mon 23 Feb 10:34:12 |Scheduler request succeeded: got 0 new tasks Mon 23 Feb 10:35:13 |Sending scheduler request: To fetch work. Requesting 385 seconds of work, reporting 0 Mon 23 Feb 10:35:18 |Scheduler request succeeded: got 1 new tasks |
Send message Joined: 19 Feb 09 Posts: 33 Credit: 1,134,826 RAC: 0 |
Yes, it would be helpful to add more WUs to the pool :) My GPU client runs dry sometimes and MW is the only project I am crunching now. |
Send message Joined: 18 Feb 09 Posts: 8 Credit: 2,424,453 RAC: 0 |
I'm getting a lot of "requesting new tasks - request completed: got 0 new tasks" messages on all of my machines. So they sit idle, then request again. Looking at the logs i'm suprised i'm able to even connect to the server since I can't be the only one hammering it many times a minute. |
Send message Joined: 20 Jan 09 Posts: 9 Credit: 17,289,621 RAC: 0 |
I have been running dry periodically, so increasing the work available would be nice, that and changing the work unit per core quota to double. Would give you more units in progress, and lower the chance that some machine ends up idle |
Send message Joined: 7 Sep 07 Posts: 444 Credit: 5,712,523 RAC: 1 |
I'm getting a lot of "requesting new tasks - request completed: got 0 new tasks" messages on all of my machines. So they sit idle, then request again. Looking at the logs i'm suprised i'm able to even connect to the server since I can't be the only one hammering it many times a minute. Has it occurred to you that the problems may be caused by the people who are hammering the server many times a minute? ;) |
Send message Joined: 6 Apr 08 Posts: 2018 Credit: 100,142,856 RAC: 0 |
I'm getting a lot of "requesting new tasks - request completed: got 0 new tasks" messages on all of my machines. So they sit idle, then request again. Looking at the logs i'm suprised i'm able to even connect to the server since I can't be the only one hammering it many times a minute. It seems to me that it's all the Cylon's fault (who won't share with humans). It seems to me that if MW is the only project running it will request work. If say CPDN is also running, at times it won't request any more MW. If I suspend CPDN, MW tasks are almost immediately requested. |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
It seems to me that it's all the Cylon's fault (who won't share with humans). It seems to me that if MW is the only project running it will request work. If say CPDN is also running, at times it won't request any more MW. If I suspend CPDN, MW tasks are almost immediately requested. On the server side, it appears that we never drop below 300+ available workunits at any given time. |
Send message Joined: 27 Aug 07 Posts: 647 Credit: 27,592,547 RAC: 0 |
It seems to me that it's all the Cylon's fault (who won't share with humans). It seems to me that if MW is the only project running it will request work. If say CPDN is also running, at times it won't request any more MW. If I suspend CPDN, MW tasks are almost immediately requested. Hm, but since today I'm getting the same on my lappy: 23.02.2009 14:51:38 Milkyway@home Sending scheduler request: To fetch work. It's sitting idle (only a little FreeHAL WU is running but that's non-cpu intensive so Milkyway does request work but the lappy isn't getting some... ) Lovely greetings, Cori |
©2024 Astroinformatics Group