This is the first time the boards have actually allowed me to post on the SETI message boards :)

I was wondering.... sometimes my SETI WU come in as "High Priority".

I was wondering why that is.
Is it because of the size of the WU, or backlog, or something else ??

Cheers
Dee
xx

---

This is the first time the boards have actually allowed me to post on the SETI message boards :)

I was wondering.... sometimes my SETI WU come in as "High Priority".

I was wondering why that is.
Is it because of the size of the WU, or backlog, or something else ??

Cheers
Dee
xx

I see you have an Astropulse unit you are working on. That will take longer than a multibeam unit. The due date is in mid October. BOINC will put a work unit into high priority mode if it is worried about finishing the unit before the due date. I would guess that the Astropulse unit will take a few days, which would make BOINC concerned that it won't finish on time. There will be an outage starting today, so you might want to go to your account page, and click on when and how BOINC uses your computer. Down near the bottom of the page that comes up, you will see how many days work your system keeps on hand. You may want to edit that number to at least 3 days, 5 to be safer. that way when the 3 day outage starts, you will have enough work to carry you until the servers come back up on Friday.

Steve

---

Warning, addicted to SETI crunching!
Crunching as a member of GPU Users Group.
GPUUG Website

This happens when the time is near for completion based upon your computers capabilities. The High Priorities units are done first to try to assure they are finished prior to the return deadline.

This is the first time the boards have actually allowed me to post on the SETI message boards :)

I was wondering.... sometimes my SETI WU come in as "High Priority".

I was wondering why that is.
Is it because of the size of the WU, or backlog, or something else ??

Cheers
Dee
xx

---

"It is better to be hated for what you are then to be loved for what you are not"
- Andre Gide (1869-1951)

Thanks for that.
I looked at the running time/work hours for the project and it`s nearly 270 hours, so I`ve suspended my other WU to get on top of it :)

Cheers
Dee

---

I have received an Astropulse with deadline October 16 and it is running high priority. It is now at 11% after 3.5 hours, ok. But I also have an AQUA with deadline 23 September at 95%, with 7 hours to go, not running high priority. When it starts running it uses both cores of my Opteron 1210 since AQUA is multithreading. It will be interesting to see an AQUA/Astropulse conflict.
Tullio
AQUA wins.

---

Hi Dee, welcome to the boards.
You should see the time on that astropulse dropping fast as you process it. It takes awhile for SETI to learn your machine. You have to complete 10 AP units before it will give you the right completion time estimate. The same holds true for the multibeam tasks but you are well past that by now. It takes longer for AP tasks because the SETI servers are fairly stingy about sending them out.

---

PROUD MEMBER OF Team Starfire World BOINC

Welcome to the boards Dee. I see you are running a P4. I have one also it does an AP in about 40 hours using lunatics Unified installer. version .37 i believe.

I see you are running stock apps. Give Lunatics a try. It will increase your RAC.

And dont hesitate to ask questions on the board. There are all sorts of folks who will be glad to help out.

---

[/quote]

Old James

I have received an Astropulse with deadline October 16 and it is running high priority. It is now at 11% after 3.5 hours, ok. But I also have an AQUA with deadline 23 September at 95%, with 7 hours to go, not running high priority. When it starts running it uses both cores of my Opteron 1210 since AQUA is multithreading. It will be interesting to see an AQUA/Astropulse conflict.
Tullio
AQUA wins.

Tullio, you might want to upgrade from Boinc 6.10.56 to 6.10.58, the latest recommended build,
6.10.56 has a bug so it does multi threaded task in preference to EDF tasks,
not sure how it would play out when both tasks require EDF,
6.10.57 fixes this with:

•client: don't promote multithread jobs ahead of EDF jobs

Claggy