|
|
Pages: [1] 
|
 |
|
 Author
|
Topic: Short results reshedule for CPU (Read 350 times)
|
|
Pepi
|
Rescheduling tool is great, but tell me is this possible.
When tool scan result and determine what result is VLAR and what is not, is also possible to add option to take short ones ( below few minutes computing on cuda) to reschedule on CPU?
Maybe this is foolish, but I would love to have that option.
|
|
|
|
|
Logged
|
|
|
|
|
Raistmer
|
Such tasks called VHARs and Re-schedule 1.9 already has required option.
|
|
|
|
|
Logged
|
|
|
|
|
Pepi
|
Such tasks called VHARs and Re-schedule 1.9 already has required option.
So VHARS are always short ones? |
|
|
|
|
Logged
|
|
|
|
|
Raistmer
|
Such tasks called VHARs and Re-schedule 1.9 already has required option.
So VHARS are always short ones? Yes. VHAR means very high angle range. If angle range is big, telescope moved fast and any single signal source stayed in observation field relatively short time. That is, no some signal accumulating procedures are possible. Generally, we can extract much less from such tasks. Hence smaller processing times. There is just less work to do on those data. |
|
|
|
|
Logged
|
|
|
|
|
Pepi
|
In that case I am very confused!!!!  Reschedule tools shows only 105 VHARS and I have got 240 task with short times. ( all times are same cca 7 minutes) |
|
|
|
|
Logged
|
|
|
|
|
Raistmer
|
In that case I am very confused!!!! Reschedule tools shows only 105 VHARS and I have got 240 task with short times. ( all times are same cca 7 minutes) That is, VHAR threshold used in Re-schedule just differs from threshold used to estimate running time. What strange in that? If you would see let say 200 VHARs and only 100 "short" tasks - that would sound strange to me  |
|
|
|
|
Logged
|
|
|
|
|
Pepi
|
In that case I am very confused!!!! Reschedule tools shows only 105 VHARS and I have got 240 task with short times. ( all times are same cca 7 minutes) That is, VHAR threshold used in Re-schedule just differs from threshold used to estimate running time. What strange in that? If you would see let say 200 VHARs and only 100 "short" tasks - that would sound strange to me That option I would love to have  To determine only short ones between VHARS |
|
|
|
|
Logged
|
|
|
|
|
Raistmer
|
Again, ALL VHARs are short. Reverse isn't true, not all shorties are VHARs.
If AR of task near VHAR threshold it will have small execution time too, but will be not VHAR from re-scheduler point of view.
You could adopt my perl script to move short tasks to CPU if you want.
|
|
|
|
|
Logged
|
|
|
|
|
Pepi
|
Perl script never work for me ( win 7 64 bit) so I left that script and wait fro CPU2GPU tools But thanks for answer, maybe I came back to script |
|
|
|
|
Logged
|
|
|
|
|
Pages: [1] 
|
|
|
|
Quote!
Nothing is as easy as it looks.- Murphy's Law
|
 |  |  |
| |
| Site Statistics |
| Total Members: | 2,310 |
| Total Posts: | 25,586 |
| Total Topics: | 811 | | Downloads |
| Apps |
| Windows R-1.x | 0 |
| Windows R-2.0 | 0 |
| Windows R-2.2 | 0 |
| Linux 32bit 1.x | 0 |
| Linux 32bit 2.2 | 0 |
| Linux 64bit 2.2 | 0 |
| Alpha/IA64 | 1,759 |
| FreeBSD | 0 |
| HPUX | 0 |
| Subtotal: | 0 |
| Source packs: | 5,345 |
| Tool/WU packs: | 9,541 |
| Total: | 86,100 | | GBs dl'd: | 365.82 | | Pages served |
| Today: | 5,985 |
| Total: | 7,179,585 |
| (since 6/26/2006) |
| 173 Donations to S@H |
| U.S. Dollars: | 3,196.59 |
| Euros: | 863.90 |
| Last 24h: | $ 0.00 |
| Avg./24h: | $ 3.81 |
| Estim. total: | $ 4,319.66 |
Latest Member:
shepperd |
| |
 | |  |
| | |
| |
Online users/last 15m
46 Guests, 9 Users (1 Hidden)
ecki, Franz, Morten, Raistmer, gjpivko, jrusling, Darwin MLP, Rectifier
73 Members/last 24hecki, Franz, Morten, Raistmer, gjpivko, jrusling, Darwin MLP, Rectifier, benool, sunu, cjdigresia, Wild6-NJ, algoodman, ic451uk, mechtheist, Avatar1966, perryjay, kit344, ScitechGrid, clockman, bobcat79568, Urs Echternacht, Purple Rabbit, Lazydude, Richard Haselgrove, John Galt 007, Mortlake, aabenroi, Frizz23, HMN, Frawe, Jason G, mr.mac52, nenym, Kinguni, Byron Leigh Hatch @ team Carl Sagan, k6xt, crazyrabbit1, glk63, corsair, The Grinch, jasonpilkin, KarVi, Tye, Claggy, arkayn, Devaster, viper666, firefox, bcvv28, JohnDK, ralph_ellis, -ShEm-, Ondar, fumi, kararom, Josef W. Segur, norway415, _heinz, Tamzina, Geek@Play, Pishi, Vyper, Pepi, Trucido, talman, Ghost0210, PatrickV2, glennaxl, benson007, Matthias Lehmkuhl, mandy24e, zvonas
| |
| | |
|
