Posts by Tom*
|
1)
Message boards :
News :
Testing Some New Plan Classes
(Message 67230)
Posted 8 Mar 2018 by Tom* Post: Thank you Jake, most of my workunits failing to successfully complete are due to wingers without DP. Bill |
|
2)
Message boards :
News :
Nbody 1.68 release
(Message 67068)
Posted 11 Feb 2018 by Tom* Post: Think we need a new version of the application that can process both v166 and v168 data file formats. PLEASE I have only been getting v166 lately is there a pointer the the v166 application? |
|
3)
Message boards :
News :
Nbody 1.68 release
(Message 67026)
Posted 2 Feb 2018 by Tom* Post: Mossy, The problem is when the V168 application tries to process V166 data You have same issue as I if you look inside the stderr de_nbody_1_13_2018_v166_20k__optimizerparameters_diff_seedruns_3_1516211024_96183_4 this is the data version. As Sidd says it only takes 2 or 3 seconds to fail, so if there are no other ramifications (like not getting new tasks:-)) just let them run. otherwise you have to highlite the task in the task list in BOINC then choose properties to see the version v166 or v168 |
|
4)
Message boards :
News :
Nbody 1.68 release
(Message 67023)
Posted 1 Feb 2018 by Tom* Post: Thanks Sidd |
|
5)
Message boards :
News :
Nbody 1.68 release
(Message 67019)
Posted 31 Jan 2018 by Tom* Post: Sidd, All three of my systems running 1.68 get the following error, some more than others. <core_client_version>7.6.22</core_client_version> <![CDATA[ <message> The data is invalid. (0xd) - exit code 13 (0xd) </message> <stderr_txt> <search_application> milkyway_nbody 1.68 Windows x86_64 double OpenMP, Crlibm </search_application> Using OpenMP 4 max threads on a system with 8 processors Error evaluating NBodyCtx: [string "-- /* Copyright (c) 2016 Siddhartha Shelton..."]:81: bad argument #1 to 'create' (Missing required named argument 'BetaSigma') Failed to read input parameters file 14:52:09 (8888): called boinc_finish(13) My I7 systems get very few but my I5 system gets a ton. Yet are completed ok by other systems on a resend |
|
6)
Message boards :
News :
Update on This Weeks Errors
(Message 66776)
Posted 8 Nov 2017 by Tom* Post: Not if the card reports or BOINC thinks its a 6800 series AMD AMD Radeon HD 6790/6850/6870 series (Barts) (1024MB) driver: 1.4.1848 OpenCL: 1.2 |
|
7)
Message boards :
News :
Update on This Weeks Errors
(Message 66772)
Posted 6 Nov 2017 by Tom* Post: Leif Your GPU does not support double precision so cannot be used for this project. from STDERR Board: AMD Radeon HD 6800 Series Driver version: 1800.11 (VM) Version: OpenCL 1.2 AMD-APP (1800.11) Compute capability: 0.0 Max compute units: 12 Clock frequency: 820 Mhz Global mem size: 1073741824 Local mem size: 32768 Max const buf size: 65536 Double extension: (none) Device doesn't support double precision |
|
8)
Message boards :
Number crunching :
6970 tweaking
(Message 66586)
Posted 5 Sep 2017 by Tom* Post: I think your 6970 is performing very well Since last November each work unit is 5 times longer than before. This was implemented to cut down on number of accesses to database as the database was overloaded and timing out our requests. Everything is operating optimally since then. Oh and the credits per WU is also 5 times |
|
9)
Message boards :
Number crunching :
AMD/ATI 4800 series
(Message 66550)
Posted 1 Aug 2017 by Tom* Post: Just a wag but all my successful GPU tasks show Double extension: cl_khr_fp64 yet info on the web says hd 4850 and hd4870 do NOT implement the full set of cl_khr_fp64 need HD 5xxx to implement this fully. may or may not be a cause |
|
10)
Message boards :
Number crunching :
M@H seems to be freezing the progress of other tasks
(Message 66293)
Posted 12 Apr 2017 by Tom* Post: Why is the M@H task consuming 100% of the CPU if it's supposed to be a GPU task? short answer because you are using NVidia and not ati longer answer calls to gpu opencl spinwait on the cpu so there is no delay in queing up another call to opencl. other projects tackle this issue by giving parameters to have the cpu sleep instead of spinwait if the sleep interval is not correct the gpu task completion time suffers. AMD ati gpu implementation of opencl uses magic to wait for a callback from the gpu so the cpu does not have to spinwait Limit yourself to 1 or 2 tasks running on the gpu or else all the instances will be spinning on multiple cpu's waiting for the gpu to finish a call. |
|
11)
Message boards :
Number crunching :
(RX 480) GPU tasks looping a couple of times before completing
(Message 66281)
Posted 8 Apr 2017 by Tom* Post: Feature Back in November 2016 ?? Jake consolidated 5 tasks into one task so the servers would not get overloaded, (Which they were) It solved a major problem The restart is just a feature of this consolidation. We only have short tasks now so the consolidation does give 5 times the credit from before the consolidation. |
|
12)
Message boards :
News :
GPU Issues Mega Thread
(Message 66236)
Posted 20 Mar 2017 by Tom* Post: According to this url https://en.wikipedia.org/wiki/Radeon_HD_6000_Series The HD6870 does not support double precision. |
|
13)
Message boards :
Number crunching :
Out Of Work?
(Message 65425)
Posted 11 Oct 2016 by Tom* Post: R9 280X so nothing *extremely* powerful there ^^ As far as AMD GPU's are concerned the Tahiti based GPU's are still the most Powerful Double Precision GPU's available. Only the NVIDIA Titan is more powerful. |
|
14)
Message boards :
News :
Updated Server Daemons and Libraries
(Message 65380)
Posted 4 Oct 2016 by Tom* Post: Thanks Jake, Looks like its back to normal at my end. Really appreciate your hard work. |
|
15)
Message boards :
News :
Updated Server Daemons and Libraries
(Message 65293)
Posted 26 Sep 2016 by Tom* Post: Sorry Cynthb but the AMD HD6800 series has no double precision capability. you need an HD69xx or higher to work here in Milkyway. Error from STDERR Double extension: (none) Device doesn't support double precision |
|
16)
Message boards :
Number crunching :
"Completed, can't validate" - what's this?
(Message 64787)
Posted 1 Jul 2016 by Tom* Post: Ret question Why aren't Aborted by User errors ignored? |
|
17)
Message boards :
Number crunching :
GPU error while computing
(Message 64273)
Posted 26 Jan 2016 by Tom* Post: SORRY BUT SPECS SAY NO DOUBLE PRECISION for that APU error compiling also failed on double precision |
|
18)
Message boards :
Number crunching :
What is the cause of these 'validate errors'
(Message 63805)
Posted 15 Jul 2015 by Tom* Post: Results so far for 7.6.6 State: All (2458) · In progress (40) · Validation pending (0) · Validation inconclusive (92) · Valid (2326) · Invalid (0) · Error (0) Application: All (2458) · MilkyWay@Home (1331) · MilkyWay@Home N-Body Simulation (0) · Milkyway@Home Separation (0) · Milkyway@Home Separation (Modified Fit) (1127) Thanks Keith and Richard for pushing the workaround |
|
19)
Message boards :
Number crunching :
Never Ending WU's + Invalid "Separation"
(Message 63712)
Posted 14 Jun 2015 by Tom* Post: re Invalids due to stderr blankness. Alot of us get 5% to 7% invalids using HD7950's on my ATI FX-8350 the whole stderr is always blank, on my Haswell-E with HD7950 it always cuts off the same portion of the STDERR, just at a different spot than [AF>EDLS]zOU FWIW my amd fx-8350 always truncates the whole stderr. But My Haswell (also running an HD7950) truncates the stderr after the Initial wait always at the same place. Although the FX-8350 had many more errors per day than the Haswell. Iteration area: 560000 Chunk estimate: 1 Num chunks: 2 Chunk size: 559104 Added area: 558208 Effective area: 1118208 Initial wait: 16 ms </stderr_txt> ]]> This is where my Haswell-E always truncates the STDERR Never had a hang though |
|
20)
Message boards :
Number crunching :
What is the cause of these 'validate errors'
(Message 63546)
Posted 9 May 2015 by Tom* Post: Hi Keith, I view the validate errors on my Haswell i7 a little differently Yes my amd fx-8350 always truncates the whole stderr. But My Haswell (also running an HD7950) truncates the stderr after the Initial wait always at the same place. Although the FX-8350 had many more errors per day than the Haswell. Iteration area: 560000 Chunk estimate: 1 Num chunks: 2 Chunk size: 559104 Added area: 558208 Effective area: 1118208 Initial wait: 16 ms </stderr_txt> ]]> |
Next 20
©2024 Astroinformatics Group