Message boards : Questions and problems : How many processors?
Message board moderation
Author | Message |
---|---|
![]() Send message Joined: 17 Oct 13 Posts: 3 ![]() |
So here's a new one for me. I've been running BOINC on Windows for well over a decade, and recently added a couple of Mac OSX machines. The first OSX m/c went on fine. The second one I added earlier this evening and I get this start-up in the Event Log - Thu Oct 17 20:44:13 2013 | | No config file found - using defaults Thu Oct 17 20:44:13 2013 | | Starting BOINC client version 7.0.65 for x86_64-apple-darwin Thu Oct 17 20:44:13 2013 | | log flags: file_xfer, sched_ops, task Thu Oct 17 20:44:13 2013 | | Libraries: libcurl/7.26.0 OpenSSL/1.0.1c zlib/1.2.5 c-ares/1.9.1 Thu Oct 17 20:44:13 2013 | | Data directory: /Library/Application Support/BOINC Data Thu Oct 17 20:44:13 2013 | | Processor: 4 GenuineIntel Intel(R) Core(TM) i5 CPU M 540 @ 2.53GHz [x86 Family 6 Model 37 Stepping 2] Thu Oct 17 20:44:13 2013 | | Processor features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 PCLMULQDQ DTES64 MON DSCPL VMX SMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 SSE4.2 POPCNT AES Thu Oct 17 20:44:13 2013 | | OS: Mac OS X 10.8.5 (Darwin 12.5.0) Thu Oct 17 20:44:13 2013 | | Memory: 8.00 GB physical, 409.65 GB virtual Thu Oct 17 20:44:13 2013 | | Disk: 464.93 GB total, 409.41 GB free Thu Oct 17 20:44:13 2013 | | Local time is UTC +1 hours Thu Oct 17 20:44:13 2013 | | OpenCL: NVIDIA GPU 0: GeForce GT 330M (driver version 8.16.74 310.40.00.10f02, device version OpenCL 1.0, 256MB, 256MB available, 106 GFLOPS peak) Thu Oct 17 20:44:13 2013 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 9214856; resource share 100 Thu Oct 17 20:44:13 2013 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7118107; resource share 100 Thu Oct 17 20:44:13 2013 | | General prefs: from http://climateprediction.net/ (last modified 03-Jan-2012 15:53:18) Thu Oct 17 20:44:13 2013 | | Host location: none Thu Oct 17 20:44:13 2013 | | General prefs: using your defaults Thu Oct 17 20:44:13 2013 | | Reading preferences override file Thu Oct 17 20:44:13 2013 | | Preferences: Thu Oct 17 20:44:13 2013 | | max memory usage when active: 4096.00MB Thu Oct 17 20:44:13 2013 | | max memory usage when idle: 7372.80MB Thu Oct 17 20:44:13 2013 | | max disk usage: 100.00GB Thu Oct 17 20:44:13 2013 | | (to change preferences, visit a project web site or select Preferences in the Manager) Thu Oct 17 20:44:13 2013 | | Not using a proxy Thu Oct 17 20:44:14 2013 | Einstein@Home | Restarting task p2030.20130203.G204.12-00.97.N.b0s0g0.00000_103_1 using einsteinbinary_BRP4 version 133 (BRP4SSE) in slot 0 Thu Oct 17 20:44:14 2013 | Einstein@Home | Restarting task h1_0755.50_S6Directed__S6CasAf40a_755.65Hz_760_1 using einstein_S6CasA version 105 in slot 1 Thu Oct 17 20:44:14 2013 | Einstein@Home | Restarting task h1_0755.50_S6Directed__S6CasAf40a_755.65Hz_759_0 using einstein_S6CasA version 105 in slot 2 Thu Oct 17 20:44:14 2013 | Einstein@Home | Restarting task h1_0533.90_S6Directed__S6CasAf40a_534.55Hz_48_1 using einstein_S6CasA version 105 in slot 3 I'm running BOINC 7.0.65 on a MacBook Pro with this config - Hardware Overview: Model Name: MacBook Pro Model Identifier: MacBookPro6,2 Processor Name: Intel Core i5 Processor Speed: 2.53 GHz Number of Processors: 1 Total Number of Cores: 2 L2 Cache (per Core): 256 KB L3 Cache: 3 MB Memory: 8 GB Processor Interconnect Speed: 4.8 GT/s Boot ROM Version: MBP61.0057.B0F SMC Version (system): 1.58f17 Serial Number (system): W80411KTAGV Hardware UUID: 788E6ED9-FC07-59B3-A333-A65209EBD70E The behaviour that's unusual is that while 4 tasks run, only one actually does any work. The remaining 3 count up/down time, but don't record any % work done. Closing/reopening BOINC Manager resets these 3 'strange tasks', whilst leaving the 'regular' task ticking away nicely. Looking at those listings I copied in, this m/c seems to think of itself as having 1 processor with dual core, while BOINC Manager detect 4 processors at start-up. So BM kicks-off 4 tasks, of which only 3 really run. This is a new one to me - I've never seen this on any Win m/c, or on the other Mac. Any ideas? I did all the usual things - restarted, uninstalled / reinstalled. No effect. Many thanks |
![]() Send message Joined: 29 Aug 05 Posts: 15640 ![]() |
When looking at http://ark.intel.com/products/50072/Intel-Core-i5-2540M-Processor-3M-Cache-up-to-3_30-GHz, it shows that the CPU has 2 cores, but runs 4 threads. It also shows that this processor supports Hyper-Threading technology, which means that for each physical core it can make a logical core that the operating system will see as a real core. Thus why BOINC says you've got 4 processors, or more correctly, 4 cores. See http://en.wikipedia.org/wiki/Hyper-threading for the ins and outs. BOINC will by default use all the processors that the operating system says there are available, which is why you have 4 tasks starting up. When applications do not show any percentage of work done, you best report this at the affected project's web site. The actual project applications are developed and maintained by the project, not by BOINC. So in this case, you best post about it in the Problems and Bug Reports forum at Einstein. |
![]() Send message Joined: 17 Oct 13 Posts: 3 ![]() |
Many thanks Jord, I'll do that. However on other (Windows) machines I have Einstein@Home running quite happily on all 4 cores. Cheers |
![]() Send message Joined: 29 Aug 05 Posts: 15640 ![]() |
About that... http://einstein.phys.uwm.edu/hosts_user.php?userid=639474 is your list of computers at Einstein. Host 9214856 and 8806913 are essentially the same system. 9214856 all of a sudden had the 12.5.0 version of OS X and you've aborted all the work on that system; 8806913 couldn't finish a lot of the work it's been given in time for it to validate. It was already resent to other systems that had returned it before your system did. So something fishy is going on anyway. What did you do with that system after the 27 Sep 2013? Something that went wrong and did you put a backup back? |
![]() Send message Joined: 17 Oct 13 Posts: 3 ![]() |
Ah, no - a bit siller / more-complex than that...! Assuming these host ids are allocated chronologically, then 8806913 was running unattended and not network-connected at home (long story) and I didn't get back home for a fortnight - unpredicted absence - so you're right, the tasks went out of time and I lost them. The other m/c (9214856) I just acquired - not aware of having upgraded the OS myself, at least not explicitly (though I'm no Mac user!), so I think it's been on that level since I acquired it last week. And yes, when I saw the behaviour I described above, I simply aborted the lot - why run tasks that don't seem to work... (I thought). So all in all, maybe not as fishy as you think! I'd also have to say that I've just seen the earlier m/c (8806913) also displaying this characteristic of running E@H tasks without seeming to make any progress. And a restart of BM simply zeros these tasks back to start - again and again (so yep, I aborted that lot too!). Is there something about how E@H tasks are built that means they have to run for a while before they start to register task progress? But you're probably not the best person to ask...! Cheers - Richard And thanks for your continued interest - it's encouraging to get feeedback so swiftly. |
![]() Send message Joined: 29 Aug 05 Posts: 15640 ![]() |
Is there something about how E@H tasks are built that means they have to run for a while before they start to register task progress? But you're probably not the best person to ask...! Hence why I pointed out the E@H forums earlier. :-) It's their application, they ought to know its ins and outs. Thanks for the clear-up on the systems. Two systems then, check. :-) |
Copyright © 2025 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.