climateprediction.net home page
Posts by Jesse Viviano

Posts by Jesse Viviano

1) Message boards : Number crunching : Transitioner is down (Message 58905)
Posted 23 Oct 2018 by Jesse Viviano
Post:
Never mind. The transitioner is back up. I had to detach and reattach to get the work unit reissued.
2) Message boards : Number crunching : Transitioner is down (Message 58904)
Posted 23 Oct 2018 by Jesse Viviano
Post:
I recently reported result 21337131 as a compute error. It still is listed as in progress. I then noticed that the transitioner is down. Could someone please bring it back up if it is not down due to maintenance? I would like my work unit to be reissued so that the computation error can be dealt with. I also need to know if I need to detach and reattach if the report is lost so that it can be marked as abandoned, so that the work unit can be reissued.
3) Message boards : Number crunching : Impossible to detach project climateprediction.net after changes in URL (Message 58426)
Posted 19 Jul 2018 by Jesse Viviano
Post:
This project does not use the BOINC validation system, so you do not have to worry about jobs waiting for validation.
4) Message boards : Number crunching : Upload server is out of disk space (Message 58416)
Posted 18 Jul 2018 by Jesse Viviano
Post:
I am getting the following errors when I try to upload some result files. Sample error messages are listed below:

7/18/2018 2:19:10 AM | climateprediction.net | Started upload of wah2_nam50_pdhh_200912_13_735_011566091_0_r348673607_1.zip
7/18/2018 2:20:05 AM | climateprediction.net | [error] Error reported by file upload server: can't open file wah2_nam50_pdhh_200912_13_735_011566091_0_r348673607_1.zip: No space left on device
7/18/2018 2:20:05 AM | climateprediction.net | Temporarily failed upload of wah2_nam50_pdhh_200912_13_735_011566091_0_r348673607_1.zip: transient upload error
7/18/2018 2:20:05 AM | climateprediction.net | Backing off 00:10:56 on upload of wah2_nam50_pdhh_200912_13_735_011566091_0_r348673607_1.zip
7/18/2018 2:26:53 AM | climateprediction.net | Started upload of wah2_nam50_pdbv_200812_13_735_011565889_0_r1787333323_1.zip
7/18/2018 2:27:49 AM | climateprediction.net | [error] Error reported by file upload server: can't open file wah2_nam50_pdbv_200812_13_735_011565889_0_r1787333323_1.zip: No space left on device
7/18/2018 2:27:49 AM | climateprediction.net | Temporarily failed upload of wah2_nam50_pdbv_200812_13_735_011565889_0_r1787333323_1.zip: transient upload error
7/18/2018 2:27:49 AM | climateprediction.net | Backing off 04:03:43 on upload of wah2_nam50_pdbv_200812_13_735_011565889_0_r1787333323_1.zip
5) Message boards : Number crunching : "No tasks available" for windows - despite 29454 in task queue (Message 58155)
Posted 27 Apr 2018 by Jesse Viviano
Post:
If you checked the project's server status page, you will see that it was last updated on April 1, 2018. Therefore, its contents are so stale that it should be disregarded for now.
6) Message boards : Number crunching : Downloads Failing (Message 58138)
Posted 22 Apr 2018 by Jesse Viviano
Post:
I enabled some debug messages in the event log, and determined that the download server's domain is download.cpdn.org, and its IP address is 129.69.193.73. I then tried to ping it. The pings failed. The download server is too busy, is down, or is unreachable.
7) Message boards : Number crunching : Model uploaded/finished/reported, but still in progress on the web (Message 56211)
Posted 12 May 2017 by Jesse Viviano
Post:
Here is one more task whose completion was reported but now lost: https://www.cpdn.org/cpdnboinc/result.php?resultid=20378119
8) Message boards : Number crunching : Model uploaded/finished/reported, but still in progress on the web (Message 55451)
Posted 8 Jan 2017 by Jesse Viviano
Post:
BOINC requires that the BOINC client report the completion of tasks before they are marked as complete to the server in order to prevent the server from trying to process tasks whose uploads are not yet complete (e.g. the internet connection failed during the middle of uploading one of the files of the result). Have you verified that the work unit is not sitting in your list of tasks waiting to be reported? If it is, then you could click the update button after selecting the project in the projects tab to force the BOINC client to transmit the completion message. Otherwise, you could wait for it to automatically contact the project, and the completion message will be transmitted then. If the task is not in the project tab and there are no more tasks for the project in the queue, you could detach from and reattach to the project to cause the server to mark the task as abandoned, so the work unit can be reprocessed by another computer. Doing this while there are other Climateprediction.net tasks left in the queue will cause you to lose progress in those tasks, so you should set the project to no new tasks mode to ensure that your computer does not get any work and that your computer could drain the queue of remaining tasks. You will still get credit for the lost task due to the trickles for the lost task that were reported.
9) Message boards : Number crunching : Credit (Message 55335)
Posted 11 Dec 2016 by Jesse Viviano
Post:
I think that the simplest way to display credit would be to use the sample_trivial_validator at https://boinc.berkeley.edu/trac/wiki/ValidationIntro with the --no_credit command line argument in order to let the trickle system grant credit with no interference from the validator. This would mark units as validated, allowing the scripts to display the credit instead of "pending".
10) Questions and Answers : Wish list : VirtualBox support (Message 51404)
Posted 13 Feb 2015 by Jesse Viviano
Post:
Since you have some 32-bit Linux projects, VirtualBox could be used to run those on Windows and Mac OS hosts. It could also allow you to run these Linux projects on Linux hosts that do not include the required libraries (e.g. 64-bit Linux hosts which do not have the required 32-bit libraries installed).
11) Questions and Answers : Windows : False positive (Message 51285)
Posted 20 Jan 2015 by Jesse Viviano
Post:
Please report the false positive to https://submit.symantec.com/false_positive/.
12) Questions and Answers : Wish list : Advanced CPU support (Message 51269)
Posted 18 Jan 2015 by Jesse Viviano
Post:
BOINC does supply a mechanism to supply different applications to different computers as seen in http://boinc.berkeley.edu/trac/wiki/AppPlanSpec, so supplying the best binary to each computer is just a configuration job for the server. Therefore, each version of the application will not need to worry about backwards compatibility because the scheduler will assign the best version of each application to each computer if it is configured correctly. However, I will acknowledge the difficulty and expense of developing different versions of each application for different computers. This is why I posted this in the wish list and do not expect it to be granted unless some major donor shows up or the OeRC budget is increased. Even if one of the two happens, my proposal would have to compete against every other item in the wish list.
13) Questions and Answers : Wish list : Advanced CPU support (Message 51266)
Posted 18 Jan 2015 by Jesse Viviano
Post:
I think that many crunchers would want support for more advanced CPUs. Since you already require SSE2 according to your system requirements page, many people would want support for more advanced CPU instruction sets to speed up the applications even more.

  • 64-bit versions of applications: This can help in three areas. The first and more important area is that x86 in 32 bit modes are register starved, and AMD fixed this issue when designing AMD64. Its 32-bit mode was designed when memory ran at the same speed as the processor, so memory operations were cheap back then. They are quite expensive today because DRAM and most caches are slower than the CPU core. Therefore, one study on some Pentium Pro processors cited in one of my old college textbooks found that they spent over half of their time waiting for the memory subsystem when executing code. Having the additional registers added by the AMD64 architecture allows the core to stay busy doing more real work and spend less time waiting for the memory system, and can sometimes keep 64-bit capable NetBurst CPUs from entering the pathologically energy-wasting replay mode by keeping more data in the registers rather than only in the memory system where a failure to keep data in the level 1 cache will guarantee entry into replay mode. The second area is that programs can directly use more than 4 gibibytes of DRAM. The third area is that 64-bit integers are supported, which is probably worthless for this application.
  • SSE3: This adds some flexibility to the 128-bit wide vector unit that might help maintain a higher consistent operation rate in some situations depending on the code, and therefore might or might not be helpful depending on your code.
  • AVX: This doubles the width of the floating point vector unit to 256 bits as compared to the 128-bit SSE/SSE2/SSE3 instruction sets. The integer vector unit is not affected by this instruction set.
  • FMA4: This instruction multiplies two numbers and keeps all of the bits of the product without rounding, adds a third number to the product, and then finally rounds the result, with the whole thing done in one cycle as an atomic operation. This instruction therefore doubles the peak floating point operations per second figure if an FMA operation is counted as two floating point operations. This instruction is required for AMD's Bulldozer processors to perform decently in floating point, because otherwise their floating point units are pathologically slow.
  • FMA3: This instruction does the same thing as FMA4, but requires that one of the source variables is overwritten with the result. AMD's Piledriver processors and above support this instruction as well as FMA4, and require either of them to be used to perform floating point at an acceptable speed because all of the processors of the Bulldozer family have floating point units that are otherwise garbage. Intel's Haswell processors and above support this instruction. Poor coordination between AMD and Intel and the discovery by Intel that FMA4 would require an extensive rework of its vector unit generated this confusion between which FMA instruction should be supported.
  • AVX2: This does two major things: it doubles the width of the integer vector unit to 256 bits as compared to the 128-bit SSE/SSE2/SSE3 instruction sets, and includes the FMA3 instruction. It is found in Intel's Haswell processors and above, and in AMD's Excavator and above.


EDIT: Explain that Piledriver and above members of the Bulldozer family require either FMA4 or FMA3.

14) Message boards : Number crunching : Scheduler process down? (Message 51178)
Posted 10 Jan 2015 by Jesse Viviano
Post:
Please report the false positive to Avast. These instructions might help you report it.
15) Message boards : Number crunching : Scheduler process down? (Message 51177)
Posted 10 Jan 2015 by Jesse Viviano
Post:
Thanks for fixing this.
16) Message boards : Number crunching : Scheduler process down? (Message 51163)
Posted 7 Jan 2015 by Jesse Viviano
Post:
My finished work unit result now has been reported. However, trickles that are newer than the scheduler outage are not showing up in the work unit result logs, as written about in this thread.
17) Message boards : Number crunching : Scheduler process down? (Message 51157)
Posted 6 Jan 2015 by Jesse Viviano
Post:
Thanks!
18) Message boards : Number crunching : Scheduler process down? (Message 51150)
Posted 6 Jan 2015 by Jesse Viviano
Post:
Here is the event log in regards to my attempting to report the trickle up message and the completed work unit.
1/6/2015 8:57:31 AM |  | cc_config.xml not found - using defaults
1/6/2015 8:57:31 AM |  | Starting BOINC client version 7.4.36 for windows_x86_64
1/6/2015 8:57:31 AM |  | log flags: file_xfer, sched_ops, task
1/6/2015 8:57:31 AM |  | Libraries: libcurl/7.39.0 OpenSSL/1.0.1j zlib/1.2.8
1/6/2015 8:57:31 AM |  | Data directory: C:\ProgramData\BOINC
1/6/2015 8:57:31 AM |  | Running under account Jesse Viviano
1/6/2015 8:57:31 AM |  | CUDA: NVIDIA GPU 0: GeForce GTX 580 (driver version 347.09, CUDA version 7.0, compute capability 2.0, 3072MB, 2933MB available, 1843 GFLOPS peak)
1/6/2015 8:57:31 AM |  | OpenCL: NVIDIA GPU 0: GeForce GTX 580 (driver version 347.09, device version OpenCL 1.1 CUDA, 3072MB, 2933MB available, 1843 GFLOPS peak)
1/6/2015 8:57:31 AM |  | Host name: JesseViviano-PC
1/6/2015 8:57:31 AM |  | Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU       X 980  @ 3.33GHz [Family 6 Model 44 Stepping 2]
1/6/2015 8:57:31 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt aes syscall nx lm vmx tm2 pbe
1/6/2015 8:57:31 AM |  | OS: Microsoft Windows 7: Ultimate x64 Edition, Service Pack 1, (06.01.7601.00)
1/6/2015 8:57:31 AM |  | Memory: 11.99 GB physical, 23.98 GB virtual
1/6/2015 8:57:31 AM |  | Disk: 223.47 GB total, 123.24 GB free
1/6/2015 8:57:31 AM |  | Local time is UTC -5 hours
1/6/2015 8:57:31 AM |  | VirtualBox version: 4.3.20
1/6/2015 8:57:31 AM | climateprediction.net | URL http://climateprediction.net/; Computer ID 1350458; resource share 100
1/6/2015 8:57:31 AM |  | Preferences:
1/6/2015 8:57:31 AM |  | max memory usage when active: 9208.27MB
1/6/2015 8:57:31 AM |  | max memory usage when idle: 11049.92MB
1/6/2015 8:57:31 AM |  | max disk usage: 111.73GB
1/6/2015 8:57:31 AM |  | (to change preferences, visit a project web site or select Preferences in the Manager)
1/6/2015 8:57:31 AM |  | Resetting file projects/pogs.theskynet.org_pogs/pogs_image01.png: md5 checksum failed for file
1/6/2015 8:57:31 AM |  | Not using a proxy
1/6/2015 8:57:32 AM | climateprediction.net | Sending scheduler request: To send trickle-up message.
1/6/2015 8:57:32 AM | climateprediction.net | Not requesting tasks: don't need (CPU: not highest priority project; NVIDIA GPU: )
1/6/2015 8:57:36 AM | climateprediction.net | Scheduler request failed: HTTP internal server error
1/6/2015 9:16:22 AM | climateprediction.net | Sending scheduler request: To send trickle-up message.
1/6/2015 9:16:22 AM | climateprediction.net | Not requesting tasks: don't need (CPU: not highest priority project; NVIDIA GPU: )
1/6/2015 9:16:25 AM | climateprediction.net | Scheduler request failed: HTTP internal server error
1/6/2015 10:08:07 AM | climateprediction.net | Sending scheduler request: To send trickle-up message.
1/6/2015 10:08:07 AM | climateprediction.net | Not requesting tasks: don't need (CPU: not highest priority project; NVIDIA GPU: job cache full)
1/6/2015 10:08:09 AM | climateprediction.net | Scheduler request failed: HTTP internal server error
1/6/2015 10:33:43 AM | climateprediction.net | Started upload of hadam3p_anz_m8ia_2012_1_009308664_0_12.zip
1/6/2015 10:36:09 AM | climateprediction.net | Finished upload of hadam3p_anz_m8ia_2012_1_009308664_0_12.zip
1/6/2015 10:42:17 AM | climateprediction.net | Started upload of hadam3p_anz_m8ia_2012_1_009308664_0_13.zip
1/6/2015 10:42:19 AM | climateprediction.net | Computation for task hadam3p_anz_m8ia_2012_1_009308664_0 finished
1/6/2015 10:50:37 AM | climateprediction.net | Finished upload of hadam3p_anz_m8ia_2012_1_009308664_0_13.zip
1/6/2015 11:41:31 AM | climateprediction.net | update requested by user
1/6/2015 11:41:38 AM | climateprediction.net | Sending scheduler request: Requested by user.
1/6/2015 11:41:38 AM | climateprediction.net | Sending trickle-up message
1/6/2015 11:41:38 AM | climateprediction.net | Reporting 1 completed tasks
1/6/2015 11:41:38 AM | climateprediction.net | Not requesting tasks: don't need (CPU: not highest priority project; NVIDIA GPU: not highest priority project)
1/6/2015 11:41:39 AM | climateprediction.net | Scheduler request failed: HTTP internal server error
19) Message boards : Number crunching : Scheduler process down? (Message 51146)
Posted 6 Jan 2015 by Jesse Viviano
Post:
The server status page is up and shows the scheduler as running, but I still get HTTP errors when BOINC tries to make a scheduler request. Is it overloaded, or is something else keeping scheduler requests from succeeding?
20) Message boards : Number crunching : Scheduler process down? (Message 51141)
Posted 5 Jan 2015 by Jesse Viviano
Post:
My BOINC client is trying to send a trickle-up message, and keeps getting HTTP errors. Second, the server status page returns a blank page as of this writing. (It has the correct HTML formatting when I viewed its source, but nothing else.) Has the scheduler process failed?


Next 20

©2024 climateprediction.net