Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 6
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 2337 times and has 5 replies Next Thread
immortal
Cruncher
USA
Joined: Nov 4, 2018
Post Count: 8
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
sloooooow

Hey all.

So I have a T5500 with dual Xeon 5675's in it and it runs pretty well. About 70,000 points a day 135 results. 4 hour or so MCM job.

I was thinking about upgrading my crappy T3500 system that the kids use sometimes but mostly just sits there doing nothing. After thinking about parts cost and electricity etc I figured I would just grab another cloud dedicated server. I've had dedicated servers before but I have never found 12 core/24 thread machines. Usually its a 4 or 8 core 2 GHz machine that doesn't really put up much productivity wise, not worth the cost.

I found a dual Xeon 5650 for $30 a month and fired up boinc and it runs pretty quick, basically the same as my T5500 points/results wise. 4 hour MCM job.

Then rented another identical 5650 :) It runs great.

Then a third (which is $5 a month cheaper) except this one has a 120 GB SSD drive while the other two have 1TB mechanical drives. Otherwise they seem identical. The third machine is slower than crap. Its only averaging 20,000 points and 15 results a day. I pulled up BoincTasks to see whats up and its taking 20+ hours for each task. One MiP is 80% complete and has been running 30 hours now.

Any idea on what to check for? The settings seem identical to the other machines. If I ssh into the machine it is slower than heck. Its showing 100% CPU usage, 3GB of 8 used, but the CPU temp is only around 55 degrees. The other machines are near 70 so its not running quite as hot. All my machines other than my main one are running Ubuntu 18.04 headless.



Thanks in advance
----------------------------------------
[Edit 1 times, last edit by immortal84660 at Jun 17, 2020 2:11:07 AM]
[Jun 17, 2020 2:07:45 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Falconet
Master Cruncher
Portugal
Joined: Mar 9, 2009
Post Count: 3315
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: sloooooow

Elapsed time to CPU time looks good.
Throttling?

I would also limit MIP to a max of 4 tasks (1 per 4 MB of L3 cache on that CPU, you can do this in the device profiles). It's a known issue with MIP that after a certain number of tasks running, it brings efficiency down quite a bit.


What is the name of this cloud provider, btw?
----------------------------------------


- AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W
- AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W
- AMD Ryzen 7 7730U 8C/16T 3.0 GHz
[Jun 17, 2020 8:41:24 AM]   Link   Report threatening or abusive post: please login first  Go to top 
William Albert
Cruncher
Joined: Apr 5, 2020
Post Count: 41
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: sloooooow

Check that the runtime clock speeds are what you'd expect. The CPU may be throttled by either Linux's governor, or the system's BIOS.
[Jun 17, 2020 2:25:45 PM]   Link   Report threatening or abusive post: please login first  Go to top 
immortal
Cruncher
USA
Joined: Nov 4, 2018
Post Count: 8
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: sloooooow

I will update the profile to only allow 4 MIP tasks.

I don't believe it is being throttled. I can't check the BIOS but it looks as expected from the terminal. Seems to be running at 100%.

My three 5650 servers are hosted at joesdatacenter.com. The other two servers are running great, its only this particular one that is having issues.

steve@joe3:~$ inxi -Fx
System: Host: joe3 Kernel: 4.15.0-106-generic x86_64 bits: 64 gcc: 7.5.0 Console: tty 0
Distro: Ubuntu 18.04.4 LTS
Machine: Device: server System: Cisco Systems product: R200-1120402W v: 0 serial: N/A
Mobo: Cisco Systems model: R200-1120402W v: 74-7340-02 serial: N/A
BIOS: Cisco Systems v: C200.1.4.1.0.071320110702 date: 07/13/2011
CPU(s): 2 6 core Intel Xeon X5650s (-MT-MCP-SMP-) arch: Nehalem rev.2 cache: 24576 KB
flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 64000
clock speeds: max: 2661 MHz 1: 2933 MHz 2: 2933 MHz 3: 2933 MHz 4: 2933 MHz 5: 2933 MHz 6: 2933 MHz
7: 2933 MHz 8: 2933 MHz 9: 2933 MHz 10: 2933 MHz 11: 2933 MHz 12: 2933 MHz 13: 2933 MHz 14: 2933 MHz
15: 2933 MHz 16: 2933 MHz 17: 2933 MHz 18: 2933 MHz 19: 2933 MHz 20: 2933 MHz 21: 2933 MHz
22: 2933 MHz 23: 2933 MHz 24: 2933 MHz
Graphics: Card: Matrox Systems MGA G200e [Pilot] ServerEngines (SEP1) bus-ID: 0e:00.0
Display Server: N/A driver: mgag200 tty size: 207x61 Advanced Data: N/A out of X
Network: Card-1: Intel 82576 Gigabit Network Connection driver: igb v: 5.4.0-k port: 1020 bus-ID: 0b:00.0
IF: ens1f0 state: up speed: 1000 Mbps duplex: full mac: 50:57:a8:af:25:94
Card-2: Intel 82576 Gigabit Network Connection driver: igb v: 5.4.0-k port: 1000 bus-ID: 0b:00.1
IF: ens1f1 state: down mac: 50:57:a8:af:25:95
Drives: HDD Total Size: 120.0GB (5.7% used)
ID-1: /dev/sda model: TEAM_T253X1120G size: 120.0GB
Partition: ID-1: / size: 106G used: 2.8G (3%) fs: ext4 dev: /dev/sda1
ID-2: /boot size: 465M used: 59M (14%) fs: ext4 dev: /dev/sda2
ID-3: swap-1 size: 4.05GB used: 0.00GB (0%) fs: swap dev: /dev/sda5
RAID: No RAID devices: /proc/mdstat, md_mod kernel module present
Sensors: System Temperatures: cpu: 58.0C mobo: N/A
Fan Speeds (in rpm): cpu: N/A
Info: Processes: 300 Uptime: 22:20 Memory: 3414.0/7909.6MB Init: systemd runlevel: 5 Gcc sys: N/A
Client: Shell (bash 4.4.201) inxi: 2.3.56
[Jun 17, 2020 11:16:25 PM]   Link   Report threatening or abusive post: please login first  Go to top 
immortal
Cruncher
USA
Joined: Nov 4, 2018
Post Count: 8
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: sloooooow

top - 17:18:08 up 22:26, 1 user, load average: 25.24, 25.12, 25.04
Tasks: 295 total, 25 running, 137 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.1 us, 0.6 sy, 99.3 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 8099472 total, 3460608 free, 3389552 used, 1249312 buff/cache
KiB Swap: 3951612 total, 3951612 free, 0 used. 4401248 avail Mem

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
5110 boinc 39 19 73700 72920 2384 R 100.0 0.9 1109:50 wcgrid_mcm1_map
6407 boinc 39 19 73700 72920 2384 R 100.0 0.9 702:00.50 wcgrid_mcm1_map
7124 boinc 39 19 73700 72916 2384 R 100.0 0.9 495:05.47 wcgrid_mcm1_map
5307 boinc 39 19 195796 134516 2496 R 99.7 1.7 1043:22 wcgrid_opn1_aut
6064 boinc 39 19 175132 113480 2496 R 99.7 1.4 805:56.50 wcgrid_opn1_aut
6274 boinc 39 19 73700 72920 2384 R 99.7 0.9 741:58.62 wcgrid_mcm1_map
6843 boinc 39 19 174240 112692 2496 R 99.7 1.4 591:51.35 wcgrid_opn1_aut
7092 boinc 39 19 392324 317188 53848 R 99.7 3.9 507:00.38 wcgrid_mip1_ros
7295 boinc 39 19 73700 72920 2384 R 99.7 0.9 456:09.78 wcgrid_mcm1_map
7299 boinc 39 19 73700 72920 2384 R 99.7 0.9 455:08.70 wcgrid_mcm1_map
7418 boinc 39 19 73700 72920 2384 R 99.7 0.9 423:01.25 wcgrid_mcm1_map
7805 boinc 39 19 73700 72916 2384 R 99.7 0.9 318:48.95 wcgrid_mcm1_map
7931 boinc 39 19 392264 316224 53832 R 99.7 3.9 279:54.66 wcgrid_mip1_ros
8030 boinc 39 19 451732 376620 53864 R 99.7 4.6 246:00.07 wcgrid_mip1_ros
3949 boinc 39 19 73700 72916 2384 R 99.4 0.9 1275:05 wcgrid_mcm1_map
5736 boinc 39 19 73700 72920 2384 R 99.4 0.9 903:56.82 wcgrid_mcm1_map
6491 boinc 39 19 430716 355604 53864 R 99.4 4.4 671:45.38 wcgrid_mip1_ros
6771 boinc 39 19 73700 72920 2384 R 99.4 0.9 618:43.76 wcgrid_mcm1_map
7460 boinc 39 19 186816 125560 2496 R 99.4 1.6 412:09.21 wcgrid_opn1_aut
8390 boinc 39 19 203704 138148 47872 R 99.4 1.7 138:04.54 wcgrid_mip1_ros
5779 boinc 39 19 73700 72916 2384 R 98.4 0.9 892:00.77 wcgrid_mcm1_map
5452 boinc 39 19 73700 72920 2384 R 98.1 0.9 998:01.20 wcgrid_mcm1_map
5920 boinc 39 19 397504 322304 53780 R 96.8 4.0 840:52.29 wcgrid_mip1_ros
6987 boinc 39 19 458196 382852 53848 R 95.9 4.7 542:29.07 wcgrid_mip1_ros
9344 steve 20 0 42344 3744 2884 R 4.7 0.0 0:01.23 top
855 boinc 30 10 301400 22128 12888 S 2.2 0.3 41:33.72 boinc
[Jun 17, 2020 11:21:13 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Falconet
Master Cruncher
Portugal
Joined: Mar 9, 2009
Post Count: 3315
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: sloooooow

I know it's 1 month late but I have tried this service provider and now have have 2 machines running. Both are dual X5650 running at 2.93 GHz.

I think I got the same 120 GB SSD machine as you did because it had the same issue you described.
I contacted support and they said it was a throttling setting that a previous customer had asked for. It is now running as it should.
----------------------------------------


- AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W
- AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W
- AMD Ryzen 7 7730U 8C/16T 3.0 GHz
[Jul 17, 2020 9:11:16 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread