Since I didn't want to totally hijack the Moodle BenchMark thread I thought I'd start a new post here.
Original Post https://moodle.org/mod/forum/discuss.php?d=335357#p1359485
Note: I'm a history teacher by trade, who ran a personal moodle site hosted by professional hosting companies who moved into IT last year when my school adopted iPads for all students. It was decided that I would run a moodle server for the school and that it would be hosted on one of our servers on campus. Last year I ran the server on 1 VM with 4 cpu cores and 24 gigs of ram, but since we got some new servers for other services, I was given a whole server for moodle.
Server Info
HP Proliant DL 360p Gen 8
2x Intel E5-2620v2 (6 cores + hyperthreading on each)
128 Gigs of Ram total
2x 7,200 RPM SAS 1tb drives (Raid 1 configuration) (datastore 1)
2x 12,00 RPM SAS 1.8tb drives (Raid 1 configuration) (datastore 2)
VMware vSphere 6 Hypervisor
- Hardware is split into 2 VM's - VM
VM1 - Moodle Web server / Moodledata on this server.
6 CPU cores, 48 gigs of ram, datastore 1
OS - Lubuntu 14.04, Apache, mySQL, Php5
VM2 - Moodle Database server
6 CPU cores, 64 gigs of ram, datastore 2
OS - Lubuntu 14.04, Apache, mySQL, Php5
Note: I split the server into 2 VM's because I can't allocate more than 10 CPU cores to 1 VM, and I wanted to use all 12 Cores. I am also a novice at VMware configuration. We are using the free version and so have some limitations in configuration options/tools available from what I understand. I may very well have something misconfigured at the VM level that is causing the issues I'm having.
I am willing to do almost anything to reconfigure this to get the performance that I should be getting - however, I am leading some training sessions on Moodle starting Monday August 1 and so I won't be able to pull the site down for a few days. I did attempt to upgrade to Moodle 3.1 and upgrade to Ubuntu 16.04 and PhP7 but ran into some problems with Ubuntu 16.04 on the server with the display being locked to 800x600, which isn't an issue most of the time but there are times when the GUI is useful. After being unable to resolve the issue I reset the server with Ubuntu 14.04 and did a fresh install of everything.
The moodle site itself handles the school's 1200 students and about 300 classes total. The most stressful time on the server last year came when we had about 500 students using the Quiz module at the same time. This year, I expect more teachers to use it and that number to climb to about 900 - 1000 using the Quiz module at the same time for semester/final exams.
Results of the Moodle BenchMark
UnixBench Results
System: : GNU/Linux
OS: GNU/Linux -- 3.13.0-92-generic -- #139-Ubuntu SMP Tue Jun 28 20:42:26 UTC 2016
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 4: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 5: Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz (4189.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
09:35:29 up 10 days, 23:54, 2 users, load average: 0.26, 0.10, 0.07; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Sun Jul 31 2016 09:35:29 - 10:03:54
6 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 23512033.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3033.3 MWIPS (9.7 s, 7 samples)
Execl Throughput 1526.6 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 690641.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 204572.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1593391.7 KBps (30.0 s, 2 samples)
Pipe Throughput 1538048.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 30865.1 lps (10.0 s, 7 samples)
Process Creation 4424.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4324.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1783.5 lpm (60.0 s, 2 samples)
System Call Overhead 2726923.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 23512033.0 2014.7
Double-Precision Whetstone 55.0 3033.3 551.5
Execl Throughput 43.0 1526.6 355.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 690641.7 1744.0
File Copy 256 bufsize 500 maxblocks 1655.0 204572.2 1236.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 1593391.7 2747.2
Pipe Throughput 12440.0 1538048.9 1236.4
Pipe-based Context Switching 4000.0 30865.1 77.2
Process Creation 126.0 4424.6 351.2
Shell Scripts (1 concurrent) 42.4 4324.8 1020.0
Shell Scripts (8 concurrent) 6.0 1783.5 2972.5
System Call Overhead 15000.0 2726923.7 1817.9
========
System Benchmarks Index Score 932.3
------------------------------------------------------------------------
Benchmark Run: Sun Jul 31 2016 10:03:54 - 10:32:34
6 CPUs in system; running 6 parallel copies of tests
Dhrystone 2 using register variables 156478170.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 20502.5 MWIPS (9.7 s, 7 samples)
Execl Throughput 15183.4 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 833217.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 239819.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2230343.1 KBps (30.0 s, 2 samples)
Pipe Throughput 9994904.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1482861.6 lps (10.0 s, 7 samples)
Process Creation 30072.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 20509.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4632.4 lpm (60.0 s, 2 samples)
System Call Overhead 6365870.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 156478170.2 13408.6
Double-Precision Whetstone 55.0 20502.5 3727.7
Execl Throughput 43.0 15183.4 3531.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 833217.9 2104.1
File Copy 256 bufsize 500 maxblocks 1655.0 239819.3 1449.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 2230343.1 3845.4
Pipe Throughput 12440.0 9994904.8 8034.5
Pipe-based Context Switching 4000.0 1482861.6 3707.2
Process Creation 126.0 30072.7 2386.7
Shell Scripts (1 concurrent) 42.4 20509.8 4837.2
Shell Scripts (8 concurrent) 6.0 4632.4 7720.6
System Call Overhead 15000.0 6365870.6 4243.9
========
System Benchmarks Index Score 4111.2