qemu with wok?

I forgot the point of this post. So I’ll post it..

Hit my head a few months ago now (I keep doing that..). Ended up “recycling” my now 10 year old pc into a general purpose play-house. I will get to the QEMU and WOK soon – I promise. Just let me rant a bit about the fun parts of this build before it ended up as a virtual server.

Now, it gets interesting. I of course dug the cleaning kit, remounted and buffed the system a bit – and stuck a few additional components in it. All components needs a home ya know.

So, spec-wise it’s not the powerhouse it used to be.

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              4
On-line CPU(s) list: 0-3
Thread(s) per core:  1
Core(s) per socket:  4
Socket(s):           1
NUMA node(s):        1
Vendor ID:           GenuineIntel
CPU family:          6
Model:               30
Model name:          Intel(R) Core(TM) i5 CPU         750  @ 2.67GHz
Stepping:            5
CPU MHz:             1399.282
CPU max MHz:         2793.0000
CPU min MHz:         1197.0000
BogoMIPS:            5360.25
Virtualization:      VT-x
L1d cache:           32K
L1i cache:           32K
L2 cache:            256K
L3 cache:            8192K
NUMA node0 CPU(s):   0-3
Flags:               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 ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt lahf_lm pti ssbd ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid dtherm ida flush_l1d

Well – the cpu still can do something – after a BIOS-battle with the gigabyte P55 motherboard, it actually rocks 3.1 boosts on all cores when it wants too (note that lscpu greps cpu max without knowing boost).

Here comes the more fun part. I found tons of old drives – and just threw them in there. Ended up with a raid 5 with spinning old disks, two older SSD’s as cache-ish disks, and two additional spinning as “cache” disks as well. Since it has 16GB of ram, I do have a 4gb RAM-drive as well ;).

The fun part is the speed of using tempfs in RAM … even with this old hog;

# sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.407901 s, 2.6 GB/s
root@cruncher:/cache4# dd if=tempfile of=/dev/null bs=1M count=1024
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB, 1.0 GiB) copied, 0.217579 s, 4.9 GB/s

But, I still wanted more fun. Ended up finding a new GTX 1050ti “cheap”, the results are in a previous post. Wow. The compute power in this one now *eherhm*!

# ./bandwidthTest 
[CUDA Bandwidth Test] - Starting...
Running on...

 Device 0: GeForce GTX 1050 Ti
 Quick Mode

 Host to Device Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)	Bandwidth(MB/s)
   33554432			6073.5

 Device to Host Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)	Bandwidth(MB/s)
   33554432			5988.2

 Device to Device Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)	Bandwidth(MB/s)
   33554432			95273.4

Result = PASS
# ./UnifiedMemoryPerf 
GPU Device 0: "GeForce GTX 1050 Ti" with compute capability 6.1

Running ........................................................

Overall Time For matrixMultiplyPerf 

Printing Average of 20 measurements in (ms)
Size_KB	 UMhint	UMhntAs	 UMeasy	  0Copy	MemCopy	CpAsync	CpHpglk	CpPglAs
4	  0.297	  0.362	  0.391	  0.053	  0.112	  0.089	  0.105	  0.081
16	  0.262	  0.422	  0.542	  0.123	  0.156	  0.127	  0.181	  0.116
64	  0.587	  0.581	  0.645	  0.271	  0.303	  0.265	  0.331	  0.246
256	  1.241	  1.245	  1.474	  1.222	  0.849	  0.935	  0.742	  0.888
1024	  4.483	  4.370	  5.259	  7.363	  3.303	  3.371	  2.897	  2.817
4096	 18.399	 17.240	 20.935	 51.331	 16.100	 15.677	 14.981	 15.001
16384	 90.533	 87.024	103.719	379.888	 82.126	 82.115	 79.934	 79.662

NOTE: The CUDA Samples are not meant for performance measurements. Results may vary when GPU Boost is enabled.

Ok ok.. So lets get down to the QEMU parts. The CPU is not rocking HT, and that’s really not that of a bad thing to be honest. But, I needed something for a virtual host running a game-server (wreckfest ;) ).

So, main host is err, ubuntu I believe. Getting QEMU in is kinda standard. New to me tho. Then I found Wok, to serve it all over web since I figured I could use it for remote access, control and just easier navigation. Guess what. One gets spoiled a bit with VMware and other setups quite quickly. Wok hasn’t been maintaned really for quite a while – the defaults assumes a bit to much of the setup and the redirects with nginx plays flipper with your request for access…