HI
We have been asked by our IT department to move from our current physical solution to a VM environment. I am not that clued up on VM.
I looked from some benchmark tests to run so i can see a comparison between our live and new VM we have been presented. Please see below for results.
To me the VM looks slower. Can someone advice please and offer advice on any other tests we can run please.
We are running Redhat Enterprise 6.4.
Thanks in Advanced
Treds
[root@VIRTAL_SERVER SSD]# dd if=/dev/urandom of=store bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 131.26 s, 8.0 MB/s
[root@VIRTAL_SERVER SSD]# dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.502956 s, 2.1 GB/s
[root@VIRTAL_SERVER SSD]# dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.4697 s, 2.2 GB/s
[root@VIRTAL_SERVER SSD]# dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.461993 s, 2.3 GB/s
Existing physical. 172.27.90.3
[root@PHYSICAL_SERVER ~]# dd if=/dev/urandom of=store bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 128.746 s, 8.1 MB/s
You have new mail in /var/spool/mail/root
[root@PHYSICAL_SERVER ~]# *dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.551658 s, 1.9 GB/s
You have new mail in /var/spool/mail/root
[root@PHYSICAL_SERVER ~]# dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.273119 s, 3.8 GB/s
[root@PHYSICAL_SERVER ~]# dd if=store of=/dev/null bs=100M
10+0 records in
10+0 records out
1048576000 bytes (1.0 GB) copied, 0.274492 s, 3.8 GB/s
I would be the first surprised if it were faster ( the VM...) but the point is more : Why? what argument was presented to migrate? The architecture of impementation is far more important over relative performance e.g. gain on downtime when maintenance, possibilities of swapping servers between sites etc...
Many will recommend iozone (can take a long time to run). But from my own experience, you can get away with a simple bonnie++ in 99% of all cases with regards to accurate performance measurements. For Bonnie++ use the RW statistics as your "general performance" guideline.
The only benchmark that makes any sense to me is your normal workload.
Suppose IT gave you a VM that allowed twice as many CPUs and/or a worderful array of SSD disks as you have on your physical machine, and that it allowed your workload to be processed in far less time on the VM. That would be a definite plus.
At a university computer center, whenever we would make changes, we would run a sample of user applications. If we saw a large increase or decrease in real time, we would look closely to try to determine the cause (it was almost always a failure of new code, many cases would crash, decreasing the real time, etc.).
Running benchmarks for a few megabytes would not be meaningful for me. I have run bonnie++ on different configurations (various RAID combinations, for example), so that might be useful if you let it run long enough -- 20-60 minutes on your physical machine, then compare it to runs on the VM.
Synthetic benchmarks are useful if they are similar to your day-to-day work, otherwise, they are perhaps better suited for water-cooler discussions.