Before I get too far into this, I primarily blame SAV 10 for being a resource hog, but I'm wondering if there's a way I can boost the performance to an acceptable level.
In brief, I set up a new PC the other day running 64-bit Ubuntu, installed 64-bit Workstation 7 for Linux on it, and set up a 32-bit XP VM with 2 VCPUs, 1 GB RAM and a 40-GB preallocated virtual disk. So far, so good -- performance was such that I could watch video in the VM with only minor stuttering.
Then, so I could meet the host requirements for our corporate VPN and use it for remote access to work, I downloaded and installed SAV 10 from our intranet. Immediately the performance went in the toilet even as the VM's CPU usage doubled. I was waiting several minutes after clicking a button or icon to see the effect. I tried disabling auto-protect but it only had marginal benefit and actually seemed to reactivate itself quickly. Uninstalling SAV completely resolved all issues (I could actually see performance improve during the uninstall process as each component was shut down).
Is there a simple known fix for this issue, or is it just a case of Symantec being an irredeemable pig? I've read that Symantec does not support SAV 10 on a VM, but I don't know what the extent of that non-support is.
Try it again with 1 vCPU.
Newer Linux kernels have a "problem" (or should I say bug?) scheduling vSMP VMs.
... is it just a case of Symantec being an irredeemable pig?
There is nothing to add
___________________________________
VMX-parameters- Workstation FAQ -[ MOA-liveCD|http://sanbarrow.com/moa241.html] - VM-Sickbay
Try it again with 1 vCPU.
Newer Linux kernels have a "problem" (or should I say bug?) scheduling vSMP VMs.
Disabling Auto-Protect can help, but I don't know if that would be allowed by your corporate policy.
David Strebel
If you find this information useful, please award points for "correct" or "helpful"
Try it again with 1 vCPU.
Newer Linux kernels have a "problem" (or should I say bug?) scheduling vSMP VMs.
That worked perfectly! I'm actually getting better performance with SAV now than I was without it before.
Can anybody identify where you found some detail on this for the Linux kernel?The kernel thread archives don't seem to have anything that leads me to believe they have seen the problem. I have to agree however, after dropping to a single CPU and single thread in my VM the responsiveness of the VM has been generally better.
Running Win7 X64 under WS7 on 2.6.31-17 with SAV 11.0 in the VM.