• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

vSphere Virtual Flash?

TeeJayHoward

Limpness Supreme
2FA
Joined
Feb 8, 2005
Messages
13,283
Could someone please explain to me how Virtual Flash Host Swap Cache and Virtual Flash Read Cache work? I get that they speed up VMs somehow through the use of a local SSD. What I want to know is what's behind the magic. HOW does it work?
 
vFRC is a local host based read cache.

Host swap cache allows the host to swap (which should be avoided) to an SSD rather than to a presumed slower datastore.
 
vFRC is a local host based read cache.
Could you please explain this to me like I'm five? Does it just store a copy of the .vmdk on the local SSD and execute reads from there? What block size do I want? What happens if I give it 1GB of vFRC for a 16GB disk? What about 16GB of vFRC? 32GB?

Host swap cache allows the host to swap (which should be avoided) to an SSD rather than to a presumed slower datastore.
So if you overprovision RAM and it needs to swap due to the host running out of physical memory, it will do it to an SSD rather than the .vswp file on the other datastore?
 
Could you please explain this to me like I'm five? Does it just store a copy of the .vmdk on the local SSD and execute reads from there? What block size do I want? What happens if I give it 1GB of vFRC for a 16GB disk? What about 16GB of vFRC? 32GB?

You assign an eligable (SSD & Local) drive to the FRC pool of a host which can then be allocated to the disks of VMs running on that host. You give each disk of each vm an amount of cache and the block size for it to use. So it isn't necessarily a one click feature to enable you need to monitor the guest to be given cache to see it's access patterns to ensure that you don't waste cache space allocating too large a block which won't be efficiently filled.

The reads carried out by the VM will then be cached so that it can be retrieved (if not modified) without going to shared or slower local storage.

So if you overprovision RAM and it needs to swap due to the host running out of physical memory, it will do it to an SSD rather than the .vswp file on the other datastore?

Once it has gone through all the other stages (compression, page sharing, balloning, guest swapping) then yes. Although if you have to swap then it's better to do it in the guest (where it knows what it is swapping) rather than the host which should be a last resort. So I can't imagine this is a highly used feature.
 
Once it has gone through all the other stages (compression, page sharing, balloning, guest swapping) then yes. Although if you have to swap then it's better to do it in the guest (where it knows what it is swapping) rather than the host which should be a last resort. So I can't imagine this is a highly used feature.

Don't forget that it only does those stages if VMTools is installed on the VM. If it doesn't, then the hypervisor goes straight for "I own your RAM"
 
Back
Top