The migrating went just fine btw. In a few days I will replace my two gigabit sync channels with a single 10GBE crossover link. The two links have ip-address 10.199.198.1 (and .2 on partner) and 10.199.199.1 (and .2 on partner). If I install the 10GBE cards on one server at the time, then change the...

Is it possible from Vsphere to use more than 1 fixed path? I thought i could only choose RR (all paths) or fixed (one path). MRU is also just one path I think. I have 4 paths to the fast SSD storage and two to the "slow" SAS storage. Is it possible to use the 4 paths (RR) to the SSD storag...

I will just update this after some days of usage. This solution works GREAT! Database performance is MANY TIMES better than before for some workloads because of the SSD drives. I present single SSD's (image file) as single LUN's, with the SAS server as secondary. The only catch in this design is tha...

Perfect!
Thanks!

Is it possible to backup the Starwind configuration and reinstall Windows + import settings after? I have a secondary node in a HA pair that I need to reinstall (disk corruption on OS disk). It would be nice to be able to reinstall Windows on a new disk and just import the Starwind setup. Recreating...

I did a quick test, and this works just as I hoped. When I use fixed path to the SSD HA-node I get very nice IO ( >= 4000 read AND random IOs). When using the SATA -path I get <=500 IO's). Write speed is now limited by my 1Gbit sync channel. Is it possible to use more than one nic for sync? I might ...

Thank you for the reply. The load is actually not going to be very high, but I want high performance for the terminal services VM's. I think average load will be very low, with short bursts only. I have 2 SSD datastores now with a single SW host, and the performance of the few VM's hosted on these S...

To be clear. I want to use Starwind HA to have a replica of my SSD datastores, but I do not want to use SSD's on the secondary HA node. Reads should primarily go from the SSD HA-node, so no RR load balancing (but failover to the other HA node if the primary goes down).

I have a design question. I do not know if my thinking is correct, so please comment. I'm thinking of trying something like this if this works the way I hope it will: Server 1 (primary) with 16x 256GB SSD (MLC), utilizing maybe 200GB of each. No RAID, just single drives as LUN's. Server 2 (secondary...