And too high latency!
Hi all
I'm having "issues" with testing my storage solution with Jetstress 2013.
Setup is: VMware VM with 8 vCPUs, 24GB RAM and 500GB raw mapped device from VMAX box.
Device is thin provisioned on RAID 7+1 FC SCSI disk group.
Jetstress setup is as follows:
disk subsystem throughput test,thread autotuning, test for performance, enabled multihost and BDM
| Thread Count | 150 |
| Minimum Database Cache | 128,0 MB |
| Maximum Database Cache | 1024,0 MB |
| Insert Operations | 40% |
| Delete Operations | 20% |
| Replace Operations | 5% |
| Read Operations | 35% |
| Lazy Commits | 70% |
| Run Background Database Maintenance | True |
| Number of Copies per Database | 2 |
With this setup I'm getting about 4500IOPS, but 24msec and 14msec for write/read operations.
EMC guy suggest droping IOPS requirements and then I would get great response times... Suppose I would accomplish that with manually setting thread count? Is the thread count total or per database?
Can anybody explain what is going on? Front-end directors are not choked...
