Hi mathieugont
Thank you for providing your hardware specs; just to quote them again:
- 1x local ATA disk (1.8TB HDD) per server (Western Digital?), partition format GPT
- ESXi-6.7.0.13006603
- Each server's HDD is a datastore based on VMFS-6.82
- no vSan (i guess)
I see two problems here:
1. You don't have vSAN, as you guessed, but that is not a requirement. What is required is shared storage (i.e. some kind of SAN, iSCSI, NFS, etc.).
Per the VMmark User's Guide:
System Under Test Storage Requirements
Configure the system under test with enough shared datastore capacity to hold the disks and paging files for
all the virtual machines required for the VMmark Benchmark runs. This is approximately 891GB per tile (not
counting the storage space required by the prime and tile clients, addressed in “Storage Requirements for
VMmark Clients” on page 35).
NOTE In order to provide source and target datastores for the storage relocation operations that are part of
the benchmark, VMmark requires a minimum of two datastore partitions.
If using VMware vSAN™ as the primary storage solution, a secondary storage solution will be needed for
infrastructure operations.
Additionally, the benchmark requires that all ESXi hosts used in a test have access to the same shared storage.
2. The fact that you only have one local (not shared) HDD per ESXi host / server is problematic for several reasons:
- 1 x local ATA 1.8TB HDD = probably 10K RPM (not SSD) - the IOPS (throughput) will be very low (~500?) - this is well below the requirements of VMmark.
Per the User's Guide:
The VMmark benchmark needs high-throughput, low-latency storage. While the exact bandwidth
requirements will vary based on other aspects of the environment, a single VMmark tile can drive about 3500
IOPS (Input/Output Operations Per Second), with additional tiles typically each driving somewhat less. The
latency requirements will also vary based on other aspects of the environment; a review of published VMmark
results will provide a sense of the storage solutions that work with VMmark. Thus in addition to ensuring that
you have enough storage capacity, you should also make sure your storage system will have adequate
performance.
So, I think a lot of the problems you have with the VMmark3 benchmark is due to the lack of a high-performance, shared storage array. I would recommend looking at VMmark 3.x Results -- those PDFs have a section on their storage hardware, most of which consistent of all-flash storage arrays.
