White Papers

VMware storage features
66 Dell EMC SC Series: Best Practices with VMware vSphere | 2060-M-BP-V
16.2 Storage distributed resource scheduler (SDRS)
SDRS is a feature introduced in ESXi 5.0 that automatically load balances virtual machines within a datastore
cluster based on capacity or performance. When creating SDRS datastore clusters with SC Series storage,
remember a few guidelines:
• Group datastores with similar disk characteristics, such as replicated, non-replicated, storage profile,
application, or performance.
• Use SDRS for initial placement based on capacity. When placing virtual machines on the datastore
cluster, it places them based on which datastore has the most space available.
• Set the automation level to manual mode. SDRS will make recommendations about moving virtual
machines, but not move them automatically. As a best practice, run SDRS in manual mode to
examine all recommendations before applying them. There are a few items to keep in mind before
applying the recommendations:
- Virtual machines moved between datastores automatically will have their data progression history
reset. Depending on the storage profile settings, the entire virtual machine could be moved back
to tier 1 RAID 10 if the volume was set to use the recommended storage profile.
- When a virtual machine moves between datastores, its location at the time the snapshot was
taken may make the virtual machine harder to find during a recovery. For example, if a VM moves
twice a week while daily snapshots are being taken, the volume with the good snapshot of the
virtual machine may be difficult to locate.
- If the SCOS version in use does not support VAAI, the move process could be slow (without full
copy) or could leave storage allocated on its originating datastore (without dead space
reclamation). See the section 16.3 on VAAI for more information.
• Storage DRS could produce incorrect recommendations for virtual machine placement when I/O
metric inclusion is enabled on an SC Series system using Data Progression. When a datastore is
inactive, the SIOC injector performs random read tests to determine latency statistics of the
datastore. With Data Progression enabled, the blocks that SIOC reads to determine datastore
performance, could potentially reside on SSD, 15 K, or even 7 K drives. This random sampling could
ultimately skew the latency results and decrease the effectiveness of the SRDS recommendations.
Disabling I/O metric inclusion
To reiterate, use SDRS only for capacity recommendations, set it to manual automation mode, and disable
the I/O metric inclusion. These recommendations will allow administrators to take advantage of SDRS
capacity placement recommendations, while still allowing Data Progression to manage the performance of the
data at the block level.