How to address bottlenecks in VMware infrastructure storage

The first consideration to this question should really be "is the storage the actual bottleneck?" Many VMware performance issues are mistakenly blamed on storage, when in reality the true problem lies elsewhere. For the sake of this question, however, we will assume that the bottleneck is the storage. In such a case, there are three main possible causes of the problem:

1. Storage is not fit for purpose/poorly performing

2. Misconfiguration of the storage

3. Overloading the storage with too many devices

Each of the issues above can be attributed to either unrealistic expectations of what can be achieved, or a complete misunderstanding of how the VMware environment uses storage. VMware Profile Driven Storage can help alleviate these bottlenecks by matching virtual machine (VM) performance requirements with the appropriate datastore.

Organisations should think of the I/O (input/output) profiles of their VM workloads and what SLAs are required for them, for example, which storage type VMs requiring high I/O should reside on and how to differentiate between different workloads and applications.

By configuring storage profiles, administrators know exactly which profile to select when creating or migrating a virtual machine. They can therefore check for profile compliance and realign any workloads that do not match the required policy.

By constructing datastores on different types of underlying storage, a virtual machine's VMDK (disk file) can reside on the most appropriate storage. Furthermore, different VMDKs for a virtual machine can reside on different datastores, in order to provide the best level of performance and scalability for individual volumes.

An example of storage profiles is shown in the table below, where workloads can be characterised by their demand on I/O.

Surjit Randhawa is a senior consultant at GlassHouse Technologies (opens in new tab).