Veritas InfoScale™ Virtualization Guide - Linux on ESXi
- Section I. Overview
- About Veritas InfoScale solutions in a VMware environment
- Section II. Deploying Veritas InfoScale products in a VMware environment
- Getting started
- Understanding Storage Configuration
- Section III. Use cases for Veritas InfoScale product components in a VMware environment
- Application availability using Cluster Server
- Multi-tier business service support
- Improving storage visibility, availability, and I/O performance using Dynamic Multi-Pathing
- How DMP works
- Improving data protection, storage optimization, data migration, and database performance
- Protecting data with Veritas InfoScale product components in the VMware guest
- Optimizing storage with Veritas InfoScale product components in the VMware guest
- Migrating data with Veritas InfoScale product components in the VMware guest
- Improving database performance with Veritas InfoScale product components in the VMware guest
- Setting up virtual machines for fast failover using Storage Foundation Cluster File System High Availability on VMware disks
- About setting up Storage Foundation Cluster File High System High Availability on VMware ESXi
- Configuring coordination point (CP) servers
- Section IV. Reference
Improving I/O performance with Dynamic Multi-Pathing in the hypervisor
When you install DMP in the hypervisor, DMP provides visibility into the I/O paths and performance. You can use the datacenter view or the individual statistics to determine if certain LUNs are not performing as well as other LUNs. If you determine a bottleneck, you can move some of the I/O workload (and thus Virtual Machines) to another datastore to alleviate the I/O performance contention.
The following scenario describes a typical use case for improving I/O performance.
Use the datacenter view of DMP to determine which storage arrays are attached to which ESXi servers. In this scenario, one of the storage arrays is not performing well.
You can select the storage array that is not performing well to view additional I/O performance details at the VMDK level. You can use this information to understand I/O loads within individual guests and the actual subcomponents of the disks assigned to the guests.
Figure: Example I/O performance statistics shows the statistics for several LUNs. In the "Avg Time Write" column for the clariion0_5 device, its performance is significantly slower than the clariion0_1, despite similar workloads reflected by the "virtual machine" column.
For this example, DMP is not running in the guests. If you also have DMP running in the individual guests, you can correlate the I/O statistics to the particular guests.
If you determine a high I/O workload for a particular LUN, you may need to move some of the I/O workload (and thus Virtual Machines) to another datastore to alleviate the I/O performance contention.