Veritas InfoScale™ 7.4.1 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 InfoScale product components in the VMware guest
- Optimizing storage with InfoScale product components in the VMware guest
- Migrating data with InfoScale product components in the VMware guest
- Improving database performance with 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
How DMP monitors I/O on paths
DMP maintains a pool of kernel threads that are used to perform such tasks as error processing, path restoration, statistics collection, and SCSI request callbacks.
One kernel thread responds to I/O failures on a path by initiating a probe of the host bus adapter (HBA) that corresponds to the path. Another thread then takes the appropriate action according to the response from the HBA. The action taken can be to retry the I/O request on the path, or to fail the path and reschedule the I/O on an alternate path.
The restore kernel task is woken periodically (by default, every 5 minutes) to check the health of the paths, and to resume I/O on paths that have been restored. As some paths may suffer from intermittent failure, I/O is only resumed on a path if the path has remained healthy for a given period of time (by default, 5 minutes). DMP can be configured with different policies for checking the paths.
The statistics-gathering task records the start and end time of each I/O request, and the number of I/O failures and retries on each path. DMP can be configured to use this information to prevent the SCSI driver being flooded by I/O requests. This feature is known as I/O throttling.