Health, Utilization, Capacity, and Performance
IO’s value lies in its ability to proactively detect deviations from your normal levels of Health, Utilization, Capacity, and Performance and provide tools for investigation and remediation. IO provides visibility into metrics pertaining to infrastructure health, utilization, and performance
Health
The health of your application's infrastructure depends on its configuration, how it communicates with other devices, and the cleanliness of its physical layer. Issues and anomalies that impact any of these items can lead to availability and performance issues.
IO observes and reports on these common health issues:
Physical layer issues: CRC errors, code violation errors, frame errors, loss of sync, loss of signal
Communication issues: Class 3 discards, link resets, link failures, aborts, buffer-to-buffer credit exhaustion
Configuration issues: Zoning, incorrect HBA queue depth settings, multi-pathing failure, cluster balance, bully VMs, zombie VMs
Utilization
Utilization is measured as a rate, e.g., MB/s, or a percentage of the available resources. High and low utilization can be a problem for performance, but it depends on multiple aspects. Measurement of one portion of the environment doesn’t take into consideration of other bottlenecks that can be in the environment as well (ISL bandwidth, storage array limitations, bandwidth within the server, available CPU resources, etc.) which all can have an impact on overall performance.
Performance
The performance of your application is dependent on the performance of your underlying infrastructure. CPU contention and memory pressure are two factors that can critically impact your application performance.