Storage performance: IOPS, latency and throughput
Relate IOPS, block size, latency and queue depth, account for RAID write penalties and rebuild times, and find the real bottleneck in storage complaints.
- Level
- Advanced
- Length
- About 55 minutes
- Contents
- 3 lessons · 1 video · final exam
- Status
- Published · updated 1 Oct 2026
Skills you'll practise
- Convert between IOPS, block size and throughput
- Apply Little's law (queue depth ÷ latency) to estimate IOPS
- Estimate front-end IOPS with RAID write penalties and usable capacity by RAID level
- Estimate rebuild time and explain rebuild risk
- Read latency graphs and datastore statistics to find saturation
Course outline
- 1.Storage performance: IOPS, latency and throughputVideo · 2 min
- 2.The three numbers and how they relateLesson · 25 min
- 3.Rebuilds, capacity and troubleshootingLesson · 20 min
- 4.Benchmarking storage honestlyLesson · 20 min
- 5.Storage performance: IOPS, latency and throughput: knowledge checkKnowledge check · 26 questions
- 6.Degraded array with a second disk warningScenario
- 7.Final exam10 questions · passing it completes the course, so people who already know the material can test out
Sources it draws on
The lessons and questions are written from these references, so learners can go back to the original.
- SNIA Dictionary: IOPS, latency, throughput
- Little's law (queueing theory): L = λW
- Storage vendor sizing guides: RAID write penalty (RAID 10 = 2, RAID 5 = 4, RAID 6 = 6)
See it with your own jobs and topics
Tell us about your team and we'll walk you through setup, from choosing jobs to your first skills check.