DRAM & Memory Design · All levels

DRAM Timing & Command Protocols: Tricky Q&A

Senior interview and review questions for DRAM Timing & Command Protocols.

Section Q&A bank

Use these drills after completing all topics in DRAM Timing & Command Protocols. Answer with workload context, mechanism proof, artifact, owner, and release decision.

Why does a controller track both tRRD and tFAW instead of only one activate limit?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: Why does a controller track both tRRD and tFAW instead of only one activate limit?

A:
tRRD constrains spacing between consecutive ACT commands, while tFAW limits how many ACTs fit in a rolling window; both are required to prevent over-aggressive row activation bursts.

FOLLOW-UP TRAP: Treating tFAW as optional if tRRD already passes.

What does tRCD practically protect in the ACT -> READ/WRITE sequence?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: What does tRCD practically protect in the ACT -> READ/WRITE sequence?

A:
It guarantees enough delay after ACT before column commands so sense-amplifier and row activation are stable for data access.

FOLLOW-UP TRAP: Issuing READ/WRITE immediately after ACT because row is 'logically open.'

How are tRAS, tRP, and tRC related during same-bank reuse?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: How are tRAS, tRP, and tRC related during same-bank reuse?

A:
A row must stay active at least tRAS, then PRE must complete for tRP, and the next ACT to that bank cannot violate the full row-cycle interval tRC.

FOLLOW-UP TRAP: Optimizing PRE timing while forgetting same-bank ACT spacing.

Why can read/write turnaround dominate bandwidth loss even when row-hit rate is high?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: Why can read/write turnaround dominate bandwidth loss even when row-hit rate is high?

A:
Direction changes on the DQ bus introduce mandatory dead cycles and recovery delays, so frequent R<->W switching creates bubbles despite good row locality.

FOLLOW-UP TRAP: Assuming row-hit optimization alone removes throughput penalties.

What should be included in a robust DRAM timing guardband strategy?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: What should be included in a robust DRAM timing guardband strategy?

A:
Programmable margin on key timings (tRCD, tRP, tRAS, tRC, tRRD, tFAW and turnaround knobs) tied to PVT, SI uncertainty, and PHY training spread.

FOLLOW-UP TRAP: Applying one fixed margin without speed-bin or condition sensitivity.

Which artifact best proves scheduler-visible timing correctness?

diagram
[INT][DRAM][TIMING-PROTOCOLS]

Q: Which artifact best proves scheduler-visible timing correctness?

A:
A command-level legality checker with assertion coverage showing no timing-window violations under stress traffic and mixed bank-group scenarios.

FOLLOW-UP TRAP: Relying only on average bandwidth metrics.

Q&A drill guide

diagram
WORKLOAD -> DRAM SYMPTOM -> TIMING/QUEUE METRIC -> ROOT CAUSE -> FIX -> REGRESSION

Sketch while answering

diagram
COMMAND TIMING SEQUENCE

ACT -> tRCD -> READ/WRITE -> tRAS(min) -> PRE -> tRP -> next ACT

Key takeaways

  • Always tie controller and PHY counter shifts to application latency and throughput outcomes.

  • Lock firmware timing profile, thermal condition, and DIMM state before comparing DRAM captures.

Common pitfalls

  • Chasing peak bandwidth while ignoring p99 latency and fairness tails.

  • Changing timing guardbands without separating SI noise from scheduling issues.

  • Declaring closure without reliability gates, fault injection, and regression replay.