NUMA and Memory Affinity: Worked Example

Worked example

Worked Example for NUMA and Memory Affinity centers on remote memory access ratio, NUMA latency delta, and workload affinity score. Tie every claim to a measurable artifact and an owner-controlled action.

A regression flags remote memory access ratio, NUMA latency delta, and workload affinity score. Correct triage isolates first failing stage, confirms mechanism, then applies one reversible change and validates blast radius.

System view

CPU PIPELINE VIEW - NUMA and Memory Affinity

fetch -> decode -> rename -> dispatch -> execute -> retire
  |        |         |          |         |         |
icache   uop flow   map table  queueing  FU ports  ROB commit

steady-state goal:
keep every stage supplied without bubbles or flush storms

Focus: front-end to retire flow
Metric tracked: remote memory access ratio, NUMA latency delta, and workload affinity score

Local vs remote memory path lengths

CPU CACHE + MEMORY HIERARCHY - NUMA and Memory Affinity

                 [ L1I ]   [ L1D ]
               32-64KB, ~4 cycles
                      \     /
                       [  L2  ]
                 512KB-2MB, ~12 cycles
                           |
                         [ L3 ]
               shared LLC, 30-60 cycles
                           |
                    [ DDR/HBM memory ]
                    80-150ns effective

Optimization lens: contrast same-node access against remote-node latency penalties

1. Capture baseline and failing trace under fixed environment tags.

2. Classify stage loss and identify dominant mechanism.

3. Collect NUMA locality profile, page-placement log, and latency percentile report.

4. Apply one bounded fix with ownership signoff.

5. Re-run validation matrix and decide ship/rollback.

CPU deep dive

Scaling across cores is limited by coherence and interconnect behavior before compute saturation on many workloads.

Concept diagram

MULTICORE SYSTEM VIEW

cores + private caches <-> LLC slices <-> interconnect <-> memory controllers

Metric graph

SCALING EFFICIENCY

ideal scaling        ███████████
observed under load  ███████
after policy tuning  █████████

Reports and artifacts

• coherence traffic matrix

• NoC congestion map

• NUMA locality profile

• synchronization contention report

Mini case study

A lock-heavy service regressed at higher core counts because coherence invalidations and NoC hotspots dominated.

Debug branches

• Classify traffic as coherence, demand miss, or synchronization

• Measure hotspot links instead of aggregate NoC throughput

• Validate thread and page affinity before hardware changes

Senior review question

Ask: which CPI/latency evidence proves this topic is truly closed beyond synthetic benchmarks?

Key takeaways

• Always connect microarchitectural counter changes to product workload outcomes.

• Lock binary, compiler, firmware, and thermal metadata before comparing CPU traces.

Common pitfalls

• Treating average IPC as sufficient proof while ignoring latency tails and outliers.

• Applying predictor or prefetch tweaks without first-failing-stage attribution.

• Declaring closure without reproducible perf, correctness, and power gates.

Worked-example reasoning

Suppose remote memory access ratio, NUMA latency delta, and workload affinity score regresses on a production workload. A shallow response tweaks one predictor knob or compiler flag. A deeper response compares baseline and regressed evidence, then identifies the first repeated loss mechanism in Thread and page placement policy controls whether cores access local or remote memory; poor affinity silently adds latency and coherence overhead to otherwise efficient software..

If bad-speculation counters dominate, inspect target/direction quality and recovery bandwidth. If queue pressure dominates, inspect scheduling and port contention. If memory dominates, inspect cache/TLB/coherence plus locality policy.

Only then choose a bounded fix: software layout, predictor policy, queue tuning, cache/prefetch change, microarchitectural update, or physical closure adjustment.