Edition of Sunday 15 November 2026

Uber's Rate Limiting System

Comments on “Uber's Rate Limiting System”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

Uber describes its Global Rate Limiter, which runs in the service mesh and replaces Redis-backed and ad hoc limiters. Proxies drop requests probabilistically using drop ratios that regional controllers recompute every second from aggregated load, at the cost of 2–3 seconds of enforcement lag. Uber says it handles about 80 million requests per second across 1,100+ services, and a companion configurator rewrites limits from traffic history. The latency and surge figures are Uber's own, from selected services, and no independent source backs them.

Read the original at uber.com. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 9 of 62 with a score of 0.65. The panel rated it highest on new information (0.92) and technical depth (0.82), lowest on audience breadth (0.54). Placed No. 9, No. 14 by score alone: the placement terms moved other stories past it.

AI judgeJudge 2claude-haiku-5-5
Detailed engineering write-up of a company's distributed rate limiter, with architecture, design tradeoffs, and measured operational results. Strong technical depth; results are self-reported and unverifiable, and it is one company's system.
AI judgeJudge 3gpt-5.6-luna
Detailed engineering account of hierarchical, probabilistic global rate limiting and automated configuration at very high scale. It includes useful architectural trade-offs and operational results, though evidence is mostly internal metrics and figures without reproducible implementation details.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.65total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.64
  • Reader interest0.77
  • Audience breadth0.54
  • Practical relevance0.73
  • Substantiated0.65
  • Technical depth0.82
  • New information0.92
  • Counts against: Promotional0.14
  • Counts against: Overclaim0.24

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.640.550.720.640.17+0.24+0.153
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.620.850.840.770.23+0.26+0.200
Audience breadthHow many people who build technology does it concern?0.350.600.680.540.33+0.13+0.071
Practical relevanceHow useful is it to someone who builds software or tech products?0.670.700.820.730.15+0.22+0.161
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.630.700.620.650.08+0.05+0.033
Technical depthHow technically substantive is it?0.890.800.780.820.11+0.05+0.041
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.920.900.940.920.04+0.05+0.046
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.090.150.180.140.09−0.19−0.027
OverclaimDo the headline or conclusions claim more than the body supports?0.290.200.220.240.09−0.11−0.026
Sum of contributions+0.651
Claim-check penalty0.000
Total, as published0.651

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #9 on Rubric, 15 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-15/9/"><img src="https://rubric.hrih.si/badge/2026-11-15/9.svg" alt="Ranked #9 on Rubric, 15 Nov 2026" width="238" height="28"></a>

Claim check

Claim check: 8 no source given. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. No source givenNumber

    Uber's GRL (Global Rate Limiter) now processes around 80 million requests per second across more than 1,100 services.

    The article gives no link or citation for it.

    The article says

    “GRL now processes around 80 million requests per second across more than 1,100 services”

    The source says

    No source to compare with.

  2. No source givenNumber

    Uber's service-oriented architecture processes hundreds of millions of RPCs per second across thousands of services.

    The article gives no link or citation for it.

    The article says

    “Uber’s service-oriented architecture processes hundreds of millions of RPCs (remote procedure calls) per second across thousands of services”

    The source says

    No source to compare with.

  3. No source givenBenchmark result

    At the P99.5 tail, requests that previously took several hundred milliseconds were reduced to tens of milliseconds after migrating to GRL, up to a 90% improvement.

    The article gives no link or citation for it.

    The article says

    “representing up to a 90% improvement in the slowest response times”

    The source says

    No source to compare with.

  4. No source givenNumber

    A critical service survived a 15× traffic surge (22K to 367K RPS) without degradation thanks to GRL.

    The article gives no link or citation for it.

    The article says

    “A critical service survived a 15× traffic surge (22 K → 367 K RPS) without degradation”

    The source says

    No source to compare with.

  5. No source givenNumber

    Because decisions depend on globally aggregated data, GRL enforcement can lag up to 2 to 3 seconds behind real-time traffic.

    The article gives no link or citation for it.

    The article says

    “enforcement can lag by up to 2 to 3 seconds behind real-time traffic conditions”

    The source says

    No source to compare with.

  6. No source givenCausal conclusion

    Moving rate limiting into the service mesh data plane and removing the Redis-backed limiter shifted requests toward lower-latency buckets for a large critical service.

    The article gives no link or citation for it.

    The article says

    “requests shifted toward lower-latency buckets immediately after removing its Redis-backed limiter and migrating to GRL”

    The source says

    No source to compare with.

  7. No source givenCausal conclusion

    A centralized Redis counter model could not scale to Uber's fleet and would require hundreds of Redis clusters even with sharding and replication.

    The article gives no link or citation for it.

    The article says

    “hundreds of Redis clusters would be required to maintain accurate global state in real time”

    The source says

    No source to compare with.

  8. No source givenFirst, only or fastest

    The team concluded a fully distributed architecture, with local proxies deciding from aggregated load, was the only way to achieve both low latency and global scalability.

    The article gives no link or citation for it.

    The article says

    “was the only way to achieve both low latency and global scalability”

    The source says

    No source to compare with.

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 3 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 15 Nov 2026, 20:46 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “Uber's Rate Limiting System”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.