Run of 17 Nov 2026

20261117T2323Z-d82776-legacy

Reference Preliminary · v0

Not the default run. These numbers are from the run of 17 Nov 2026 on Intel Xeon W-2133 · 6 cores / 12 threads · bare metal. The rest of the site shows the default run of 18 Nov 2026 on the same machine. Open this page for the default run

Preliminary numbers. Measured with the preliminary methodology (v0) in only 3 passes, so there are no confidence intervals. Not measured in this run: CPU per request, CPU efficiency index. Methodology versions

What this run measured

Pre-release shell suite: throughput, p50/p99 latency, errors, start-up and memory only; floating image tags; peak memory recorded once per run; 3 s warm-up and 12 s measurement per scenario.

Run note: Pre-release shell suite: wrk 3 threads, 128 connections, 3 s warm-up + 12 s measured per scenario, 5,000 req/s fixed-rate run with oha, floating image tags.

Run

Started
17 Nov 2026, 23:23 UTC
Finished
18 Nov 2026, 01:39 UTC · 2 h 16 min
Status
complete
Passes
3
Provenance
Reference
Methodology
Preliminary methodology (v0)
Engine
import_legacy (1.0.0)

Machine

Name
Intel Xeon W-2133 · 6 cores / 12 threads · bare metal
CPU
Intel(R) Xeon(R) W-2133 CPU @ 3.60GHz
Memory
62.5 GiB
OS
Ubuntu 24.04.5 LTS
Fingerprint
d827760100bb

Options

passes
3
quick
no
implementations
all
scenarios
all

Pre-release shell suite: wrk 3 threads, 128 connections, 3 s warm-up + 12 s measured per scenario, 5,000 req/s fixed-rate run with oha, floating image tags.

CPU layout

Which logical CPUs each role was pinned to during this run.

  1. CPU 0: Service under test
  2. CPU 1: Service under test
  3. CPU 2: Environment (database)
  4. CPU 3: Environment (database)
  5. CPU 4: Load generator
  6. CPU 5: Load generator
  7. CPU 6: Kept idle
  8. CPU 7: Kept idle
  9. CPU 8: Environment (database)
  10. CPU 9: Environment (database)
  11. CPU 10: Load generator
  12. CPU 11: Load generator
  • Service under test 0, 1
  • Environment (database) 2, 3, 8, 9
  • Load generator 4, 5, 10, 11
  • Kept idle 6, 7

SMT is on. The service gets one logical CPU per physical core; the SMT siblings of those cores stay idle.

Implementations

15 implementations with their build and check status as recorded in this run.

Implementation Functional check Image size Flags Note
TypeScript · Bun passed 283.8 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
C# · ASP.NET Core 10 LTS · Native AOT passed 198.8 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
C# · ASP.NET Core 10 LTS · JIT passed 342.2 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
C# · ASP.NET Core 11 · Native AOT passed 244.8 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
C# · ASP.NET Core 11 · JIT passed 392.4 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Elixir · Plug + Bandit passed 249.4 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Go · net/http passed 21.8 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Java · Spring Boot · JVM + AOT cache passed 632.0 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Java · Spring Boot · JVM passed 560.7 MB 1 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Java · Spring Boot · Native image passed 248.1 MB 1 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
TypeScript · Express passed 375.3 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
TypeScript · Fastify passed 393.3 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
PHP · FrankenPHP passed 875.2 MB 1 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Python · FastAPI passed 258.2 MB 0 Built from floating tags at run time; versions match the pinned ones closely but not exactly.
Rust · axum passed 46.5 MB 1 Built from floating tags at run time; versions match the pinned ones closely but not exactly.

Quality flags

Warnings raised while measuring. Flagged results stay in the data and are marked wherever they appear.

Java · Spring Boot · JVM

  • Cold load: throughput varies 11 % between passes

Java · Spring Boot · Native image

  • Single query, 512 connections: throughput varies 11 % between passes

PHP · FrankenPHP

  • List: throughput varies 27 % between passes

Rust · axum

  • JSON POST: throughput varies 17 % between passes

Results

Explore every scenario and metric of this run.

2 runs published

Showing Throughput index in Overall, index, higher is better

Derived from several scenarios

Overall

Summary metrics computed from the scenario medians of this run.

Throughput index

higher is better · Geometric mean of throughput over the six workload scenarios, each relative to the fastest implementation in that scenario of the same run. 100 means fastest in all six; the leader of a run scores below 100 when it does not lead every workload.

Computed: Geometric mean of “Throughput” over JSON, Single query, 20 queries, Updates, List, JSON POST, each relative to the best implementation in that scenario (best = 100). Only an implementation that is best in every one of them reaches 100, and only implementations with a usable result in every one of them get an index.

Derived values are computed from scenario medians and carry no interval. Ranks count all 15 implementations.

Not measured in this run: Soak (5 minutes).

Implementations in this run

15 implementations · 11 scenarios measured. Every page lists all scenarios and metrics of one implementation.