Platforms · Measurement record
Recordmi355x-2026-11-05Outputs validatedFixed
AMD Instinct MI355X
Measured on 5 November 2026 on one GPU, with the published method. 26 double-precision measurements of the tuned code for this platform: five launches each, every launch recorded.
One whole GPU, unpartitioned, with 192 host threads. Power cap 1,400 W, the maker's rated board power, ECC on, clocks not locked.
- Published method
- Run and checked by Xcelerit
- Suite version: November 2026 round
- Double precision
- Accelerator
- AMD Instinct MI355X
- Power limit
- 1,400 W
- Driver
- amdgpu 7.1.3
- Host CPU
- AMD EPYC 9005 (Turin), 192 threads
- Host memory
- 2,015 GiB
- Operating system
- Ubuntu 24.04.5 LTS
- Compiler
- ROCm 7.2.4, gfx950
- Libraries
- hipRAND/rocRAND
Results at a glance
The sizes the results rank at. Time is the mean of the middle three of five launches; throughput is the problem size over that time (Black-Scholes: options × 100 passes, the work of one call). Sizes in K and M are binary (1K = 1,024); throughput is per second, in decimal units. Tuned code: tuned for this chip under the same rules. Select a row to see the launches behind it.
| Workload | Size | Time (ms) | Throughput (per s) | Energy | Within-launch CI | Launch spread | |
|---|---|---|---|---|---|---|---|
| Swaption portfolio, LMM | 1,024K paths | 5.45 | 192,367,632 | 1.726 Wh/bn | ±1.00% | 0.54% | Inspect |
| Barrier options, Monte Carlo | 16,000 options | 5.90 | 2,710,863 | 0.09743 Wh/M | ±0.36% | 0.12% | Inspect |
| American options, binomial lattice | 512K options | 72.21 | 7,260,760 | 37.27 Wh/bn | ±0.21% | 0.32% | Inspect |
| European options, Black-Scholes formula | 128M options | 195.58 | 68,625,741,546 | 3.509 Wh/tn | ±1.00% | 0.05% | Inspect |
All 26 measurements
| Workload | Code | Size | Time (ms) | Within-launch CI | Launch spread | Average warm calls per launch |
|---|---|---|---|---|---|---|
| Swaption portfolio, LMM | Tuned code | 16K | 0.438 | ±0.09% | 2.61% | 4,567 |
| Swaption portfolio, LMM | Tuned code | 32K | 0.450 | ±0.05% | 3.08% | 4,444 |
| Swaption portfolio, LMM | Tuned code | 64K | 0.620 | ±0.08% | 0.29% | 3,228 |
| Swaption portfolio, LMM | Tuned code | 128K | 0.710 | ±0.44% | 2.28% | 5,533 |
| Swaption portfolio, LMM | Tuned code | 256K | 1.363 | ±1.00% | 4.64% | 11,908 |
| Swaption portfolio, LMM | Tuned code | 512K | 2.544 | ±1.00% | 3.88% | 7,877 |
| Swaption portfolio, LMM | Tuned code | 1,024K | 5.451 | ±1.00% | 0.54% | 2,797 |
| Swaption portfolio, LMM | Tuned code | 2,048K | 10.529 | ±0.74% | 0.38% | 776 |
| Barrier options, Monte Carlo | Tuned code | 2,000 | 1.618 | ±0.19% | 0.04% | 1,237 |
| Barrier options, Monte Carlo | Tuned code | 4,000 | 2.149 | ±0.13% | 0.21% | 931 |
| Barrier options, Monte Carlo | Tuned code | 8,000 | 3.373 | ±0.22% | 0.36% | 593 |
| Barrier options, Monte Carlo | Tuned code | 16,000 | 5.902 | ±0.36% | 0.12% | 339 |
| Barrier options, Monte Carlo | Tuned code | 32,000 | 10.833 | ±0.69% | 0.26% | 185 |
| American options, binomial lattice | Tuned code | 16K | 9.415 | ±0.45% | 3.60% | 213 |
| American options, binomial lattice | Tuned code | 32K | 11.572 | ±0.27% | 0.50% | 173 |
| American options, binomial lattice | Tuned code | 64K | 15.839 | ±0.23% | 0.15% | 127 |
| American options, binomial lattice | Tuned code | 128K | 24.357 | ±0.31% | 0.49% | 82 |
| American options, binomial lattice | Tuned code | 256K | 40.440 | ±0.27% | 0.64% | 50 |
| American options, binomial lattice | Tuned code | 512K | 72.208 | ±0.21% | 0.32% | 28 |
| American options, binomial lattice | Tuned code | 1,024K | 134.339 | ±0.15% | 0.37% | 15 |
| American options, binomial lattice | Tuned code | 2,048K | 258.891 | ±0.23% | 0.19% | 10 |
| European options, Black-Scholes formula | Tuned code, pinned host memory | 16M | 28.309 | ±1.00% | 0.02% | 273 |
| European options, Black-Scholes formula | Tuned code, pinned host memory | 32M | 52.876 | ±1.00% | 0.05% | 211 |
| European options, Black-Scholes formula | Tuned code, pinned host memory | 64M | 100.583 | ±1.00% | 0.04% | 160 |
| European options, Black-Scholes formula | Tuned code, pinned host memory | 128M | 195.579 | ±1.00% | 0.05% | 151 |
| European options, Black-Scholes formula | Tuned code, pinned host memory | 256M | 386.446 | ±0.69% | 0.47% | 104 |
Outputs validated before speed
A time only counts if the output is right. At every problem size, the option price the tuned code computes is validated against the reference implementation's on the same machine.
| Workload | Against the reference code | Result |
|---|---|---|
| Swaption portfolio, LMM | Within 0.91% at 128K paths, on this platform's own random-number generator | Pass |
| Barrier options, Monte Carlo | Within 0.79% at 32,000 options, on this platform's own random-number generator | Pass |
| American options, binomial lattice | Identical to double precision | Pass |
| European options, Black-Scholes formula | Identical to double precision | Pass |
Binomial and Black-Scholes are deterministic: the tuned code gives the reference code's prices. Monte Carlo: on GPUs the barrier and LMM codes generate their random numbers differently, so their prices differ only by sampling: within 4 standard errors of the reference implementation's, and fed the same random numbers they agree with the reference implementation's to within 1e-9 (relative).
Every launch behind one measurement
Pick a workload and a size. Each dot is one launch: the mean of its warm calls, run until the 95% confidence interval was within ±1% or until the call limit. The fastest and slowest launch are dropped and the middle three averaged into the result.
Zoom: drag a box on the chart or use the slider under it; reset brings it back.
| Launch | Warm calls | Warm mean (ms) | Range (ms) | Used |
|---|
How time grows with the problem
Time against size on log scales, for every size measured. The dashed line is perfect linear scaling from the smallest size: points under it mean the machine was not yet full at small sizes.
What the record holds
The measurement record behind this page: every launch, the machine and the output validation, and the results as a table.
Full record (JSON) Results (CSV)
Free to cite with attribution: Xcelerit Finance Benchmarks, record mi355x-2026-11-05.
SHA-2562e9c72dfb75b3b65c83fd0a20bf16608b28519c9f5ef926eca6999fc1a6def18
Records are fixed: once published, a record is never edited. A correction is published as a new record, linked from the old one. Checksums of every record