PantheonSim / GPUs / NVIDIA A100 80GB PCIe

Simulated NVIDIA A100 80GB PCIe: a GPU profile for CUDA and ROCm testing

PantheonSim, the GPU simulator for CUDA and ROCm, simulates the NVIDIA A100 80GB PCIe (Ampere, 8.0) as the profile nvidia/a100-80gb-pcie. The PCIe card version of the A100, with 80 GB of HBM2e, for training and HPC in standard servers. Its profile is built from public documents and not yet measured against a physical card.

The profile at a glance

These are the numbers a program can ask this simulated GPU, and the values nvidia-smi show for it. They come from the profile file in the simulator's repository, pinned to one commit of it.

Property	Value
Profile id	`nvidia/a100-80gb-pcie`
Vendor and architecture	NVIDIA, Ampere
Compute capability	`8.0`
Device memory	80.0 GiB, HBM
Streaming multiprocessors	108
Warp size	32
Max threads per multiprocessor	2048
Shared memory per block (opt-in maximum)	163 KiB
L2 cache	40 MiB
Memory bus width	5120-bit
Power limit (profile)	300 W
Maximum clocks (profile)	1410 MHz shader, 1512 MHz memory
ECC memory	yes
PCIe link reported	Gen4 x16
Tensor or matrix cores	yes
Verified (measured on a physical card)	not yet measured

What this GPU is for

The PCIe card version of the A100, with 80 GB of HBM2e, for training and HPC in standard servers.

Ampere covers two compute capabilities in these profiles: 8.0 for the GA100 data-center parts and 8.6 for the GA10x parts. The simulator runs the SASS a binary carries for the architecture and falls back to PTX when it carries none.

Among the 37 NVIDIA profiles it ranks 14th by device memory (80.0 GiB) and 19th by SMs (108), counting the largest as first.

Is this profile verified?

No. This profile is marked not verified: it is built from public documents only, and nothing in it was measured on a physical card. Some figures are derived or copied from the nearest profile where the vendor publishes none, so treat them as approximate, and memory sizes as upper bounds. Programs still run on it, and it still answers every call as the architecture's rules require, but a value such as a clock or a limit may differ from a real card's.

Simulate this GPU

Build the simulator (see the docs) or open the playground, then:

nvcc -cudart shared -arch=compute_80 my_app.cu -o my_app
vgpu run --gpu nvidia/a100-80gb-pcie ./my_app
vgpu shell -y --gpu nvidia/a100-80gb-pcie --count 4
vgpu info --gpu nvidia/a100-80gb-pcie

This profile is compute capability 8.0. A kernel built for compute_75 runs on it and on every other NVIDIA profile; a kernel built for a newer architecture is refused by a profile that is older than it. Link the CUDA runtime shared (-cudart shared) so the simulator can stand in for it.

To use it every time, put gpu: nvidia/a100-80gb-pcie in a pantheonsim.yaml settings file, or in a GitHub Actions job give the PantheonSim action gpu: nvidia/a100-80gb-pcie (see CUDA and HIP tests in GitHub Actions). The simulator checks behaviour and has no timing model, so a run on a physical GPU before a release still counts. The CUDA guide explains the rest.

Where the numbers come from

Each profile file begins with a note on its sources. This is the note for NVIDIA A100 80GB PCIe, as written in the file:

Source note for nvidia/a100-80gb-pcie

Public values; verified: false, nothing here has been measured on a card. Where each value comes from: - compute capability 8.0 and every per-SM limit: read from a physical A100 SXM4 80GB (nvidia/a100) and 40GB (nvidia/a100-sxm4-40gb); the PCIe card is the same GA100 die with the same 108 SMs, so the limits are inherited from those profiles. - memory 80 GB HBM2e, 1935 GB/s, max TDP 300 W, PCIe Gen4: NVIDIA's A100 page (80GB PCIe column). The SXM4 card read 85095874560 bytes of an "80 GB" part; this profile keeps the label's 81920 MiB as an upper bound until a PCIe card is read. - memory clock: 1935 GB/s on the 5120-bit bus is 1512 MHz (derived); the SXM4's 2039 GB/s is 1593 MHz, which a card reports. - PCI device 0x20B5, "NVIDIA A100 80GB PCIe", from the name table in NVIDIA's open GPU kernel modules (g_nv_name_released.h, main at ca2d03e, 2026-10-07).

The GPU index lists every profile, and this one is not yet measured.

Other Ampere profiles: