400Ping opened a new issue, #691:
URL: https://github.com/apache/mahout/issues/691
### Summary
We now have a `dataloader_throughput` example in `qdp-core` that measures
end-to-end encode throughput for a QML-style DataLoader pipeline on QDP.
As a follow-up, we should add a small benchmark to **compare this DataLoader
throughput against equivalent workloads in other frameworks** (e.g. Qiskit,
PennyLane, CUDA-Q) on the same GPU and configuration.
### Goals
- Define a simple, canonical workload (batch size, vector length, #qubits,
#batches).
- Run the DataLoader/encode loop in:
- QDP (`dataloader_throughput`),
- 1–2 competitor frameworks (Qiskit / PennyLane / CUDA-Q to start).
- Report throughput in vectors/sec for each framework under comparable
settings.
### Deliverables
- A benchmark script or example that runs the QDP DataLoader test in
“benchmark mode”.
- Minimal competitor scripts that implement the same workload and log
throughput.
- A short markdown summary (table of results + a brief discussion) under
`docs/`
or `docs/benchmarks/`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
