-
### Motivation
Single kernel cost measurement is crucial to the holistic performance of the LLM. I wonder if lmdeploy supports benchmarking either solely quantized, int8, or int16 gemm operations?
…
-
following https://github.com/cairo-book/cairo-book/pull/996
add a new chapter 10.3 specifically for benchmarking. it's a bit too much for hte "write tests" section
-
Hi there!
I'd like to understand what's the recommended approach to Profile a model. For instance, I'd like to understand which ops take more time when running (even for a Forward Pass)
This is …
-
**Describe the bug**
When running trino ingestion with specific table name mentioned under filter --> tables --> table name(fully qualified), pipeline is not considering the filter specified.
**To…
-
I'd like to do a one-off performance profiling test (or maybe a pytest) with a high number of entities to test memory/CPU usage and make sure there's nothing we're doing anywhere that is particularly …
-
-
We have CPU profiling available right now and some open areas of improvement like
* https://github.com/getsentry/sentry-dotnet/issues/2316
* https://github.com/getsentry/sentry-dotnet/issues/2315
…
-
**Use case**
Efficiently querying continuous profiling data in ClickHouse to drive flamegraph visualization would be a great addition to the [logging, metrics and traces use case](https://clickhous…
yawnt updated
2 months ago
-
This issue is meant to serve as a placeholder with regards to optimization and profiling of the code and how this may be impacted by our design choices (e.g., use of deepcopy of SamplerState in places…
-
profiling.h has method with name "emit" and strongly conficts with Qt keyword emit() used for signal/slot mechanism; Please, rename this method (problem starts from one api and absents in parallel stu…