The PALISADE-CPU backend is a Reference backend engine for HEBench which is a shared library that implements the required functions specified in either the hebench::APIBridge or hebench::cpp wrapper. It is intended only as a reference on how to implement a PALISADE backend for HEBench and it is not fully optimized for performance.
Apache License 2.0
6
stars
2
forks
source link
Many Threads Spinning Not doing work on Latency (batch-sized) tests #26
On latency tests that are batch-sized based, threads are allocated but aren't actually doing any work.
As such, performance takes a significant hit.
To Reproduce
Steps to reproduce the behavior:
Run any Latency batch-size based test, and see that the CPU is significantly > Wall time when it should be around the same as 1 thread is being used.
Expected behavior
It's expected that CPU and Wall time on those tests will be relatively close to each other, and performance will be better since the workload isn't having to deal with unused threads.
Describe the bug
To Reproduce
Steps to reproduce the behavior: Run any Latency batch-size based test, and see that the CPU is significantly > Wall time when it should be around the same as 1 thread is being used.
Expected behavior
It's expected that CPU and Wall time on those tests will be relatively close to each other, and performance will be better since the workload isn't having to deal with unused threads.
Screenshots
N/A
Additional context
N/A