Configurable Software System Tuning Benchmark
Project description
CSSTuning - Configurable Software Configuration Optimization Benchmark
CSSTuning is a comprehensive benchmark tool that caters to various real-world software configuration optimization problems. It offers a simple and rapid setup process for creating a software configuration optimization environment on Linux systems, accurately replicating real software configuration challenges.
Introduction
Optimizing software configurations is a crucial aspect of performance tuning for a wide array of applications and systems. CSSTuning aims to address this need by providing an extensive collection of benchmark scenarios, each representing authentic software configuration optimization problems. Whether you are a developer, researcher, or enthusiast, CSSTuning provides an invaluable resource for assessing and optimizing the performance of diverse software configurations.
Benchmark List
Compiler
Data base manage system (DBMS) //to do
Web Server //to do
Big data system //to do
Features
- Diverse Benchmark Scenarios: CSSTuning presents a rich set of benchmark scenarios that accurately emulate real-world software configuration optimization problems for multiple software systems.
- Multi-objective Optimization: For each software system, CSSTuning incorporates various officially recognized performance evaluation metrics, allowing the construction of multi-objective optimization problems.
- Containerized Benchmarks: All benchmarks in CSSTuning are containerized, eliminating dependency issues and ensuring easy reproducibility across different environments.
- Industrial Data-driven Tuning: CSSTuning goes beyond traditional benchmarks by providing the ability to optimize software systems using real-world industrial data, making the optimization process more relevant and practical.
installation
Prerequisites:
- POSIX System (i.e. Linux, MacOS)
- Docker installed
- Python 3.8+
git clone https://github.com/neeetman/csstuning.git
cd path/to/csstuning
#modify PAPI permissions for compiler benchmark
sudo sh -c "echo -1 > /proc/sys/kernel/perf_event_paranoid"
##installation
python setup.py install --user
Usage
1.Compiler Benchmarks
Use command line
GCC compiler
# View the available benchmarks for GCC #
csstuning compiler:gcc list benchs
# View the available flags for GCC #
csstuning compiler:gcc list flags
# Help command #
csstuning --help
# Run GCC with specific benchmark and flags #
csstuning compiler:gcc run benchs=cbench-automotive-bitcount flags="ftree-loop-vectorize,ftree-partial-pre"
LLVM compiler
# View the available benchmarks for LLVM #
csstuning compiler:llvm cc list benchs
# View the available flags for LLVM #
csstuning compiler:llvm list flags
# Help command #
csstuning --help
# Run LLVM with specific benchmark and flags #
csstuning compiler:llvm run benchs=cbench-automotive-bitcount flags="vetor-combine"
Use in python
2.DBMS Benchmarks
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file csstuning-0.0.3.tar.gz.
File metadata
- Download URL: csstuning-0.0.3.tar.gz
- Upload date:
- Size: 20.0 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.8.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
112e847198fb64d73f93f137f44b008b6314fccbd3b2c55b20c3fd38a0190c8b
|
|
| MD5 |
c2a3f8ab5050eb824dd854393397f380
|
|
| BLAKE2b-256 |
6d5a89f52db425c85d20820ede6b919bc5495b63f69254c61d9bb0d490c633f8
|