apache / apache/incubator-graphar
Implement an LRU cache mechanism to the reader in the C++ library
- Dominant language
- C++
- Stars
- 373
- Forks
- 93
- Avg merge
- 7d 21h
- Merged PRs (30d)
- 7
Description
**Is your feature request related to a problem? Please describe.**
Least Recently Used (LRU) is a common caching strategy. To improve the performance of reading GAR chunks, this strategy could be applied in the implementation of the reader in the C++ library of GraphAr.
**Describe the solution you'd like**
Applying the LRU mechanism in the implementation of reader, for caching GAR chunks in the memory. And comparing the performance gap with disabling this mechanism using representative applications.
**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.
**Additional context**
Add any other context or screenshots about the feature request here.
Contributor guide
Research direction
Start by locating the C++ reader implementation and the path that reads GAR chunks. Determine how an LRU cache would fit there, then compare representative applications with caching enabled and disabled. Done means the reader uses the cache and the performance difference is documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- data
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100