EPF Dev Update - Week 9 & 10
# EPF Dev Update - Week 9 & 10
1. **Log Instructions:**
- Made progress on the PR. Due to time constraints, I couldn't fully optimize the implementation, so one of my mentors assisted with the PR.
- I was informed that, as the project was getting delayed, I should start working on opcode statistics in week nine.
- The PR received two approvals, but I still need to address some of the comments.
2. **Opcode Statistics:**
- Read numerous papers related to the task.
- At the start of week 10, I modified the client to analyze the code database and initially provided just two opcode stats.
3. **Basic Tool for Implementation Testing:**
- Initially developed a very basic tool for implementation testing.
- This tool collects recent execution n-grams and performs statistics on them using a shell script.
- A sample of 100 blocks was shared with my mentors.
4. **Opcode Stats Single Pass Tool:**
- Began developing a tool that identifies heavy hitters using probabilistic data structures, specifically Count Min Sketch (CMS).
- CMS is known for its good performance in big data scenarios and enables one-pass implementations of count queries.
- Implemented CMS and researched how to optimize the available parameters, including the number of buckets and the number of hash functions.