EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Lobsters·4h ago·5 min read
512 bytes at a time: reconstructing IBM i save-file checksums
TL;DR
A Python implementation was developed to reconstruct and repair IBM i save-file (SAVF) checksums, enabling modifications to SAVFs while maintaining data integrity. The process successfully validated and repaired both per-record and global checksums after byte modifications.
✦ Why It Matters
Implement this checksum reconstruction method to safely modify IBM i save files without corrupting data integrity.
Key Takeaways
- Understand the structure of IBM i save files for effective modifications.
- Use Python to replicate checksum algorithms for SAVFs.
- Ensure both per-record and global checksums are updated after modifications.
- Validate modified SAVFs before importing to maintain data integrity.
How It Works
The reconstruction process involves two main checksum algorithms: one for individual records and another for the entire logical stream. The per-record checksum is calculated using a combination of the record sequence and payload, processed through a series of reductions and transformations.
The global checksum is computed over the complete logical stream, utilizing a lookup table and specific state management to ensure accurate results. This allows for precise updates to the checksums after any byte modifications.