Faye Amacker

I removed "Less network bandwidth needed for chunk data packs (TBD)" because I didn't add it to Success Criteria list. Seems risky to add to Success Criteria after work was...

I added @janezpodhostnik to the DACI Consulted (aka Contributors) list. He provided answers and valuable insights when asked about fees, etc. and also contributed in other ways. 👍

UPDATE: With Jan 31, 2025 mainnet state, this optimization reduces state size by 11.1%, eliminates 28.2% of mtrie vertexes (leaf and non-leaf nodes), and also reduces payload count by 28.2%....

Closing this as completed. Zero-downtime (aka on-the-fly) migrations completed on mainnet. All accounts are using storage account format v2. Storage health check, etc. didn't report any new problems on mainnet...

Awesome work everyone! This was initially an idea for a regular migration and then it became more ambitious (HCU, feature flag, on-the-fly migration) yet the team managed to make everything...

I'm told this reduced peak RAM use on Execution nodes by ~30% (roughly 3x the state size reduction). 🎉

@j1010001 @janezpodhostnik @peterargue @zhangchiqing Good meeting today (sorry it was ~30 minutes longer than expected). NEXT STEPS: I will use Approach G to create a PoC. Approach G combines some...

Quick update on this, since today's backlog meeting was canceled and we had no standup yesterday. I resumed work on this yesterday (July 9). Also very briefly started to resume...

Update: I created a new LRU using a different approach than discussed because the approach discussed on May 21 would require more locks and it would add overhead to `get`,...

@SophisticaSean As a quick sanity check, I tried building and running a simple program that has dependencies specifying different versions of fxamacker/cbor: - program: cbor v2.7.1-0.2025... (updated feature/stream-mode branch) -...