The question
How can expired state be identified and reclaimed with bounded memory overhead and minimal latency spikes?
This is a proposed research direction. No novel algorithm, completed benchmark or performance gain is claimed.
Why it matters
Session stores, caches and streaming operators need cleanup that keeps pace with state growth.
Search space
Time buckets, timing wheels, expiry indexes and adaptive cleanup budgets.
What would count as progress?
- Correct expiry visibility
- memory per entry
- CPU
- p99 operation latency
- reclamation delay.
A useful result improves a measured trade-off against strong baselines on held-out workloads, with reproducible code and documented limitations.
Baselines and boundaries
Compare periodic scans, expiry heaps and bucketed cleanup. Distinguish logically expired entries from physically reclaimed storage; test idle periods, expiry bursts and TTL refreshes.
Starting reference
Where you could contribute
Help define representative, legally shareable workloads; critique the correctness contract; implement baselines; propose candidates; or reproduce measurements. Negative results and failure cases are useful contributions.
A detailed research brief is being prepared. Access requests will become available when that brief and the request service are ready.