m3taversal
be8ff41bfe
Wrote sourced_from: into 414 claim files pointing back to their origin source. Backfilled claims_extracted: into 252 source files that were processed but missing this field. Matching uses author+title overlap against claim source: field, validated against 296 known-good pairs from existing claims_extracted. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2.7 KiB
| type | domain | description | confidence | source | created | supports | reweave_edges | sourced_from | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| claim | internet-finance | Simulation-based scheduling optimizes the responsiveness-efficiency tradeoff in systems with time-varying arrival rates | proven | Simio / WinterSim 2018, Resource Scheduling in Non-Stationary Service Systems | 2026-03-11 |
|
|
|
Non-stationary service systems require dynamic worker allocation because fixed staffing wastes capacity during low demand and creates bottlenecks during peaks
Service systems with time-varying arrival rates face a fundamental tradeoff: fixed worker counts either waste capacity during quiet periods or create unacceptable wait times during demand spikes. The WinterSim 2018 paper demonstrates that simulation-based approaches can optimize this tradeoff by modeling realistic arrival patterns and testing staffing policies before deployment.
The key insight is that without server constraints there would be no waiting time, but this wastes capacity since arrivals are both stochastic (random within any time window) and nonstationary (the average rate changes over time). Traditional queueing theory assumes stationary arrivals, making it unsuitable for real-world systems where demand varies by hour, day, or season.
The paper validates discrete-event simulation as the method for determining optimal server counts as a function of time, measuring queue depth and adjusting workers dynamically rather than using static scheduling.
Evidence
- WinterSim 2018 paper explicitly addresses "the gap between theoretical queueing models (which assume stationarity) and real systems (which don't)"
- Paper states: "Without server constraints there would be no waiting time, but this wastes capacity since arrivals are stochastic and nonstationary"
- Simulation-based approach tests staffing policies against realistic arrival patterns to optimize responsiveness vs efficiency
Relevance to Teleo Pipeline
This directly validates the Living Capital pipeline architecture choice to use dynamic worker scaling based on queue depth rather than fixed MAX_WORKERS or cron-based scheduling. The paper's framework maps precisely to the agent task processing problem: LLM API calls are the "servers", task arrivals are nonstationary (bursty during market hours, quiet overnight), and the goal is minimizing latency without wasting compute capacity.
Relevant Notes:
- domains/internet-finance/_map
Topics:
- domains/internet-finance/_map