博士論文"Scaling Telemetry Workloads in Cloud Applications: Techniques for Instrumentation, Storage, and Mining"で論じられるテレメトリーシステムにおけるワークロード増大に対するスケーリング問題のこと。 ## Related Papers ### Layer 1: Instrumentation #### Metrics - PMF: [[2024__CLOUD__Enabling Programmable Metric Flows]] #### Traces - [[2025__ICPE__Towards eBPF Overhead Quantification - An Exemplary Comparison of eBPF and SystemTap]] - [[Transtracer]] - [[2025__ICPE-Companion__Investigating Performance Overhead of Distributed Tracing in Microservices and Serverless Systems]] ### Layer 2: Collection ## Layer 3: Storage #### Metrics - [[HeteroTSDB]] #### Logs - ClickHouse: [[2024__VLDB__ClickHouse - Lightning Fast Analytics for Everyone]] ### Layer 4: Mining #### Metrics - [[MetricSifter]] - KIMetrics: [[2025__arXiv__Metric Criticality Identification for Cloud Microservices]] - [[2025__arXiv__MicroRCA-Agent - Microservice Root Cause Analysis Method Based on Large Language Model Agents]] - > 変動幅が5%未満の安定メトリックを自動フィルタリング(コンテキスト長を約50%削減) #### Logs #### Traces --- [[研究所テックトーク2025春 アウトライン]]