Chainscope pub

Author: gallomassimo

_bibliography/references.bib

@@ -134,4 +134,14 @@ @inproceedings{azorin2024PracticalDL
   booktitle={In AAAI Workshop on Practical Deep Learning in the Wild},
   howpublished = {/docs/2023PracticalDL.pdf},
   series = {PracticalDL 24}
+}
+
+@article{hong2026Chainscope,
+  author = {Ruipeng, Hong and Gabriele, Castellano and Jingrun, Zhang and Chen, Sun and Pengfei, Chen and Massimo, Gallo},
+  title = {ChainScope: Balancing Accuracy and Overhead in Non-intrusive Distributed Tracing of Microservices},
+  year = {2026},
+  journal = {Proceedings of the ACM on Networking},
+  abstract = {Distributed tracing is essential for monitoring and debugging services deployed on top of microservice architectures. However, existing solutions face a fundamental trade-off between non-intrusiveness, language/protocol independence, overhead, and accuracy. We present ChainScope, a distributed tracing system that leverages eBPF to establish a universal tracing framework within the kernel. By implementing kernel-level context propagation, ChainScope enables trace-level sampling in a non-intrusive manner, effectively controlling overhead while maintaining high accuracy across diverse service processing models. Evaluation results show that ChainScope achieves better accuracy-overhead trade-offs than existing non-intrusive tracing approaches with up to 2.2× better accuracy, and 1.6× higher end-to-end service throughput in complex scenarios such as asynchronous processing and RPC (Remote Procedure Call) protocols. Furthermore, we demonstrate that ChainScope accurately identifies the root cause of long-tail latency in a realistic use case.},
+  howpublished = {/docs/2026Conext.pdf},
+  series = {CoNEXT 26}
 }

_posts/2026-03-27-Conext.md

@@ -0,0 +1,12 @@
+---
+title:  "Chainscope accepted at PACMNet (CoNEXT) 2026."
+search: true
+author_profile: true
+classes: wide
+---
+
+Our paper titled *"ChainScope: Balancing Accuracy and Overhead in Non-intrusive Distributed Tracing of Microservices "* will be presented at Conext 2026
+
+Congrats to the team, especially Ruipeng, and Gabriele.
+
+*Abstract:* Distributed tracing is essential for monitoring and debugging services deployed on top of microservice architectures. However, existing solutions face a fundamental trade-off between non-intrusiveness, language/protocol independence, overhead, and accuracy. We present ChainScope, a distributed tracing system that uses eBPF to establish a universal tracing system in the kernel that enables trace-level sampling in non-intrusive tracing frameworks to control overhead. ChainScope operates independently of application languages and protocols, and introduces kernel-level context propagation for high-accuracy and efficient trace reconstruction across diverse service processing models. Evaluation results show that ChainScope achieves better accuracy-overhead trade-offs than existing non-intrusive tracing approaches with up to 2.2× better accuracy, and 1.6× higher end-to-end service throughput in complex scenarios such as asynchronous processing and RPC (Remote Procedure Call) protocols. Furthermore, we demonstrate that ChainScope accurately identifies the root cause of long-tail latency in a realistic use case.