Deriving Semantic Checkers from Tests to Detect Silent Failures in Production...
summary
Claims an automated pipeline T2C that transforms existing system tests into runtime semantic checkers using static slicing, dynamic tracing, symbolization and mutation, plus four-stage validation. Study covers 210 tests across four popular distributed systems and derives tens–hundreds of checkers that detect 15 out of 20 reproduced real-world silent failures. Generated checkers run with a bounded trace buffer, a precondition index, and incur about 4% overhead on YCSB while keeping low false positives.
AI-written summary · about
chapters (9)
- 0:00Silent failures in distributed systems
- 1:01Silent failures dangers
- 3:13Limitations of semantic checkers
- 5:03T2C transforming tests to checkers
- 8:21Extracting checker functions
- 10:29Production runtime parameters
- 13:44Deploying and validating checkers
- 14:27Evaluating T2C on failures
- 17:00Conclusion and future work
related talks
38mThe Verification of a Distributed System
48mThe Verification of a Distributed System
29mScaleCheck: A Single-Machine Approach for Discovering Scalability Bugs...
25mSafe Replication through Bounded Concurrency Verification
14mUnderstanding and Detecting Fail-Slow Hardware Failure Bugs in Cloud Systems
20mSift: Using Refinement-guided Automation to Verify Complex Distributed Systems
20mAn Automated Scalable RISC-V Cache Coherency Verification Project
14mBridging the Gap between Real-World and Formal Binary Lifting through(…)