Jepsen: Bufstream 0.1(jepsen.io)
203 points by aphyr | 78 comments
First screen, captured .
Looks similar
Jepsen: MySQL 8.0.34(jepsen.io)
321 points by aphyr | 119 comments
Jepsen analysis by Peter Alvaro and Kyle Kingsbury showing MySQL 8.0.34 Repeatable Read violates internal consistency and AWS RDS MySQL clusters violate Serializability.
Jepsen: NATS 2.12.1(jepsen.io)
412 points by aphyr | 151 comments
Jepsen analysis by Kyle Kingsbury of NATS JetStream 2.12.1, finding lost writes from file corruption, power failures, and lazy fsync behavior.
Jepsen: TigerBeetle 0.16.11(jepsen.io)
193 points by aphyr | 55 comments
Jepsen: Jetcd 0.8.2(jepsen.io)
110 points by aphyr | 12 comments
Jepsen: MariaDB Galera Cluster 12.1.2(jepsen.io)
78 points by aphyr | 8 comments
Related stories
Jepsen: Datomic Pro 1.0.7075(jepsen.io)
371 points by aphyr | 84 comments
Jepsen analysis by Kyle Kingsbury of Datomic Pro 1.0.7075, evaluating its transaction safety and serializability properties.
Jepsen: Amazon RDS for PostgreSQL 17.4(jepsen.io)
581 points by aphyr | 138 comments
Jepsen analysis by Kyle Kingsbury showing Amazon RDS for PostgreSQL multi-AZ clusters violate Snapshot Isolation, exhibiting Long Fork anomalies across versions 13.15 to 17.4.
Jepsen: NATS 2.12.1(jepsen.io)
412 points by aphyr | 151 comments
Jepsen analysis by Kyle Kingsbury of NATS JetStream 2.12.1, finding lost writes from file corruption, power failures, and lazy fsync behavior.
Jepsen: MySQL 8.0.34(jepsen.io)
321 points by aphyr | 119 comments
Jepsen analysis by Peter Alvaro and Kyle Kingsbury showing MySQL 8.0.34 Repeatable Read violates internal consistency and AWS RDS MySQL clusters violate Serializability.
Why is Jepsen written in Clojure?(aphyr.com)
415 points by manicennui | 210 comments
Kyle Kingsbury (Aphyr) blog post explaining why the Jepsen distributed-systems testing tool is written in Clojure, citing concurrency support and JVM interop.
Some critical issues with the SWE-bench dataset(arxiv.org)
331 points by joshwa | 113 comments
arXiv paper introducing SWE-Bench+, an empirical analysis revealing data quality issues in the SWE-bench coding benchmark for LLMs, such as solution leakage and weak tests.