Hydra 2021
| title | author | length |
|---|---|---|
| Algorithms for practical distributed agreement | Naama Ben-David | 01h 09m |
A practical RDMA-based state machine replication system called MU uses Infiniband one-sided operations to achieve a one-sided round trip commit in the common… | ||
| Data parallelism from a multicore perspective | Maurice Herlihy | 01h 12m |
Surveys how to implement map‑reduce on a single multicore: split input into mapper threads that produce key/value streams and merge them with a reducer… | ||
| Distributed systems showdown — TLA + vs real code | Jack Vanlightly | 01h 12m |
Built a minimalist BookKeeper-style distributed log both as a TLA+ model and as a Java Maelstrom/Jepsen prototype; TLA+ uses two specs (~300 and ~1,000 lines)… | ||
| Serverless nature of | Andrey Fomichev | 59m |
Yandex VDB implements a NewSQL, horizontally scalable database built from tablets stored on a distributed storage layer (no local disk) and an actor framework… | ||
| Workshop. Lincheck: Testing concurrency on the JVM (Part 1) | Maria Sokolova | 01h 10m |
Lincheck is a JVM testing tool (Kotlin examples) that declares operations and a correctness contract (default: linearizability) and runs either stress or… | ||
| Workshop. Lincheck: Testing concurrency on the JVM (Part 2) | Maria Sokolova | 01h 11m |
Hands-on use of Lincheck in Kotlin with IntelliJ demonstrates concrete checks: using checkObstructionFreedom in model checking mode finds an infinite read loop… | ||
| Workshop: TLA+ in action (Part 1) | Markus A Kuppe | 01h 14m |
Workshop builds a TLA+ specification of Safra’s EWD998 termination detector, iterates from a token-free counter to the token+count version, and exercises the… | ||
| Workshop: TLA+ in action (Part 2) | Markus A Kuppe | 01h 11m |
Workshop models Safra’s EWD998 termination detector in TLA+ using TLC model checking and Lamport’s tutorial style, exercises include reducing Naturals to a… | ||
| Workshop: TLA+ in action (Part 3) | Markus A Kuppe | 01h 13m |
A runnable TLA+ model of Dijkstra/Safra EWD998 appears, modeling node color and per-node counters, a token record with position Q and color, and message… | ||
| Workshop: TLA+ in action (Part 4) | Markus A Kuppe | 01h 12m |
Workshop models EWD998 termination detection in TLA+, writes a low‑level spec that replaces counters with inboxes and a recursive sum, and checks safety and… | ||
| Simplifying global-scale strong consistency | Andras Gerlits | 01h 12m |
A prototype global-commit protocol uses token-passing logical clock and event-sourced consensus groups to assign atomic commit timestamps locally and associate… | ||
| Designing fast lock free algorithms by understanding cache coherence dynamics | Adam Morrison | 01h 08m |
Shows that cache-coherence dynamics (the MESI-style directory flow and forward/invalidation messages) make contended CAS operations serialize and incur latency… | ||
| CAP Theorem — two decades and few clouds later | Mike Kowalski | 01h 15m |
Explains that CAP remains a hard constraint but is practical and nuanced: S3 claims strong read‑after‑write while buckets stay eventually consistent and S3 SLA… | ||
| Fearless global transactions with | Nathan VanBenschoten | 01h 11m |
CockroachDB runs a shared‑nothing cluster of 512MB ranges with Raft (default 3x replication) and targets a 20ms global write budget by combining three SQL… | ||
| Building scalable and flexible cluster managers using declarative programming | Lalith Suresh | 01h 12m |
A system called Declarative Cluster Managers stores cluster state in an embedded SQL DB (H2), compiles declarative SQL constraints into Java and feeds them to… | ||
| Co-designing Raft + thread-per-core execution model for the Kafka-API | Alex Gallego | 01h 12m |
Redpanda is a Kafka‑API storage engine built around a thread‑per‑core model and a co‑designed Raft path, using the systar library, per‑core arenas with no… | ||
| The hitchhiker's guide to distributed transactions | Irfan Sharif | 01h 14m |
Surveys systems that reduce commit latency from two WAN round trips to one WAN round-trip, comparing concrete mechanisms: leader-based consensus… | ||
| The official ten-year retrospective of NewSQL databases | Andy Pavlo | 01h 11m |
Term "NewSQL" originates 2011 and means relational + ACID for scalable OLTP, aiming to combine SQL transactions with NoSQL scale. Implementations split into… | ||
| Workshop: Java Concurrency Stress (JCStress). Part 1 | Aleksey Shipilev | 01h 20m |
JCStress is an OpenJDK harness that empirically probes the Java Memory Model by running millions of tiny configurations; it reproduces an atomicity failure in… | ||
| Workshop: Java Concurrency Stress (JCStress). Part 2 | Aleksey Shipilev | 02h 01m |
JCStress walkthrough shows how the harness actually finds concurrency bugs: it forks a host JVM per test and generates task/run loops with forced inlining and… | ||
| nothing matches | ||