Apache Cassandra

→ 🦋 cassandra.apache.org Find us on bsky.app/profile/cassandra.a…

Maxim Muzafarov, author of CEP-38, a CQL-based management API, with a live demo. You can now run all nodetool commands without jmx or its port!

Francisco Guerrero on running Cassandra with mutual TLS. Certificates in place of passwords, how to rotate them, and how to migrate without an outage.

Štefan Miklošovič on dictionary compression with Zstd, CEP-54, which lands in 6.0, and probably as one of 6.0's best example of broad community collaboration, giving up to three fold storage cost stavings at 4kb chunk sizes, double at 16kb

Mariah McLaughlin and Carlos Rolo, from NetApp Instaclustr, making upgrades and migrations from 4 to 5 simple, predictable and safe.

Branimir Lambov presents CEP-57: flat keys and tries, and what they do for tombstones, partitioning, global indexes and materialised views

Youki Shiraishi, a researcher at NTT, presents comprehensive benchmark theory and results on Accord's fast path, transaction overhead, and ephemeral reads against simple, complicated and contented transactions.

Hayato Shimizu finishing the day with a live demo of CQLAI, an Apache-licensed alternative to cqlsh, built on bubbletea

Francisco Guerrero and Saranya Krishnakumar showing how the Analytics library reads SSTables through Sidecar, bypassing the read and write path entirely.

Bloomberg replaced its own in-house sidecar with the Apache one, and then started contributing back. Andrés Beck-Ruiz and Isaac Reath cover Sidecar’s architecture and features

Stefan Stoicescu from TU Delft, Software Engineering Research Group, finding non-trivial linearizability bugs by a model-guided approach that maps concrete executions to abstract TLA+ states, turning an unbounded trace-discovery graph into a usable coverage metric.

Alexandre Siret from IMDEA Software Institute on modelling Accord in TLA+

Carlos Rolo, Piotr Walczak and Sebastian Machlowski on how they moved more than four hundred Cassandra clusters, up to fifty a night.

Patrick Mcfadin on adding columnar format and using Cassandra as a HTAP database. And community health, dealing with the increased contribution rate, and shipping releases on time.

Tejal Chakeres and Abhijeet Dubey wrapping up the day with"CEP-60: Flexible Placements" Now ownership changes are linearizable, topology changes can happen in smaller steps that can be paused and resumed.

Memtables are the single largest contributor to Cassandra’s garbage collection overhead. Kamil Leśniak's talk describes the final stage of a long progression: a fully off-heap memtable implementation building on the trie memtable

Cassandra 6.0 write performance improvements from Dmitry Konstantinov the low down on how to "do less work" and "allocate less" and "contend less" and "helping the JIT"

Johnny Miller and a live demo of AIOps and what AI does for Kafka and Cassandra cluster operations today

Fedor Ryabinin getting into the depths of the Accord protocol, the first leaderless protocol deployed in a widely used production systems.

Bernardo Botella Corbi talking to native cpu profiling in Cassandra. Also with a call out to using easy-db-labs 🤩 cassandra.apache.org/_/event…

A jammed pack Cassandra Track room has started! Scott Andreas kicks it off with some great slides on "Cassandra: A Normal Database", describing how these things are getting implemented in lots of details.