ScyllaDB Engineering Resources

Under the Hood of ScyllaDB's Shard-per-Core Architecture

Learn how ScyllaDB's shard-per-core architecture takes full advantage of modern infrastructure and networking capabilities.

ScyllaDB’s Path to Strong Consistency: A New Milestone

In ScyllaDB 5.2, Raft is GA and used for the propagation of schema changes. Quickly assembling a fresh cluster, performing concurrent schema changes, updating node's IP addresses – all of this is...

What’s Next on ScyllaDB’s Path to Strong Consistency

What's next on ScyllaDB's path to strong consistency with Raft? Topology changes will be safe and fast: we won't be bound by ring delay any longer, and operator mistakes won't corrupt the cluster.

How ScyllaDB Distributed Aggregates Reduce Query Execution Time up to 20X

Explore the merits of distributing aggregates in a distributed database, delving into the benefits it provides and shedding light on the mechanisms we used to enable this transformative capability...

ScyllaDB 5.2 and Beyond

ScyllaDB co-founder and CTO Avi Kivity covers 2022 accomplishments and deliveries.

Scaling ScyllaDB Storage Engine with State-of-the-Art Compaction

Learn how ScyllaDB leverages RUM conjecture and controller theory, to deliver a state-of-the-art LSM-tree compaction for its users.

A New IO Scheduler Algorithm for Mixed Workloads

How ScyllaDB, built on the highly asynchronous Seastar library, implemented an IO scheduler optimized for peak performance on modern NVMe SSDs.

Raft After ScyllaDB 5.2: Safe Topology Changes

ScyllaDB’s drive towards strongly consistent features continues. This talk covers the upcoming implementation of safe topology changes feature: our rethinking of adding and removing nodes.

Quantifying the Performance Impact of Shard-per-core Architecture

Most software isn’t architected to take advantage of modern hardware. How does a shard-per-code and shared-nothing architecture help – and exactly what impact can it make?

Why ScyllaDB is Moving to a New Replication Algorithm: Tablets

How moving from vnode-based replication to tablets helps dynamically distribute data across the cluster, ultimately increasing ScyllaDB's elasticity.

ScyllaDB Open Source 5.2: With Raft-Based Schema Management

ScyllaDB 5.2 introduces Raft-based strongly consistent schema management,DynamoDB Alternator TTL, and many more improvements and bug fixes.

Building Next Generation Drivers: Optimizing Performance in Go and Rust

Highlights performance increases obtained using techniques available in the respective programming languages, including shaving performance off Google's B-tree implementation with Go generics and Rust

Retaining Goodput with Query Rate Limiting

How we implemented rate limiting on a per-partition basis which reduces the performance impact in such a case, and how we reduced the CPU cost of handling failed requests such as timeouts.

Implementing Highly Performant Distributed Aggregates

How we implemented aggregate functions without hammering real-time availability and performance for other read/write operations.

Keeping Latency Low for User-Defined Functions with WebAssembly

How we integrated WebAssembly into ScyllaDB in a latency-friendly manner while implementing the core runtime for user-defined functions in async Rust

How ScyllaDB Optimizes Database Architecture to Maximize Hardware Performance

An IEEE interview with ScyllaDB co-founder and CTO, Avi Kivity

The Coordinated Omission Problem