ScyllaDB Announces 6.0 Release for True Elastic Scale

ScyllaDB’s tablet architecture is a breakthrough in speedy database scale agility

SUNNYVALE, Calif., June 12, 2024 (GLOBE NEWSWIRE) — ScyllaDB, the database for predictable performance at scale, today announced ScyllaDB 6.0, the first release featuring ScyllaDB’s new “tablets” replication architecture. This new architecture, which builds upon a multiyear project to implement and extend the Raft consensus protocol, enables new levels of elasticity, speed, simplicity, and efficiency.

In the era of cloud and AI, databases need to rapidly grow and shrink as millions of users access the applications powered by them. Often, a need arises to increase the database capacity to meet the demand. Elasticity speed is measured by the time it takes for a given deployment to start handling increased throughput and volume in production.

The new release, powered by tablets, demonstrates the ability to double a cluster within 15 minutes. ScyllaDB can even quadruple the throughput of a given cluster in this same short time. For reference, DynamoDB, the first NoSQL technology with the largest customer base, can only double a cluster within a 30 minute period (using on-demand mode).

With such elastic speed, customers can tighten their deployment, use what they really need on an hourly or per-minute basis, and be prepared for any given spike in demand.

The challenge of cloud database scaling

Although cloud infrastructure is inherently elastic, most databases don’t take full advantage of that elasticity. New nodes can be provisioned fast, but there’s a significant lag before they can actually serve traffic – particularly when data distribution is static (e.g., determined exclusively by a hashing function).

The new data distribution must be completed and synchronized across the cluster before new nodes can reliably service reads and writes. Even with “autoscaling,” substantially increasing cluster capacity takes time. In addition, moving massive amounts of data is challenging because it may pin too many compute and IO resources, causing a performance bottleneck.

Smooth scaling through dynamic “tablets-based” data replication

With ScyllaDB’s new tablets architecture, data is dynamically redistributed as the workload and topology evolve. New nodes can be spun up in parallel and start adapting to the load in near real-time. This means teams can quickly scale out in response to traffic spikes – satisfying latency SLAs without needing to overprovision “just in case.”

Adaptive load rebalancing further optimizes efficiency on top of ScyllaDB’s signature shard-per-core architecture, which is known for providing predictable low latency at scale (e.g., workloads exceeding 1M ops/sec).

“We used to define ScyllaDB as the power of Cassandra at the speed of Redis,” noted Dor Laor, ScyllaDB co-founder and CEO. “With the ScyllaDB 6.0 release, we add a third pillar to the equation: the usability of DynamoDB. This is ideal for high intraday traffic variations (e.g., delivery and social media), totally unpredicted surges in demand, or major known events of unknown scope (e.g., sporting events or Black Friday sales).”

For additional details on the release, see the ScyllaDB 6.0 web page.

Additional details to be revealed at P99 CONF, a free virtual conference on “all things performance”

Additional details on the latest ScyllaDB performance optimizations will be introduced at P99 CONF, the ScyllaDB-hosted conference on low-latency engineering. On October 23-24, ~20,000 latency-obsessed engineers from around the world will converge at P99 CONF to share their latest experiments, optimizations, ideas, and lessons learned. It’s purely technical, intentionally virtual, and highly interactive. In the open source spirit, the event is collaborative and free. Registration is now open: https://www.p99conf.io/

The first round of speakers introduced for P99 CONF 24 include:

  • Michael Stonebraker, Postgres creator and MIT Professor
  • Bryan Cantrill, Co-founder and CTO of Oxide Computer
  • Avi Kivity, KVM creator, ScyllaDB co-founder and CTO
  • Liz Rice, Chief open source officer with eBPF specialists Isovalent
  • Andy Pavlo, CMU professor and co-founder of Ottertune
  • Ashley Williams, Axo founder/CEO, former Rust core team, Rust Foundation founder
  • Carl Lerche, Tokio creator, Rust contributor, and engineer at AWS

The conference will feature 60+ engineering talks on performance optimizations at Disney/Hulu, Shopify, Lyft, Uber, Netflix, Turo, American Express, Datadog, Grafana, LinkedIn, Google, Oracle, AWS, ScyllaDB, and more. There will be parallel tracks of sessions covering topics such as:

  • Database optimizations
  • Rust, Zig, Go, C, C++, Wasm
  • eBPF, io_uring, kernel
  • AI/ML and LLMs
  • Kubernetes
  • Observability
  • Cloud infrastructure

About ScyllaDB

ScyllaDB is engineered to deliver predictable performance at scale. It’s adopted by organizations that demand ultra-low latency, even with workloads exceeding 1M ops/sec. Our shard-per-core architecture leverages the power of modern infrastructure – translating to fewer nodes, less admin, and lower costs. Over 400 game-changing companies like Disney+ Hotstar, Expedia, Discord, Crypto.com, Zillow, Starbucks, Comcast, and Samsung use ScyllaDB for their toughest database challenges. ScyllaDB is available as free open source software, a fully-supported enterprise product, and a fully managed service on multiple cloud providers. For more information: https://www.scylladb.com/

Media Contact

Wayne Ariola
wayne.ariola@scylladb.com