Scylla Open Source 3.2 introduces a rich set of new production-ready features for IPv6, LIKE operators, as well as experimental features such as Lightweight Transactions (LWT), Change Data Capture (CDC) and a DynamoDB-compatible API. Already the industry’s most performant NoSQL database, Scylla Open Source includes features beyond the capabilities of Apache Cassandra.
Scylla LWTs allow stronger consistency guarantees using the Paxos consensus algorithm. They ensure requests on a distributed database are processed in a strict, linearized (serial) method, in a process known as ‘Compare and Set.’. They are also called ‘Conditional Updates,’ because they can test the databases’ existing values before submitting the update. This provides atomic consistency for single keys, allowing updates to be performed in order on a global basis. They can also be used for batches, to ensure all conditions are met before submitting a batch update.
This feature allows you to track changes made to a base table in your database for visibility or auditing. Changes made to the base table are stored in a separate table that can be queried by standard CQL. Our CDC implementation uses a configurable Time To Live (TTL) to ensure it does not occupy an inordinate amount of your disk.
Our Amazon DynamoDB-compatible API implementation, known as Project Alternator, enables you to connect your DynamoDB applications to Scylla without any changes to your client code. This gives you the multi-cloud, multi-vendor flexibility you need for resiliency and disaster recovery.
Scylla’s LIKE operator allows wildcard pattern searches. It supports both single character or string wildcards, and also verbatim and empty field matches. These additional operators mean that you can create queries on Scylla you cannot make in Cassandra. Also unlike Apache Cassandra, Scylla’s LIKE operator is not reliant on SSTable Attached Secondary Indexes (SASI).
This feature allows database administrators to escape the limitations of the Internet Protocol version 4 (IPv4) 32-bit address space established back in the 1980s. IPv4 was limited to about 4.3 billion devices on the Internet. IPv6 uses 128-bit addresses allowing 3.4×1038 addressable devices. This enables and facilitates the continued expansion of data networks.
Rebuilding a failed node is a non-trivial operation. It requires decommissioning the old node, spinning up a new one, and streaming data from the replicas to bring the new node up to speed. Scylla’s RPC streaming mechanism makes it more efficient to transfer large amounts of data, cutting time-to-rebuild dramatically.
Creates a persistent table that is a subset of a base table, using different keys and thus accelerating the search. The view and the base consistencies are managed by Scylla automatically. Materialized Views are now fully production ready in Scylla 3.0, while still experimental in Apache Cassandra.
Global Secondary Indexes (GSI) allows you to query data through non-primary key columns, minimizing the amount of data retrieved from the database and resulting in faster interactions. GSI is compatible with Apache Cassandra; however, with Scylla the index is cluster-global and search does not depend on the size of the cluster. An enhancement added in Scylla Open Source 3.1, Local Secondary Indexes (LSI) allows Scylla to optimize workloads where the partition key of the base table and the index are the same key.
Scylla Open Source 3.1 is now supported on the AWS i3en instances, where each node can store up to 30TB of data. Significantly larger databases can now be created, simplifying development, deployment and enhancing user experiences.
Speeds up repairs by working at the row level. This significantly reduces the time for repairs and lowers the volume of data that needs to be transferred between nodes when repairs are needed.
Get more query flexibility by appending queries with ALLOW FILTERING to bypass some Cassandra query restrictions. Designed for low-selectivity queries, the ALLOW FILTERING command can be executed on regular and primary key columns, reducing the network traffic back to the client and resulting in better application performance.
Now compatible with the latest Apache Cassandra file format (SSTable “mc”), which results in less disk space used and improved performance. The storage volume is reduced by as much as 3X.
Automatically recover intermediate failures by storing up to 3 hours of untransmitted data between nodes. This improves consistency and reduces the repair time and amount of work, resulting in higher overall performance for large clusters.
Increases system responsiveness and efficiency by improving locality and caching of range scans. This feature increases the throughput of range scans by 30%, reduces the amount of data read from the disk by 40% and lowers disk operations by as much as 75%.
Boost the speed of auto-scaling the cluster as well as replacing faulty nodes. These improvements result in higher performance for the overall system when there is a change in the cluster topology.
Read our press release announcing Scylla Open Source 3.2 from our Scylla Summit last fall
Getting started takes only a few minutes. Scylla has an installer for every major platform. If you get stuck, we’re here to help.