ScyllaDB Systemeigenschaften

ScyllaDB is the database for data-intensive apps that require high performance and low latency at scale. It enables teams to harness the ever-increasing computing power of modern infrastructures — eliminating barriers to scale as data grows. Unlike any other database, ScyllaDB is built with deep architectural advancements that enable exceptional end-user experiences at radically lower costs.

ScyllaDB is a highly-available, highly-scalable, highly-performant NoSQL database. It is generally used as a key-value store or wide-column model database.

Cassandra Compatibility — ScyllaDB was designed to be API-compatible with Apache Cassandra. It is completely compliant with Cassandra Query Language (CQL). It also uses the same SSTable format on disk, making migrations from Cassandra to ScyllaDB easy.

ScyllaDB took the best characteristics of Apache Cassandra and did a complete rewrite from the ground up in C++. This means no JVM tuning and no garbage collection (GC) pauses. Built on the Seastar engine, ScyllaDB provides all the best features of Cassandra with a built-for-speed engine under the hood.

DynamoDB Compatibility — ScyllaDB also offers an Amazon DynamoDB-compatible API, known as Alternator. It supports the same SDKs, modified to allow for client-side load balancing. ScyllaDB also offers an experimental feature for Alternator Streams.

This allows you to run your workloads on an open source alternative on AWS, or on any other public cloud (Azure, Google Cloud). You can also run on hybrid or multi-cloud deployments, or even on-premises.

DBaaS — ScyllaDB Cloud is the Database-as-a-Service (DBaaS) offering of ScyllaDB. It runs on AWS and is certified for use in AWS Outposts, and now has a beta for GCP.

Extensible to Graph or Time Series — With additional packages such as JanusGraph or KairosDB, ScyllaDB can serve as a graph database or a time series database (TSDB).

• Highly-performant (efficiently utilizes full resources of a node and network; millions of IOPS per node)

• Highly-available (peer-to-peer, no single-point-of-failure, active-active, automatic cross-datacenter replication with rack-awareness and datacenter-awareness)

• Highly-scalable both horizontally (automatic rebalancing of data with new nodes) and vertically (takes advantage of modern multi-CPU, multi-core architectures)

• Self-tuning performance (minimizes administration)

• Tunable consistency model

• Consistent low-latency and high throughput

• Free online training via ScyllaDB University

ScyllaDB is ideal for applications that require high throughput and low latency at scale. 

Common use cases include:

• IoT (Fleet Monitoring/Logistics, Machine Health/ Predictive Maintenance, Smart Devices, etc.) 
• Customer experience (User Profile, Messaging, Chat, User Activity, Leaderboard, Financial Trans, Betting, Customer 360, etc.)
• AI/ML Feature Store
• Fraud and Threat Detection (Malware Detection, Network Monitoring, Security Monitoring, Risk Management, Identity Management, etc.)
• Blockchain (Crypto, Smart Contract, etc.)

Discord, Epic Games, Expedia, Zillow, Comcast,  Disney+ Hotstar, Samsung, ShareChat, Palo Alto Networks, Grab, Starbucks, Tubi, Rakuten, Fanatics [see Users page]

ScyllaDB typically offers ~75% total cost of ownership savings, with ~5X higher throughput and ~20X lower latency versus similar databases. [See benchmarks]

• ScyllaDB Open Source - free open source software (AGPL)

• ScyllaDB Enterprise - subscription-based self-managed (proprietary license)

• ScyllaDB Cloud - fully-managed DBaaS (various pricing models: annual, monthly) [Pricing Calculator]