DBMS > RocksDB vs. Sadas Engine vs. ScyllaDB vs. Splice Machine vs. Tarantool

System Properties Comparison RocksDB vs. Sadas Engine vs. ScyllaDB vs. Splice Machine vs. Tarantool

Editorial information provided by DB-Engines

Name

Description

Embeddable persistent key-value store optimized for fast storage (flash and RAM)

SADAS Engine is a columnar DBMS specifically designed for high performance in data warehouse environments

Cassandra and DynamoDB compatible wide column store

Open-Source SQL RDBMS for Operational and Analytical use cases with native Machine Learning, powered by Hadoop and Spark

In-memory computing platform with a flexible data schema for efficiently building high-performance applications

Primary database model

Document store
Key-value store
Relational DBMS

Secondary database models

Key-value store

Spatial DBMS

with Tarantool/GIS extension

DB-Engines Ranking measures the popularity of database management systems
Trend Chart

Score	3.65
Rank	#85	Overall
	#11	Key-value stores

Score	0.00
Rank	#383	Overall
	#158	Relational DBMS

Score	4.75
Rank	#68	Overall
	#5	Wide column stores

Score	0.54
Rank	#250	Overall
	#114	Relational DBMS

Score	1.72
Rank	#144	Overall
	#25	Document stores
	#25	Key-value stores
	#66	Relational DBMS

Website

Technical documentation

github.com/facebook/rocksdb/wiki

www.sadasengine.com/en/sadas-engine-download-free-trial-and-documentation/#documentation

docs.scylladb.com

splicemachine.com/how-it-works

www.tarantool.io/en/doc

Developer

Facebook, Inc.

SADAS s.r.l.

ScyllaDB

Splice Machine

Initial release

2013

2006

2015

2014

2008

Current release

8.11.4, April 2024

8.0

ScyllaDB Open Source 5.4.1, January 2024

3.1, March 2021

2.10.0, May 2022

License

Commercial or Open Source

Open Source

BSD

commercial

free trial version available

Open Source

Open Source (AGPL), commercial license available

Open Source

AGPL 3.0, commercial license available

Open Source

BSD-2, source-available extensions (modules), commercial licenses for Tarantool Enterprise

Cloud-based only

Only available as a cloud service

DBaaS offerings (sponsored links)

Database as a Service

Providers of DBaaS offerings, please contact us to be listed.

Scylla Cloud: Create real-time applications that run at global scale with Scylla Cloud, the industry’s most powerful NoSQL DBaaS

Implementation language

C++

Java

C and C++

Server operating systems

Linux

AIX
Linux
Windows

Linux

Linux
OS X
Solaris
Windows

BSD
Linux
macOS

Data scheme

schema-free

yes

schema-free

yes

Flexible data schema: relational definition for tables with ability to store json-like documents in columns

Typing

predefined data types such as float or date

yes

string, double, decimal, uuid, integer, blob, boolean, datetime

XML support

Some form of processing data in XML format, e.g. support for XML data structures, and/or support for XPath, XQuery or XSLT.

Secondary indexes

yes

cluster global secondary indices

yes

SQL

Support of SQL

yes

SQL-like DML and DDL statements (CQL)

yes

Full-featured ANSI SQL support

APIs and other access methods

C++ API
Java API

JDBC
ODBC
Proprietary protocol

Proprietary protocol (CQL)

compatible with CQL (Cassandra Query Language, an SQL-like language)
RESTful HTTP API (DynamoDB compatible)
Thrift

JDBC
Native Spark Datasource
ODBC

Open binary protocol

Supported programming languages

C
C++
Go
Java
Perl
Python
Ruby

.Net
C
C#
C++
Groovy
Java
PHP
Python

For CQL interface: C#, C++, Clojure, Erlang, Go, Haskell, Java, JavaScript, Node.js, Perl, PHP, Python, Ruby, Rust, Scala
For DynamoDB interface: .Net, ColdFusion, Erlang, Groovy, Java, JavaScript, Perl, PHP, Python, Ruby

C#
C++
Java
JavaScript (Node.js)
Python
R
Scala

C
C#
C++
Erlang
Go
Java
JavaScript
Lua
Perl
PHP
Python
Rust

Server-side scripts

Stored procedures

yes, Lua

yes

Java

Lua, C and SQL stored procedures

Triggers

yes

yes, before/after data modification events, on replication events, client session events

Partitioning methods

Methods for storing different data on different nodes

horizontal partitioning

Sharding

Shared Nothhing Auto-Sharding, Columnar Partitioning

Sharding, partitioned with virtual buckets by user defined affinity key. Live resharding for scale up and scale down without maintenance downtime.

Replication methods

Methods for redundantly storing data on multiple nodes

yes

none

selectable replication factor

Representation of geographical distribution of servers is possible

Multi-source replication
Source-replica replication

Asynchronous replication with multi-master option
Configurable replication topology (full-mesh, chain, star)
Synchronous quorum replication (with Raft)

MapReduce

Offers an API for user-defined Map/Reduce methods

Yes, via Full Spark Integration

Consistency concepts

Methods to ensure consistency in a distributed system

Immediate Consistency

Eventual Consistency
Tunable Consistency

can be individually decided for each write operation

Immediate Consistency

Casual consistency across sharding partitions
Eventual consistency within replicaset partition

when using asyncronous replication
Immediate Consistency within single instance
Sequential consistency including linearizable read within replicaset partition

when using Raft

Foreign keys

Referential integrity

yes

Transaction concepts

Support to ensure data integrity after non-atomic manipulations of data

yes

Atomicity and isolation are supported for single operations

ACID

ACID, with serializable isolation and linearizable read (within partition); Configurable MVCC (within partition); No cross-shard distributed transactions

Concurrency

Support for concurrent manipulation of data

yes

yes, multi-version concurrency control (MVCC)

yes, cooperative multitasking

Durability

Support for making data persistent

yes

yes, write ahead logging

In-memory capabilities

Is there an option to define some or all structures to be held in-memory only.

yes

managed by 'Learn by Usage'

yes

in-memory tables

yes

yes, full featured in-memory storage engine with persistence

User concepts

Access control

Access rights for users, groups and roles according to SQL-standard

Access rights for users can be defined per object

Access rights for users, groups and roles according to SQL-standard

Access Control Lists
Mutual TLS authentication for Tarantol Enterprise
Password based authentication
Role-based access control (RBAC) and LDAP for Tarantol Enterprise
Users and Roles

More information provided by the system vendor

Specific characteristics

ScyllaDB is engineered to deliver predictable performance at scale. It’s adopted...
» more

Competitive advantages

Highly-performant (efficiently utilizes full resources of a node and network; millions...
» more

Typical application scenarios

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

Key customers

Discord, Epic Games, Expedia, Zillow, Comcast, Disney+ Hotstar, Samsung, ShareChat,...
» more

Market metrics

ScyllaDB typically offers ~75% total cost of ownership savings, with ~5X higher throughput...
» more

Licensing and pricing models

ScyllaDB Open Source - free open source software (AGPL) ScyllaDB Enterprise - subscription-based...
» more

We invite representatives of system vendors to contact us for updating and extending the system information,
and for displaying vendor-provided information such as key customers, competitive advantages and market metrics.