Splice Machine delivers an innovative open-source RDBMS
While the Splice Machine RDBMS is unique as an open-source, operational RDBMS powered by the dual engines of Apache Hadoop® and Apache Spark™, it’s important to understand how it compares to other databases such as SQL-on-Hadoop, traditional RDBMSs, NoSQL and in-memory only solutions.
Try Splice Machine
Get started with Splice Machine, the open-source RDBMS powered by Hadoop and Spark, today!
Splice Machine vs...
The number of SQL-on-Hadoop solutions has grown rapidly because of the popularity of both SQL and Hadoop.
However, most of them have the same shortcomings:
- Analytics only
- No real-time updates
- No transactions
- Limited SQL
Splice Machine is a transactional, real-time Hadoop
RDBMS, and it is the only general-purpose SQL database
on Hadoop with the flexibility to support both OLAP and
The established RDBMS leaders work well until database sizes reach a few terabytes, when scalability becomes a huge issue:
- The high cost of scaling up RDBMSs such as Oracle and IBM DB2
- Application rewrites needed for manual sharding of MySQL, PostgreSQL, or Microsoft SQL Server databases
- The maintenance nightmare to prune data from overwhelmed databases
The Splice Machine RDBMS delivers highly cost-effective scale out on commodity servers – along with full ANSI SQL support, real-time updates with transactional integrity and in-memory performance from Apache Spark.A Brief Guide to Scaling Oracle Databases
Inspired to solve the scalability issues of traditional RDBMSs, the NoSQL movement delivered highly scalable databases capable of handling petabytes of data.
However, companies have found serious limitations with NoSQL solutions:
- Very limited SQL language support
- A lack of transactions and joins
- Costly efforts needed to recreate database functionality in each application
Splice Machine enables application developers to enjoy the best of both worlds: standard SQL support and NoSQL’s scalability on commodity hardware.Discover When to Use NoSQL vs. NewSQL
The Splice Machine RDBMS is an innovative hybrid of in-memory technology from Spark and disk-based technology from Hadoop.
Spark has a number of advantages as in-memory architecture:
- High-performance, analytical processing framework
- Efficient, in-memory processing with spill-to-disk
- Highly parallelized, pipelined computation
- Resilient against node failures
Unlike in-memory-only databases, the Splice Machine RDBMS does not force companies to put all of their data in-memory, which can become prohibitively expensive as data volumes grow.Download Splice Machine White Paper