Harnessing the 21st Century with a Distributed Memory-Centric SQL
The 21st century is not just the beginning of a new millennium, but also the century that pushes the boundaries of technologies and frameworks that we have used for a while. Billions of devices, appliances and gizmos are connected to the Internet, generating massive and ever increasing volumes of data. Millions of payments and trades are issued and closed every second. Countless aeroplanes are buzzing above our heads taking off and landing in an attempt to push back traffic jams that are usually caused by an insignificant delay.
To keep pace with the massive and growing data volumes and workloads, some companies are trying to scale-up their infrastructure and technologies, investing significant resources, but facing the possibility of a dead end. Other companies have moved to distributed NoSQL database systems, sacrificing SQL capabilities and ACID guarantees. Distributed in-memory SQL database systems seem to offer a solution, but they are missing persistence capabilities that are undoubtedly essential.
In this session attendees will learn about the Apache Ignite Distributed Memory-Optimized SQL that addresses many of the challenges of the 21st century. This is a memory-optimized database that easily distributes data and workload across a cluster of thousands of nodes. The database easily scales to petabytes of data located both in RAM and on disk and operates equally well under OLTP and OLAP workloads. This session will cover the following:
- Apache Ignite Distributed SQL Database architecture: virtual memory, B+Tree, primary and secondary storage.
- Distributed SQL Engine internals: indexing, execution flow, distributed joins, etc.
- DML and DDL.
- APIs and tools: ODBC, JDBC and more.