Modern distributed systems require strong consistency and a powerful query language, leading to the development of distributed SQL databases.
Distributed databases primarily solve for massive internet-scale workloads and help organizations comply with data sovereignty regulations.
A distributed database presents a single logical view to applications while physically partitioning data into replicated, independently operating shards.
Clustered databases offer the lowest latency within a data center, while distributed databases provide massive geographical scale at the cost of network latency.
A global bank implemented data sovereignty for India by sharding local customer data into a dedicated region without changing its core applications.
The BlueKai platform simplified its architecture and improved performance by migrating a complex, multi-database system to a single Oracle distributed database.
Flexible data distribution methods like composite, directory-based, and duplicated tables are crucial for minimizing latency and handling data skew.
Using adaptive replication and the Raft consensus protocol provides extreme survivability and fast, automatic failover with zero data loss.
A converged database supports multiple data types and workloads in a single system, simplifying development and deployment across multi-cloud and on-premise environments.
A fully-featured distributed database combines flexible data distribution, replication, and deployment methods within a converged architecture for maximum power.
28:41Wei Hu
21:09Tobias Petry
29:27Wei Hu
26:08Maxim Fateev
24:22Jürgen Pilz
29:30Wei Hu
30:34Markus Dreseler
29:25Matthias Niehoff
.gif?w=240&auto=compress,format)
Jobs that call for the skills explored in this talk.
Interhyp Gruppe
München, Germany
Scalefree Scalefree
Deutsche Leasing AG
Clickhouse