New 5.5

Client Multi-Member Routing

Optimize geographically dispersed clusters

The highest performance and continuity for geographically dispersed (stretch) clusters

Multi-member routing works with Hazelcast partition groups. In multi-member routing mode, the client connects to all members of a single partition group. Since each partition group contains all of the data, read operations can be routed directly to the member that contains the data, improving overall throughput. In this way, it behaves similar to “smart clients” and provides many of the same benefits. As with “unisocket clients”, if access to a member in a different partition group is required, any of the members to which the client is already connected can act as a proxy.

  • High performance and stability for critical applications connecting to geographically dispersed clusters.
  • Instant client failover in the event of planned or unplanned downtime.
  • Suitable for disaster recovery, business continuity and data localisation.
  • Greater operational stability and efficiency with the new model.
Client Multi-Member Routing Diagram
Client Multi-Member Routing
  • Additional intelligence added to client
    • Partition-group aware
    • CP Subsystem aware
  • Added capability to connect based on partition group

Features

Data Locality for Clients

Clients of geographically dispersed clusters can now preferentially access members of the closest partition group while still being able to access any member when necessary.

Advanced CP Enabled

Strong Data Consistency across client multi-member configured nodes.

Easy Configuration

Client multi-member routing is easily enabled using the existing client network configuration.

High Performance Client Re-Routing

Achieve fast client failover for geographically dispersed clusters in the event of planned downtime or an outage.

Highest Levels of Connectivity and Resilience

Reduces connection overheads and improves overall performance for geographically dispersed clusters.

Take the next step

See how strong consistency, performance, resilience, and scale drive an AI-centric future.