Advanced Architecture

This article discusses XpoLog's architecture for advanced deployment scenarios, which are usually used in highly busy environments, due to enterprise regulations or multiple geographical locations.

An important part of the considerations when planning a full HA deployment including ELB, nodes array and storage replication is recovery time / tolerable loss vs. complexity and cost.

It is recommended to consult with XpoLog support prior to setting up the clustered environment.

Advanced Deployments Architecture

Full HA Installation

XpoLog cluster configuration supports high availability for both UI nodes by using a load balancer in front of the cluster that redirects users to 1 out of N nodes, and for Processor nodes by assigning an Active/Passive MASTER node and 1-M Processor nodes as described in the system architecture article.

The following diagram demonstrates XpoLog's full HA cluster including UI nodes array, Processor node array and storage replication. Multiple nodes supports full HA of the cluster usage and functionality and normally the cluster would work against its hot storage. In case of a storage failure that cluster nodes will switch to use the (replicated) cold storage which is a mirror of the Hot Storage which is normally used.

Clusters Array Installation

Some scenarios do not allow a single XpoLog cluster to process all data into a centralized silo (closed network segments, network limitation over remote geographical locations, single share storage limitations, etc.). In such cases it is possible to distribute XpoLog deployment to multiple instances / simple clusters / HA clusters across the organization.

The following diagram demonstrates XpoLog's full HA clusters array across an enterprise environment with a centralized cluster which provides access to the data stored on each remote cluster. In this case data is managed by each cluster independently and the centralized cluster only queries the remote clusters to provide users with the required information: