Distributed Event Correlation via Dynamic Partitioning and Failover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current distributed event correlation systems face challenges in efficiently managing and analyzing large volumes of event data across multiple nodes, particularly in detecting security threats and maintaining high availability in the presence of node failures.

Innovation Solution

A distributed event correlation system that implements a dynamic data partitioning scheme and rapid failover capabilities, using a master node to manage cluster membership, route events to relevant nodes, and store rules for correlation, while maintaining a standby master node for failover and utilizing independent cluster nodes for non-partitioned data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If distributed event correlation systems process large volumes of event data across multiple nodes, then the system can analyze more data in parallel, but the complexity of managing data distribution and node coordination increases

Engineering Contradiction:
Improveevent correlation throughputVSAvoidsystem management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system segments event data into different partitions distributed across multiple cluster nodes. Each node handles a specific partition, enabling parallel processing of large volumes of events while maintaining manageable complexity through clear division of responsibilities. The master node coordinates this segmentation by routing events to appropriate partitions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A master node acts as an intermediary between event sources and cluster nodes. It receives events, determines the appropriate partition based on event properties, and routes events to the correct cluster node. This intermediary layer simplifies the complexity of direct node-to-node coordination while maintaining high throughput.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the system maintains a standby master node for failover, then system availability improves during node failures, but the resource overhead and system complexity increase

Engineering Contradiction:
Improvesystem availabilityVSAvoidfailover mechanism complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

A standby master node is maintained in advance, ready to take over if the primary master node fails. This preliminary preparation ensures immediate failover capability, improving system availability. The standby node continuously monitors the primary node and can assume control without significant delay or complexity in the failover execution.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the operational state parameter of the master node from active to standby, allowing a single node to serve dual purposes at different times. This parameter change approach enables failover without requiring multiple fully redundant master nodes, reducing resource overhead while maintaining availability.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If dynamic data partitioning is implemented to balance load across nodes, then event correlation efficiency improves, but the overhead of managing partition assignments and data redistribution increases

Engineering Contradiction:
Improveevent correlation efficiencyVSAvoidpartition management overhead
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system implements dynamic data partitioning where partition assignments can be adjusted based on load conditions. The master node can redistribute events to different partitions dynamically, optimizing event correlation efficiency. This dynamic approach allows the system to adapt to changing workloads while the master node manages the overhead of partition assignments centrally.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The master node performs multiple functions: it coordinates partition assignments, routes events to appropriate partitions, monitors cluster health, and manages failover. By consolidating these functions in a single node, the system achieves efficient dynamic partitioning without requiring each cluster node to independently manage partition logic, reducing overall management overhead.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP2987090B1Distributed event correlation system
Publication Date: 2019.03.27 MICRO FOCUS LLC
  • EP2987090B1 patent drawingFigure 1
  • EP2987090B1 patent drawingFigure 2
  • EP2987090B1 patent drawingFigure 3

AI summary

According to an example, a master node is to divide an event field in events into partitions including ordered contiguous blocks of values for the event field. Each partition may be assigned to a pair of cluster nodes. A partition map is determined from the partitions and may identify for each partition, the block of the event field values for the partition, a primary cluster node, and a failover cluster node for the primary cluster node.