REDOOP 红象云腾
← Back to customer storiesOptimization case / REDOOP CASE STUDY

Big data cluster optimization solution

Big data parallel systems are complex: a single incident can involve hardware, network, operating system, applications, the platform and data scale, so failures need analysis from an overall data-processing perspective.

Big data cluster optimization solution

Design approach

Establish a monitoring and optimization system that improves stability, security and flexibility while reducing downtime and raising efficiency — with less manual intervention and more automated operations.

Design approach

Architecture

Monitoring business flows together with cluster hardware, operating systems and platform runtime data to form an end-to-end monitoring and optimization system for satellite data processing.

Architecture

Outcome

Failure events are identified from full-flow runtime data and drive a handling workflow, with modules for runtime monitoring, anomaly detection, anomaly handling, bottleneck optimization and inspection statistics.

Outcome

This page is based on the publicly published case study on the original website; anonymous customers remain anonymous. Figures and results describe the original project.

LET’S BUILD WHAT’S NEXT

The next possibility starts here.

Talk to us about the Redoop big data platform that fits your business.

Talk to us