Event-driven microservices on the MapR Converged Data Platform can act on both current and historical data. Combining file, database, and streaming services with an underlying publish-and-subscribe framework, information from recent events is integrated with deep insights from accumulated data.
Microservices represent an important application architecture in big data today because they offer tremendous agility. They are relatively simple, single-purpose applications that work in unison via lightweight communications, such as data streams. Therefore they are much easier to build, integrate, and coordinate relative to traditionally large monolithic applications and can be reused as different use cases and solutions require.
Support for event-driven microservices relies on underlying capabilities in the Converged Data Platform including:
- Agile microservices app development: Integrated file, database, and streams functionality, requiring fewer lines of code
- Simplified monitoring/management: Easier maintenance across entire deployment lifecycle, including application/data versioning
- Publish-subscribe framework: Integrated publish-and-subscribe framework to support event-driven applications
These microservices capabilities benefit from key features of the existing converged data platform including:
- Unified security with Access Control Expressions for data volumes, including streaming data
- Support for agile and containerized application development on Docker, including applications with persistent data requirements
- Converged analytic, processing and messaging services in the same physical nodes for real-time requirements and greater simplicity
- Support for hybrid cloud microservices architectures with global message service, global data namespace, bidirectional data flows with loop detection, and service scale out capabilities
- Logical and functional isolation of services. Ideal for machine learning model training and evaluation
- Continuous high availability and multi-master mission-critical disaster recovery capabilities
Microservice Support Details
Support for agile microservices application development and evolution
- Flexible, publish-subscribe framework with enterprise-grade messaging supports sharing data while maintaining logical and functional isolation of services. Ideal for machine learning model training and evaluation
- Natural infrastructure for service modification without disrupting production workflows
- Flexibility for developers to combine file, database, document, and streaming analytics functionality
Simplified microservices monitoring and management
- Comprehensive monitoring of resource usage for a single pane of glass view
- Microservices specific volumes for application versioning, simplifying the development lifecycle and production deployment
- Support for containerized applications especially in Docker provide utilization and agility advantages
- Continuous high availability and distributed multi-data disaster recovery capabilities
- Unified security with access control expressions for stream access and analytics
Integrated publish-and-subscribe framework to support event-driven applications
- Ultra scale, low latency, utility grade reliable messaging system as foundation of micro-service at the application level
- Combines global messaging service with processing services in the same physical nodes for greater efficiency and simplicity than alternative architectures
- Integrated data-in-motion and data-at-rest to support real-time applications
- Support for hybrid cloud microservices architectures with global message service, global data namespace, and service scale out capabilities.
See the Converged Application Blueprint for more information on converged applications, microservices, and other big data architectures.
And if you have any questions for us or for the community, please go to our Answers page of the Converge Community.