Pega deployment architecture
Pega Platform™ is a distributed and scalable system that supports enterprise-grade applications across multiple deployment models, including cloud and container-based environments. Understanding Pega deployment architecture helps Lead System Architects design environments that meet performance, availability, security, and operational objectives.
Pega Platform deployments share the following characteristics, regardless of deployment model:
- Distributed and clustered run-time architecture in which multiple Pega nodes communicate to form a cluster.
- Deployment across multiple Pega nodes that are organized into one or more tiers.
- Shared Rules and data in a relational database, with run-time processing performed on individual nodes.
- Support for high availability by using clustered nodes and routing user traffic through a load balancer or Kubernetes ingress.
- Integration with externalized services, such as Search and Reporting Service.
- Compatibility with cloud-native and container-based platforms that use Docker images and Kubernetes for deployment.
For more information, see the Understanding the Pega Platform deployment architecture topic on the Pega Documentation site.
Pega Platform nodes work together as a cluster, sharing Rules and data while distributing interactive and background processing workloads.
Core architectural components
A Pega deployment architecture includes the following core components, which define how the platform processes user requests, runs background workloads, stores application data, and integrates with external services.
Pega nodes and clusters
A Pega node is a run-time instance of Pega Platform. In modern environments, nodes run as containerized instances orchestrated by Kubernetes and operate as part of a Pega cluster.
Key characteristics include the following items:
- Nodes in a cluster communicate with each other.
- Nodes share Rules and access data in a common relational database.
- Workloads are distributed across nodes to support resilience and scalability.
Nodes can be organized into deployment tiers, allowing architects to scale and tune processing for different workload types.
Web and batch tiers
A processing tier is a logical grouping of Pega nodes that perform a specific workload category. Typical deployments include two primary tiers.
The web tier is responsible for:
- Handles interactive, user-initiated requests.
- Exposed through a load balancer or Kubernetes ingress.
- Optimized for responsiveness and concurrency.
The batch tier is responsible for:
- Handles background and asynchronous processing.
- Not exposed directly to user traffic.
- Supports workloads such as batch jobs, background processing, analytics, data movement, and integrations.
Separating workloads by tier allows architects to tune and scale resources based on processing characteristics.
Load balancer or Kubernetes ingress
A load balancer or Kubernetes ingress is the controlled entry point into a Pega deployment.
Its responsibilities include the following items:
- Routing requests to available web tier nodes.
- Enabling horizontal scaling of web nodes.
- Providing resilience by redirecting traffic during node failures.
- Acting as a single-entry point for user access.
Database layer
Pega Platform uses a relational database to store Rules, application metadata, and transactional data.
Key characteristics of the database layer include:
- A split schema architecture that separates Rules from data.
- A shared schema used by all nodes in the cluster.
- Centralized governance for availability and sizing.
Because all nodes depend on the database, database architecture and performance are critical considerations.
Externalized services
Modern architectures use externalized services to improve scalability and operational efficiency. These services can be Pega managed or client managed.
Examples include:
- Search and reporting services.
- Messaging and streaming services.
- Analytics and decisioning services.
Externalization enables independent scaling and supports cloud native patterns.
For more information, see the Externalization of services in your deployment topic on the Pega Documentation site.
Pega deployment models
Pega Platform supports multiple deployment models to meet enterprise and regulatory needs. These models define where the platform runs and who manages the infrastructure.
- Pega Cloud®: A Pega-managed deployment model in which platform and infrastructure services are operated by Pega.
- Client managed cloud: A deployment model in which Pega Platform runs in the client’s cloud environment, and the client team manages the infrastructure.
- Pega Cloud for Government: A deployment model that supports governmental security and compliance requirements.
Check your knowledge with the following interaction: