Instana, ADDI, and IBM Z Application Discovery

Instana's Current Capabilities

On Power Systems LPARs: Instana can effectively discover applications through agents installed on the operating system, identifying processes, services, and their dependencies.

On IBM Z Systems:

• Full support for z/Linux environments
• Limited support for traditional z/OS workloads
• Primarily focused on Java applications and modern middleware on z/OS
• Traditional mainframe applications (COBOL, CICS, IMS) have limited visibility

ADDI Integration Potential

Integration Approach: Using ADDI's APIs to enhance Instana's mainframe application discovery is technically feasible

Deployment Architecture:

• ADDI collectors on z/OS with minimal footprint
• Both ADDI and Instana servers on distributed systems
• Custom integration layer connecting them
• No need to deploy full Instana backend on mainframe

Integration Components:

• Custom middleware service for orchestration
• Data transformation module
• Authentication via IBM SSO
• Caching layer for performance optimization

Data Flow: Hybrid model with scheduled pulls from ADDI's API and push notifications for significant changes

Key Challenges

Data Model Reconciliation: Bridging the conceptual gap between mainframe and distributed application models

Resource Considerations: Managing mainframe MIPS consumption and performance impact

Organizational Alignment: Coordinating across traditionally siloed mainframe and distributed teams

Security & Compliance: Addressing mainframe-specific security requirements

Long-term Maintenance: Ensuring compatibility as both products evolve

Next Steps

A phased implementation approach starting with limited scope would be most effective, focusing first on application inventory before expanding to deeper dependency mapping.