IST AZURE DIE NEUE MAINFRAME!!!
This article is written with the sole purpose of providing a start point to those organisations who are thinking in the direction of moving IBM Mainframe workloads to Microsoft Azure.
This is not an actual migration guide but covers most aspect which will stir you to the right direction in order to do your due diligence.
For better understanding, I have categorized the article into –
Business Drivers to move IBM workloads to Azure.
Where IBM Mainframe workloads migrations stands under cloud adoption waves.
Is Azure ready for Mainframe workloads?
Reference Architectures: Moving IBM Mainframe workloads to Azure.
Azure HA Deployment Architecture.
Cost benefit on running IBM Mainframe workloads on Azure.
Mainframe Migration Components.
IBM Mainframe to Azure | Apple to Apple comparison on components.
BUSINESS DRIVER TO MOVE IBM WORKLOADS TO AZURE:-
More flexibility to respond to faster infrastructure changes.
Reduce mainframe cost balance IT cost.
Growing concern on availability of mainframe skillset.
WHERE IBM MAINFRAME WORKLOADS MIGRATION STANDS UNDER CLOUD ADOPTION WAVES:-
WAVE 1: Dev/Test Environment, Web Servers, Big Data, SaaS Platform (HRM, CRM)
WAVE 2: Infrastructure Critical Servers (Backup and Email), DRaaS, Devops, Bursty Application Servers
WAVE 3: Legacy Mainframe, ERP, Containers, DTaaS (Development and Test as a Service)
IS AZURE READY FOR MAINFRAME WORKLOADS?
Yes it is and here is why:
Mainframe workloads have been running on Open systems for decades now, the first migration on Unix Systems was in Year 1995. UNIX is supported platform in Azure.
x86 chipsets are proven capable of running Mainframe Workloads. x86 is supported chipset in Azure.
Boost Performance = Heavy Intensive IO operations. We have SSDs (Solid State Devices) Available in Azure which along with x86 chipset is now becoming a commodity Hardware.
For Availability of Application anytime anywhere, Azure has data centers across 42 regions providing Clustering and Redundancy.
REFERENCE ARCHITECTURE: MOVING IBM MAINFRAME WORKLOADS TO AZURE:-
IBM Mainframe Components:
How Mainframe Components looks in Azure:
AZURE HA DEPLOYMENT ARCHITECTURE:-
Users connect to the Azure Traffic Manager.
Azure Traffic Manager load Balances between Azure Load Balancers deployed in Each Regions (Primary and Secondary)
Azure Load Balancers distributes workloads between VMs in Availability Set.
There is an Active Geo Replication between Azure SQL.
COST BENEFIT ON RUNNING IBM MAINFRAME WORKLOADS ON AZURE:-
One of the biggest cost benefits of the Azure platform is running development and test (Dev/Test) environments. For mainframe workloads on their Z Systems, IBM has a Dev/Test environment for the x86 platform called IBM Z Development and Test Environment (IBM zD&T for short).
You should ask the below questions before considering migration of IBM Mainframe Workloads to Azure -
How application supports the business?
What Mainframe Infrastructure Consists of?
How the application is Architected?
What is the complexity of the application?
What are the dependencies between the systems?
This will allow creating a centralized Fact space/repository based on the above Artifacts.
MAINFRAME MIGRATION COMPONENTS:-
Essentials Components for Mainframe Migration to Azure include –
Micro Focus Enterprise Server = Available on Azure to Run z/OS workloads on a less expensive x86 platform.
TmaxSoft OpenFrame = Available on Azure and is used in Lift-and-Shift scenarios.
Rehost, Rebuild, Replace, or Retire? IaaS or PaaS? To determine the right migration strategy, please refer – https://docs.microsoft.com/en-us/azure/architecture/cloud-adoption/infrastructure/mainframe-migration/migration-strategies
IBM MAINFRAME TO AZURE | APPLE TO APPLE COMPARISON ON COMPONENTS:-
The below table explains as which components of IBM Mainframe correlates to Mainframe related services in Azure –