GDDR Message Affinity Evaluation for Mainframe Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing geographically dispersed disaster restart (GDDR) systems require dedicated logical partitions for each control-system instance, leading to increased hardware costs and challenges in ensuring autonomy and affinity-based communication between GDDR complexes coexisting on the same mainframe, particularly in preventing reactions to messages outside of their affinity group.
Innovation Solution
Implementing a method that ensures a particular GDDR instance does not react to messages from outside its affinity group by evaluating message identifiers, scheduling service request blocks, verifying the address space, and comparing subsystem connectors, allowing multiple GDDR complexes to share the same set of control systems and reducing the number of mainframes needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple GDDR instances coexist on the same logical partition, then hardware costs are reduced and resource utilization is improved, but message affinity management becomes complex and unintended reactions to external messages may occur
Solution Approach 1:
The system segments message handling by introducing affinity groups that partition message processing responsibilities. Each GDDR instance is associated with specific affinity groups, and the message affinity evaluation mechanism segments the overall message processing task into instance-specific subtasks, allowing multiple instances to coexist without interference.
Solution Approach 2:
The patent introduces an intermediary message affinity evaluation mechanism that sits between message reception and processing. This intermediary evaluates whether a received message belongs to the current GDDR instance's affinity group before allowing processing, preventing unintended reactions to external messages while enabling efficient coexistence.
2Reliability
If dedicated logical partitions are allocated to each GDDR instance, then message affinity and autonomy are ensured, but hardware costs increase and resource efficiency decreases
Solution Approach 1:
The patent makes the logical partition universal by allowing it to host multiple GDDR instances simultaneously. The message affinity evaluation mechanism enables the same partition to serve multiple instances with different affinity groups, making the hardware resource multi-functional and eliminating the need for dedicated partitions for each instance.
Solution Approach 2:
The system changes the parameter of message handling from dedicated-instance-based to affinity-group-based. By introducing affinity groups as an additional organizational layer, the system allows multiple instances to share resources while maintaining autonomy through parameter-based message routing decisions rather than physical partitioning.
3Speed
If GDDR instances process all received messages, then message responsiveness is improved, but unintended reactions to messages outside their affinity group occur
Solution Approach 1:
The system performs preliminary action by evaluating message affinity before processing the message. The affinity evaluation occurs immediately upon message reception, determining in advance whether the current GDDR instance should process the message. This preliminary check ensures fast response times for relevant messages while preventing unintended processing of irrelevant messages.
Solution Approach 2:
The message affinity evaluation mechanism provides feedback to the GDDR instance about whether a received message belongs to its affinity group. This feedback loop allows the instance to quickly determine processing eligibility without extensive analysis, maintaining fast response times while ensuring processing accuracy through the affinity verification feedback.
Data Source
AI summary
The system, products, and methods disclosed herein relate to an improved configuration for geographically dispersed data restart (“GDDR”) systems. Specifically, the teachings herein ensure a particular GDDR instance does not react to messages issued from address spaces outside of its affinity group. We use the term message “affinity group” to mean the combination of a particular GDDR instance and its set of Mainframe Enablers started tasks. By ensuring that the particular GDDR instance does not react to messages intended for recipients outside of its affinity group, we improve computer functionality and efficiency by facilitating further partitioning of the logical partition. Once the logical partition is further partitioned into additional address spaces, it is possible to run Mainframe Enablers and/or additional GDDR instances independent of the particular GDDR instance of concern.