SRB Redirection to zIIP Enclaves via PSA and SVT Pointer Modification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In IBM zSeries mainframe systems, Service Request Block (SRB) processes are not eligible to run on zIIP processors due to operating system limitations, leading to increased costs and inefficiencies, as only specific workloads meeting proprietary criteria can execute on zIIPs, restricting the general availability of zIIP processing capacity.
Innovation Solution
The method involves intercepting and redirecting SRB workloads from General Purpose (GP) processors to zIIP processors by modifying the Prefixed Save Area (PSA) and Supervisor Vector Table (SVT) pointers to point to intercept routines that dynamically redirect SRB invocations to zIIP-eligible enclaves, allowing SRB processes to execute on zIIPs without changing the application code or recompiling the software, and implementing a validation routine to ensure only authorized requests utilize this redirection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If SRB routines are executed on General Purpose processors, then the processes can run with standard operating system support, but the run-time costs increase and zIIP processing capacity is underutilized
Solution Approach 1:
The patent introduces an intermediary mechanism (the intercept routine and redirection infrastructure) that mediates between the SRB routine and the zIIP processor. This intermediary layer captures SRB routines that would normally execute on GPs and redirects them to zIIP-eligible enclaves, enabling cost-effective execution while maintaining OS compatibility. The intermediary resolves the contradiction by providing a bridge that allows GP-coded SRBs to run on cheaper zIIPs without requiring changes to the original application code.
Solution Approach 2:
The patent changes the execution parameter of SRB routines from GP processors to zIIP processors by modifying the PSA and SVT pointer parameters. This parameter change enables the same SRB code to execute on a different, more cost-effective processor type. The redirection mechanism changes the execution destination parameter while preserving the original routine's functionality, thus reducing run-time costs without sacrificing reliability.
2Use of energy by moving object
If SRB routines are redirected to zIIP processors, then run-time costs are reduced and zIIP capacity is enhanced, but the operating system limitations must be circumvented
Solution Approach 1:
The patent applies preliminary action by modifying the PSA and SVT pointers before SRB routines are executed. This preliminary modification of the execution environment prepares the system to redirect SRBs to zIIP-eligible enclaves. The validation routine is also prepared in advance to authorize which SRBs can be redirected. This preliminary setup circumvents operating system limitations without requiring complex runtime changes during execution.
Solution Approach 2:
The intercept routine serves as an intermediary that handles the complexity of circumventing OS limitations. It captures SRB routines, validates their eligibility for redirection, and redirects them to appropriate zIIP enclaves. This intermediary layer absorbs the complexity of working around OS restrictions, allowing the system to achieve cost reduction without exposing the complexity to the application programs.
3Adaptability or versatility
If the Prefixed Save Area and Supervisor Vector Table are modified to enable redirection, then SRB processes can execute on zIIPs, but the system requires additional validation mechanisms
Solution Approach 1:
The validation routine acts as an intermediary between the redirection mechanism and the SRB routine execution. It verifies that only authorized and eligible SRBs are redirected to zIIP processors. This intermediary validation layer ensures that the modified PSA and SVT pointers are used correctly and that only appropriate routines are redirected, thus managing the complexity introduced by the eligibility restrictions.
Solution Approach 2:
The validation routine applies local quality by selectively validating and authorizing specific SRB routines for redirection based on their eligibility criteria. Not all SRBs are treated equally; only those meeting the zIIP eligibility requirements are redirected. This localized validation approach manages complexity by applying restrictions only where necessary, while allowing other routines to execute normally on GPs.
4Ease of manufacture
If application programs are enabled to run on zIIP without code changes, then development effort is reduced and cost is lowered, but the original processor must be used if no zIIP capacity is available
Solution Approach 1:
The redirect infrastructure acts as an intermediary that enables application programs to run on zIIPs without modifying the original code. The intercept routine captures SRB calls and redirects them to zIIP-eligible enclaves, providing transparent processor migration. This intermediary mechanism achieves ease of deployment while maintaining productivity by automatically utilizing zIIP capacity when available, without requiring application-level changes.
Solution Approach 2:
The system applies dynamics by making the processor selection dynamic rather than static. Applications can dynamically execute on either GP or zIIP processors based on availability and eligibility. The redirection mechanism dynamically routes SRBs to appropriate processors, optimizing capacity utilization. This dynamic approach enhances productivity by utilizing zIIP capacity when available while maintaining ease of deployment through automatic redirection without code changes.
Data Source
AI summary
A method to redirect SRB routines from otherwise non-zIIP eligible processes on an IBM z/OS series mainframe to a zIIP eligible enclave is disclosed. This redirection is achieved by intercepting otherwise blocked operations and allowing them to complete processing without errors imposed by the zIIP processor configuration. After appropriately intercepting and redirecting these blocked operations more processing may be performed on the more financially cost effective zIIP processor by users of mainframe computing environments.