Connection Abstraction Logic for Adaptive Interconnect Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The static allocation of interconnects in data centers leads to inefficiencies due to mismatched quality of service targets for different workloads, resulting in delays and over-provisioning of communication capabilities.
Innovation Solution
A system with connection abstraction logic units that dynamically allocate interconnects based on quality of service targets, allowing workloads to communicate effectively through a single interface, optimizing interconnect utilization and offloading processing overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If interconnects are allocated statically to workloads, then device complexity is reduced and ease of operation is improved, but productivity decreases due to mismatched quality of service targets causing delays and over-provisioning
Solution Approach 1:
A connection abstraction logic unit is introduced as an intermediary between workloads and interconnects. This mediator dynamically selects appropriate interconnects based on quality of service targets, transparently managing the complexity of interconnect allocation while enabling workloads to execute at optimal speeds without directly handling allocation complexity
Solution Approach 2:
The connection abstraction logic unit automatically performs interconnect allocation and selection based on quality of service targets without requiring manual intervention or complex configuration by operators. The system self-manages the allocation process, reducing operational complexity while maintaining high productivity
2Productivity
If multiple interconnects are allocated to different workloads based on their quality of service targets, then productivity is improved by reducing delays and over-provisioning, but device complexity increases due to dynamic allocation requirements
Solution Approach 1:
The connection abstraction logic unit serves as a mediator that handles the complex task of dynamically allocating multiple interconnects based on quality of service targets. It manages the allocation complexity internally while presenting a simplified interface to workloads, thereby improving data center efficiency without exposing operational complexity
Solution Approach 2:
The system segments the interconnect allocation function into a dedicated connection abstraction logic unit, separating the complex allocation management from the workload execution. This segmentation allows the allocation management complexity to be handled in a specialized component while maintaining overall system productivity
3Ease of operation
If a single communication interface is exposed to workloads, then ease of operation is improved by simplifying the interface, but device complexity increases due to the need for connection abstraction logic units
Solution Approach 1:
The connection abstraction logic unit acts as an intermediary that provides a single simplified communication interface to workloads while handling the complexity of multiple interconnect allocations in the background. This mediator absorbs the complexity of managing multiple interfaces, presenting only a simple interface to workloads
Solution Approach 2:
The connection abstraction logic unit provides a universal single interface that works with multiple different interconnect types and configurations. This multi-functional unit handles various allocation scenarios through a unified interface, simplifying workload operations while managing the underlying complexity
Data Source
AI summary
Technologies for providing adaptive utilization of different interconnects for workloads include a compute device. The compute device includes a connection abstraction logic unit to determine a quality of service target to be satisfied in the execution of a workload that is to communicate with at least one other workload through one or more interconnects of a set of interconnects associated with the compute device, determine a quality of service property of each interconnect of the set of interconnects, and allocate, as a function of the determined quality of service property of each interconnect, one or more of the set of interconnects to the workload to satisfy the quality of service target. The compute device also includes circuitry to execute the workload and communicate with the at least one other workload through the allocated one or more interconnects. Other embodiments are also described and claimed.