Workload Capsule Generation and Broker Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Information processing systems often have unused capacity due to a lack of mechanisms to encourage utilization, leading to inefficiencies and underutilization of resources.
Innovation Solution
The implementation of workload capsules that can be distributed across processing platforms, allowing for efficient and flexible processing of workloads by utilizing unused capacity, with a broker device managing the distribution and execution of these capsules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional cloud-based systems are used, then service availability is maintained, but resource utilization efficiency deteriorates due to unused capacity
Solution Approach 1:
The patent creates a universal workload capsule format that can be executed on any processing platform regardless of its native environment. The capsule encapsulation mechanism allows workloads to be portable and platform-agnostic, enabling the same workload to run on different hardware architectures and operating systems. This multi-functionality resolves the contradiction by allowing workloads to be dynamically assigned to the most efficiently utilized resources while maintaining service availability.
Solution Approach 2:
The broker device serves as an intermediary between workload sources and processing platforms. It receives workload capsules, matches them with appropriate processing platforms based on capacity and capability, and manages the execution lifecycle. This mediator resolves the contradiction by optimizing resource allocation across the ecosystem while ensuring workloads are always executed on suitable platforms, thereby improving utilization without sacrificing productivity.
2Loss of energy
If workload distribution across platforms is implemented, then resource utilization is improved, but system complexity increases
Solution Approach 1:
The patent segments the workload distribution system into distinct functional components: workload encapsulation modules, broker devices for matching and coordination, and execution environments on processing platforms. Each component has a specific, well-defined responsibility. This segmentation resolves the complexity contradiction by creating a modular architecture where the broker handles matching logic, platforms handle execution, and capsules provide standardized interfaces, thereby simplifying the overall system despite the distributed nature.
Solution Approach 2:
The patent changes the state of workloads from platform-specific to platform-agnostic by encapsulating them in a standardized format with defined parameters and interfaces. The workload capsule structure transforms workloads into portable units with standardized metadata, resource requirements, and execution contexts. This parameter standardization resolves the complexity contradiction by enabling automatic matching algorithms to operate on consistent data structures, reducing the complexity of cross-platform coordination.
3Productivity
If spare capacity is utilized through workload offloading, then productivity is improved, but reliability may deteriorate due to distributed execution
Solution Approach 1:
The patent implements beforehand cushioning through comprehensive workload encapsulation that includes all necessary execution contexts, dependencies, and error handling mechanisms within the capsule itself. The broker performs validation and compatibility checking before assigning workloads, and the encapsulation process ensures that all required resources and configurations are bundled with the workload. This prior preparation resolves the reliability contradiction by ensuring that distributed execution does not compromise workload integrity or execution success.
Data Source
AI summary
An apparatus comprises at least one processing platform having one or more processing devices. The processing platform is configured to generate a plurality of workload capsules for transmission to a broker device over a network. A given one of the workload capsules comprises a self-contained and independent workload to be performed by another processing platform as determined by the broker device in accordance with one or more attributes specified in the workload capsule and one or more criteria specified in the workload capsule. The given workload capsule further comprises a process description identifying at least one of server resources, storage resources, network resources and physical security resources to be utilized by the other processing platform in performing the workload. The broker device may be part of a central compute depository platform that receives workload capsules from multiple source devices over the network.