Unified Workload Recovery Across Data Service Platforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale distributed systems face challenges in identifying and resolving production incidents due to non-adherence to best practices, requiring manual processes that are time-consuming and prone to errors, involving coordination across multiple teams and systems.
Innovation Solution
A system and method for automatically identifying and resolving problem instances in data service workloads using a system including a processor and a non-transitory memory configured to store instructions for identifying problem instances, determining solutions using machine learning models, and executing these solutions across multiple data service platforms to resolve issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual processes are used to identify and resolve production incidents, then human expertise and judgment can be applied, but the process becomes time-consuming and prone to errors
Solution Approach 1:
The system enables self-service incident resolution by automatically detecting problem instances, determining appropriate solutions from a catalog, and executing resolutions without requiring human intervention. The machine learning model autonomously identifies issues and selects from predefined solutions, allowing the system to resolve incidents independently and reducing both time loss and human error.
Solution Approach 2:
The patent replaces manual mechanical processes with automated machine learning-based systems. Instead of human engineers manually analyzing and resolving incidents, the system uses machine learning models to detect problems, determine solutions, and execute resolutions automatically, substituting human expertise with an automated intelligent system.
2Reliability
If manual coordination across multiple teams is used to resolve incidents, then comprehensive expertise is available, but the process becomes complex and error-prone
Solution Approach 1:
The system eliminates the need for manual coordination across multiple teams by automatically determining and executing resolutions. The machine learning model independently analyzes the incident, selects appropriate solutions from the catalog, and executes them without requiring human coordination between different teams, thereby reducing coordination complexity while maintaining resolution accuracy.
Solution Approach 2:
The machine learning model acts as an intermediary between incident detection and resolution execution. It receives problem instance information, determines the appropriate solution from the catalog, and triggers execution across different systems, mediating the complex coordination process and eliminating the need for direct human coordination between multiple teams.
3Productivity
If automated systems are used to resolve incidents, then speed and consistency are improved, but the system complexity increases
Solution Approach 1:
The system reduces complexity by performing preliminary actions during the development phase, where problem instances and their solutions are identified and cataloged in advance. This preparation allows the automated system to quickly resolve incidents during production without needing complex real-time analysis, as the solutions are pre-determined and stored in the catalog.
Solution Approach 2:
The patent uses copying by creating a catalog of predefined problem instances and their corresponding solutions. Instead of developing a completely new automated resolution system for each incident type, the system copies and reuses proven solutions from the catalog, reducing the complexity of the automated system while maintaining high resolution speed and consistency.
Data Source
AI summary
Systems and methods for automatically identifying and resolving problem instances in data service workloads are disclosed. In some embodiments, a disclosed method includes: identifying a problem instance for a workload associated with a plurality of data service platforms; determining, using at least one machine learning model, a problem solution based on the problem instance and a catalog of problem solutions; executing the problem solution including operations across the plurality of data service platforms; and recovering the workload in accordance with a determination that the problem instance is resolved by the problem solution.