Workload Pattern Detection for Server Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches fail to effectively analyze captured workloads to detect patterns that provide insight into server requests and their consequences, limiting optimization and performance improvement.
Innovation Solution
The method involves normalizing requests to create templates, mapping these templates to identifiers, ordering them based on workload criteria, and using these patterns to generate workload models that can create artificial workloads and optimize processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If captured workload data is collected and stored for analysis, then insight into server requests and performance can be gained, but system complexity and data processing requirements increase
Solution Approach 1:
The patent extracts only the essential pattern information from captured workload data by normalizing requests into templates and identifying recurring patterns. This extraction approach gains insight into server requests without processing or storing all raw data, thereby reducing system complexity while preserving critical information.
Solution Approach 2:
Instead of analyzing individual requests to gain insight, the patent inverts the approach by aggregating requests into patterns first, then analyzing the patterns. This inversion reduces the complexity of analysis while maintaining the ability to gain meaningful insights about server request behavior.
2Productivity
If workload patterns are detected and analyzed, then optimization opportunities can be identified, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary normalization of requests into templates and pre-identification of patterns from captured workload data. This preliminary action allows optimization opportunities to be identified in advance, enabling faster decision-making and reducing the time needed for real-time performance optimization.
Solution Approach 2:
The patent creates template copies of actual requests that capture the essential patterns without duplicating all the detailed data. These template patterns serve as simplified representations that can be analyzed quickly while still providing the basis for identifying optimization opportunities.
3Measurement precision
If detailed request data is processed to identify patterns, then accurate workload models can be generated, but data processing complexity and resource consumption increase
Solution Approach 1:
The patent extracts only the critical characteristics of requests needed for accurate workload modeling by normalizing into templates. This extraction maintains measurement precision for workload patterns while avoiding the complexity of processing complete detailed request data.
Solution Approach 2:
The patent segments the workload analysis process into distinct stages: request normalization into templates, pattern identification from templates, and workload model generation from patterns. This segmentation reduces processing complexity by breaking down the complex task into manageable steps that can be performed efficiently.
Data Source
AI summary
Techniques for leveraging frequent patterns identified in a captured workload are provided. In one approach, multiple frequent patterns detected in a captured workload may be ordered by frequency to determine, for example, which patterns should be targeted for optimization. In another approach, a model of a captured workload is created, where the model comprises nodes that represent templates (which in turn correspond to requests) and edges that represent transitions between templates. The model is used to create an artificial workload, such as a workload that is twice as large as the originally-captured workload. The model may also be edited before creating the artificial workload. In another approach, workload models are compared to identify errors, regressions, or security issues. In another approach, an artificial workload is created for an application that is not yet deployed and then executed to determine whether the artificial workload or the originally-captured workload executed faster.