Workload Identification System for Storage Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face challenges in identifying and addressing abnormal performance issues, particularly service level objective (SLO) violations and performance problems caused by shared resource utilization among multiple applications, where bursty workload interarrival time distributions can significantly impact response times.
Innovation Solution
A method is introduced to identify problematic workloads by analyzing measured response times across applications sharing resources, determining abnormal performance based on defined criteria, and performing remediation actions such as resource allocation, workload relocation, or I/O operation pacing to mitigate performance issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple applications share common storage resources to improve resource utilization, then system efficiency is improved, but performance variability and SLO violations increase due to bursty workload interarrival times
Solution Approach 1:
The patent segments the shared storage resources into application-specific resource sets, creating isolated resource pools for each application. This segmentation allows each application to have dedicated resources while maintaining overall resource utilization, preventing bursty workloads from one application from impacting others.
Solution Approach 2:
The patent applies local quality by providing different resource allocation strategies to different applications based on their specific workload characteristics. Applications with bursty interarrival times receive dedicated resources with guaranteed performance, while applications with more uniform workloads can share resources, optimizing both reliability and productivity.
2Loss of time
If resource allocation is optimized for average workload to improve overall system performance, then average response time is reduced, but performance during workload bursts deteriorates
Solution Approach 1:
The patent implements dynamic resource allocation that adapts to workload conditions. Resource sets are dynamically assigned to applications based on their workload characteristics, particularly their interarrival time distributions. This dynamic approach ensures that resources are allocated appropriately during both average and bursty conditions, maintaining performance consistency while optimizing average response time.
Solution Approach 2:
The patent changes the allocation parameters based on workload characteristics, specifically the interarrival time distribution. By analyzing whether workloads have uniform or bursty interarrival times, the system adjusts resource allocation parameters to provide guaranteed performance for bursty workloads while maintaining efficient resource utilization for uniform workloads.
3Measurement precision
If detailed monitoring of each application's response time is implemented to identify SLO violations, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent extracts the monitoring and analysis function into a separate workload analysis component that operates independently from the core storage system. This extraction allows detailed monitoring of each application's response time and SLO compliance without adding complexity to the core storage management system. The analysis component processes workload characteristics and identifies problematic applications without interfering with normal storage operations.
4Reliability
If remediation actions are taken for each identified problematic workload to improve its performance, then individual application performance is improved, but overall system resource utilization decreases
Solution Approach 1:
The patent creates a universal resource allocation framework that serves multiple functions: it provides dedicated resources for applications requiring guaranteed performance, enables efficient sharing for applications that can tolerate variability, and maintains overall system resource utilization. The resource set allocation mechanism is multi-functional, handling both performance guarantee and resource efficiency in a unified system.
Data Source
AI summary
Described are techniques that identify problematic workloads. Measured response times for workloads associated applications are received. Each of the applications has one of the workloads resulting in one of the measured response times for the application. The applications share a set of one or more resources. In accordance with a first set of one or more criteria, it is determined whether there is an occurrence of abnormal performance with respect to performance of the applications. Responsive to determining the occurrence of abnormal performance with respect to performance of the applications, second processing is performed that includes determining, using the measured response times and in accordance with a second set of one or more criteria, an application set of one or more of the applications having an associated workload causing the occurrence of abnormal performance. A remediation may also be taken to address or alleviate the abnormal performance.