Per-Tenant Workload Availability Determination in Multi-Cloud Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining workload availability in multi-tenant systems on a per-tenant basis is challenging due to the complexity of managing maintenance schedules and redundancy across multiple workloads.
Innovation Solution
A monitoring system that tracks metrics for workloads, considers maintenance schedules, and accounts for redundancy by generating reports that indicate availability based on specified intervals and patterns, using allowance heuristics to adjust for delays and transitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional approaches are used to determine workload availability, then the system structure remains simple, but the ability to determine availability on a per-tenant basis deteriorates due to inability to account for maintenance schedules and redundancy
Solution Approach 1:
The patent segments the workload availability determination by creating separate data structures and tracking mechanisms for each tenant. Each tenant's workloads are evaluated independently using tenant-specific availability data structures that track individual workload states, maintenance schedules, and redundancy relationships. This segmentation enables precise per-tenant availability measurement without requiring a complete system-wide redesign.
Solution Approach 2:
The patent implements preliminary action by pre-establishing availability data structures, maintenance schedules, and redundancy relationships before availability determination is needed. The system pre-processes workload information, maintains updated availability data structures, and prepares redundancy mappings in advance, allowing for rapid and accurate availability determination when required without complex real-time calculations.
2Measurement precision
If maintenance schedules are tracked for all workloads, then availability determination accuracy improves, but the complexity of managing and processing maintenance information increases
Solution Approach 1:
The patent applies local quality by maintaining maintenance schedule information at the specific level where it is needed - within each tenant's availability data structure. Rather than a centralized complex maintenance management system, each tenant's workloads have their own maintenance schedule tracking, allowing for accurate availability determination localized to each tenant without requiring complex global coordination of maintenance information.
3Measurement precision
If redundancy relationships between workloads are analyzed, then availability determination becomes more accurate, but the computational complexity and time required for analysis increases
Solution Approach 1:
The patent reduces analysis time by performing preliminary action - pre-establishing redundancy relationships and storing them in availability data structures before availability determination is needed. The system pre-processes workload relationships, identifies redundant workloads in advance, and maintains this information in ready-to-use data structures, eliminating the need for complex real-time analysis when availability is queried.
Data Source
AI summary
In general, embodiments of the technology relate to determining availability of workloads executing on a multi-tenant system on a per-tenant basis. More specifically, embodiments of the technology determine workload availability by taking into account maintenance schedules of the workloads (or the underlying physical cloud infrastructure) and/or whether a given workload is redundant.