Tenant Workload Classes for Database Resource Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-tenant database architectures face challenges in efficiently managing resource sharing between tenants, leading to unpredictable performance degradations due to the 'noisy neighbor' effect, where one tenant's resource usage affects others, and require cost-effective solutions for tenant-specific resource isolation.
Innovation Solution
Implementing a native multi-tenant database system with configurable and enforceable resource limits per tenant, using workload classes to define and manage resource usage, including thread, memory, and I/O limits, and allowing for a hierarchy of workload classes to inherit and add specific resource limits for each tenant.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single database instance is used to manage data of multiple tenants, then resource sharing efficiency is improved, but performance predictability deteriorates due to noisy neighbor effects
Solution Approach 1:
The patent segments the single database instance into multiple isolated workspaces, where each tenant's data and operations are confined to their own workspace. This segmentation allows resource sharing at the database instance level while preventing noisy neighbor effects through workspace-level isolation, thus maintaining both high resource efficiency and predictable performance.
Solution Approach 2:
The patent applies local quality by providing different resource allocation characteristics to different tenants within the same database instance. Each workspace can have customized resource limits and isolation levels tailored to specific tenant requirements, enabling differentiated service quality while sharing underlying infrastructure.
2Reliability
If a full database instance is provisioned per tenant, then tenant data isolation is improved, but cost efficiency deteriorates due to excessive resource consumption
Solution Approach 1:
The patent merges multiple tenants into a single shared database instance while maintaining logical isolation through workspaces. This combining approach eliminates the need for separate database instances for each tenant, reducing overall resource consumption while preserving data isolation through workspace boundaries and security contexts.
Solution Approach 2:
The patent creates a universal database instance that serves multiple tenants simultaneously through the workspace abstraction. The single database instance performs multiple functions - serving different tenants with different data, security requirements, and resource allocations - thereby eliminating the need for dedicated instances and reducing total resource consumption.
3Reliability
If resource limits are enforced per tenant, then performance predictability is improved, but system complexity increases due to workload class management
Solution Approach 1:
The patent uses parameter changes by defining workload classes with specific resource allocation parameters (memory limits, CPU quotas, I/O thresholds). These parameters can be adjusted to match different tenant requirements and workload characteristics, enabling flexible resource management and predictable performance through quantitative control rather than complex qualitative rules.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
Systems and methods include establishment of a first database session with a first database user, determine first session variable values associated with the first database session, the first session variable values including a first tenant value, determine a first workload class based on the first tenant value and on zero or more of the other first session variable values, the first workload class including first parameter values specifying first resource consumption limits, and manage resource consumption of the first database session based on the first parameter values.