Database Resource Reservation and Metering for Tenant Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource sharing in database services leads to performance unpredictability for users due to varying workloads from co-located customers, resulting in a negative customer experience.
Innovation Solution
Implementing a reservation system for server resources, such as CPU, memory, and I/O throughput, with independent metering to ensure accountability and guarantee resource allocation to tenants, regardless of other tenants' workloads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If resources of a database server are shared among multiple customers, then costs are reduced, but performance predictability deteriorates
Solution Approach 1:
The patent segments the shared database server resources into isolated performance domains for each customer. By implementing customer-specific performance isolation mechanisms, the system maintains cost efficiency through resource sharing while ensuring that each customer's performance metrics (CPU usage, I/O operations, memory access) remain predictable and isolated from other customers' workloads.
Solution Approach 2:
The patent introduces performance isolation mechanisms as intermediary layers between the shared physical resources and individual customers. These intermediaries (such as virtualization layers, resource managers, or isolation policies) mediate resource allocation and performance guarantees, allowing multiple customers to share infrastructure while maintaining predictable performance boundaries for each.
2Adaptability or versatility
If complex SQL queries are allowed from multiple customers, then service versatility is improved, but performance stability deteriorates
Solution Approach 1:
The patent applies local quality by implementing customer-specific performance isolation policies tailored to individual service level agreements. Each customer receives customized performance guarantees and resource allocation rules based on their specific query workloads and service requirements, allowing diverse query types while maintaining stable performance characteristics for each customer independently.
Data Source
AI summary
Techniques for tenant performance isolation in a multiple-tenant database management system are described. These techniques may include providing a reservation of server resources. The server resources reservation may include a reservation of a central processing unit (CPU), a reservation of Input/Ouput throughput, and/or a reservation of buffer pool memory or working memory. The techniques may also include a metering mechanism that determines whether the resource reservation is satisfied. The metering mechanism may be independent of an actual resource allocation mechanism associated with the server resource reservation.


