Dynamic Resource Allocation for Database Segment Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Massively parallel processing (MPP) relational databases face performance degradation and increased downtime due to the need for rebalancing and resource reallocation when a segment host fails, leading to overloaded mirror segment instances and reduced query speed and consistency.
Innovation Solution
Implementing primary and mirror segment instances using containers, allowing dynamic resource allocation and promoting mirror instances to primary roles without rebalancing, thereby reducing resource consumption and downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If mirror segment instances are allocated full computing resources to take over failed primary instances, then reliability is improved, but device complexity and performance degradation occur due to overloading
Solution Approach 1:
The patent implements dynamic resource allocation where mirror segment instances can be promoted to primary instances with automatic resource adjustment. When a primary instance fails, its mirror is promoted and the system dynamically allocates the freed resources to the promoted instance, avoiding overloading and maintaining optimal performance without complex manual intervention
Solution Approach 2:
The system changes the operational state parameter of segment instances from static roles to dynamic states. Mirror instances can transition to primary instances and their resource allocation parameters are automatically adjusted based on their new role, enabling flexible adaptation to failures while maintaining system balance
2Stability of the object's composition
If rebalancing is performed when a segment host fails, then resource distribution is improved, but productivity decreases due to downtime
Solution Approach 1:
The patent prepares mirror segment instances in advance as ready-to-promote replicas. When a failure occurs, the pre-prepared mirror can immediately take over without requiring time-consuming rebalancing operations, thus maintaining continuous query processing and avoiding downtime while ensuring proper resource distribution
3Productivity
If segment nodes operate at full capacity with both primary and mirror instances, then productivity is improved, but reliability worsens due to overload during failures
Solution Approach 1:
The system enables self-service through automatic failover where mirror instances autonomously promote themselves to primary instances upon detecting failures. This self-service mechanism ensures continuous operation at full capacity while maintaining reliability, as the promotion process is automatic and does not require manual intervention or cause system overload
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for allocating a number of first containers to implement one primary segment instance each and a number of second containers to implement one mirror segment instance each. In one example system, the second containers are configured to have less computing resources than the first containers. The containers are distributed among a number of physical computers. The system receives an indication of a failure of a physical computer hosting a particular first container implementing a first primary segment instance. In response to receiving the indication, the system promotes a second mirror segment instance that is a mirror of the first primary segment instance to be a new primary segment instance. The system allocates additional resources to the particular second container implementing the promoted mirror segment instance.