Cloud Streaming Failover With Pre-Provisioned Standby Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud failover techniques in cloud computing environments are inadequate for meeting stringent Recovery Time Objectives (RTOs, as they fail to activate applications and their dependencies in backup regions quickly enough during region failures, rendering applications vulnerable to prolonged downtime.
Innovation Solution
Active and standby applications in a cloud computing environment monitor the health of two different logical regions and switch modes dynamically to ensure seamless failover, utilizing a resilience library plugin to generate standby applications that can transition to active mode when the primary region fails, ensuring rapid activation in a healthy backup region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional cloud failover techniques are used to redirect traffic to backup regions, then application availability is improved, but the activation time of applications and dependencies in backup regions exceeds the defined Recovery Time Objective (RTO)
Solution Approach 1:
The patent applies preliminary action by pre-configuring standby applications and their dependencies in backup regions before failures occur. The standby applications are pre-provisioned with necessary resources and configurations, so when a failure is detected, they can be activated immediately without waiting for post-failure provisioning. This pre-positioning of resources resolves the contradiction by ensuring both availability (through pre-prepared backups) and rapid activation (within RTO requirements).
Solution Approach 2:
The patent uses copying by creating standby applications that are copies of the primary application, including its configuration, dependencies, and runtime environment. These copies are maintained in backup regions and can be rapidly activated when the primary application fails. The copying approach allows immediate failover without the time-consuming process of creating applications from scratch, thus resolving the time delay issue while maintaining reliability.
2Speed
If standby applications are pre-configured in backup regions to meet RTO requirements, then activation speed is improved, but system complexity and resource requirements increase
Solution Approach 1:
The patent applies universality by designing a failover system where standby applications can serve multiple purposes: they act as backup applications for the primary region, can be activated for failover, and can also serve as primary applications if the original primary region fails permanently. This multi-functionality reduces the need for dedicated single-purpose components, thereby managing complexity while enabling rapid activation through pre-configured standby instances.
Solution Approach 2:
The patent uses parameter changes by dynamically adjusting the state of standby applications based on the health of the primary region. When the primary region is healthy, standby applications remain in a low-resource state. When failure is detected, parameters change to activate the standby applications with full resources. This dynamic parameter adjustment allows rapid activation speed while managing resource consumption and system complexity through state-based resource allocation.
3Reliability
If conventional failover mechanisms redirect requests to redundant instances in different availability zones, then application reliability is improved, but the failover process cannot meet stringent RTO standards for data streaming applications
Solution Approach 1:
The patent applies preliminary action by pre-provisioning all necessary application components and dependencies in backup regions before failures occur. This includes pre-configuring message brokers, databases, and other dependencies that the application relies on. When a failure is detected, the pre-configured standby application can be activated immediately with all dependencies already in place, achieving both high reliability and adherence to stringent RTO requirements for data streaming applications.
Solution Approach 2:
The patent uses an intermediary approach by introducing a failover management system that coordinates between primary and standby applications, monitors their health, and manages the activation process. This intermediary layer abstracts the complexity of failover management and enables rapid, automated activation of standby applications, resolving the contradiction between maintaining reliability through redundant instances and meeting stringent RTO requirements.
Data Source
AI summary
Techniques are provided for performing a failover for one or more streaming data applications executing in a cloud computing environment. An active application may actively communicate with a logical region of the cloud computing environment. A corresponding standby application may operate in standby mode with a different logical region. The active application and its standby application may each monitor the health of the logical region and the different logical region by polling respective health status applications. When it is determined that the logical region has failed and the different logical region is healthy, the active application may transition to standby mode and the standby application may transition to active mode to actively communicate with the different logical region. When it is determined that the logical region is healthy again, the applications may switch back to their initial modes.