Machine-to-Machine Lease Agents for Cluster Failure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems lack efficient methods for detecting failures in interconnected machines, particularly in communication failures and application instance failures, which can lead to delayed or inaccurate detection of machine or application downtime.
Innovation Solution
Implementing a machine-to-machine lease mechanism using lease agents to establish and monitor communication between machines, allowing for timely detection of failures through 'heartbeat' messages and lease renewal processes, and employing arbitration to resolve ambiguity in failure detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional communication methods are used without lease agents, then system simplicity is maintained, but failure detection accuracy and timeliness deteriorate
Solution Approach 1:
The patent introduces lease agents as intermediary components that mediate between application instances and the lease management system. These lease agents enable accurate failure detection by managing lease states and coordinating heartbeat messages, while isolating the complexity from application instances. This resolves the contradiction by providing precise failure detection through the intermediary layer without requiring application instances to directly handle complex lease management logic.
2Loss of time
If lease renewal processes are implemented continuously, then failure detection timeliness is improved, but system resource consumption increases
Solution Approach 1:
The patent implements periodic lease renewal processes where lease agents exchange heartbeat messages at defined intervals rather than continuously. Application instances request lease renewals periodically, and lease agents update lease states at specific time points. This periodic approach maintains timely failure detection while significantly reducing system resource consumption compared to continuous monitoring, as resources are utilized only when lease renewals are due rather than constantly.
3Reliability
If multiple application instances monitor each other directly, then failure detection coverage is improved, but communication overhead increases
Solution Approach 1:
The patent segments the failure detection function by introducing lease agents that act as dedicated intermediaries between application instances. Instead of application instances directly monitoring each other, the system is segmented into application instances and lease agents, where lease agents handle the monitoring and communication tasks. This segmentation maintains comprehensive failure detection coverage while reducing communication overhead, as lease agents consolidate and manage communications efficiently rather than having multiple direct instance-to-instance communication channels.
Data Source
AI summary
Detecting failures of interconnected machines. A method includes establishing a machine to machine lease at a first machine using a first lease agent at the first machine to a second lease agent at a second machine as a result of an application on the first machine requesting to establish a lease with an application on the second machine. Using the machine to machine lease, the method further includes detecting a communication failure between the first machine and the second machine or a machine failure of the second machine.