Stateless Event Objects for Encoder Instance Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The management of a large number of encoder instances in dynamic streaming environments, such as AWS MediaLive, is impractical due to asynchronous events and the limited API transactions per second, making it difficult to monitor and manage the state of multiple encoders simultaneously.
Innovation Solution
A system and method that utilize stateless event objects and state machines to manage encoder instances, where channel information is received and associated with encoder functions to create stateless event objects, which are combined into a state machine for monitoring and workflow application, enabling efficient management of encoder instances across multiple channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional encoder management methods are used with limited API transactions, then system complexity is reduced, but the ability to manage large numbers of encoder instances efficiently deteriorates
Solution Approach 1:
The patent introduces an intermediary layer between the encoding service API and the encoder instances. This intermediary manages the state of multiple encoders and coordinates API transactions, allowing efficient management of thousands of encoders despite the 5 TPS limitation. The intermediary abstracts the complexity of individual encoder management while maintaining control over the entire encoder fleet.
Solution Approach 2:
The system segments encoder management into independent, stateless encoder instances that can be individually controlled and managed. Each encoder is treated as a discrete unit with its own state information stored externally, allowing parallel management of multiple encoders through coordinated API transactions rather than requiring complex inter-encoder relationships.
2Reliability
If the API transaction limit of 5 TPS is enforced, then system reliability is maintained, but the ability to monitor and control multiple encoders simultaneously deteriorates
Solution Approach 1:
The system performs preliminary actions by pre-planning and batching encoder management operations. Before executing API transactions, the system prepares the necessary state changes and coordinates them to minimize the number of transactions required. This allows the system to maintain API stability while improving encoder monitoring capability through efficient transaction scheduling.
Solution Approach 2:
The patent implements continuous encoder monitoring and management through persistent state tracking and background synchronization processes. Rather than relying solely on periodic API transactions, the system maintains continuous awareness of encoder states through background processes that synchronize with encoder status changes, ensuring reliable monitoring without exceeding API transaction limits.
3Adaptability or versatility
If encoder instances are managed dynamically, then adaptability to different streaming scenarios is improved, but the difficulty of tracking and monitoring encoder states deteriorates
Solution Approach 1:
The system creates and maintains digital copies of encoder state information in a centralized database or data structure. Each encoder instance has its state replicated and tracked externally, allowing the management system to monitor and control encoder states without directly querying each encoder through the limited API. This copying mechanism enables easy tracking of thousands of encoder states simultaneously.
Solution Approach 2:
The patent implements a universal state management system that handles all encoder instances through a common interface and data structure. This universal approach allows the same monitoring and control mechanisms to be applied across diverse encoder instances regardless of their specific configuration or streaming scenario, simplifying state tracking while maintaining adaptability to different use cases.
Data Source
AI summary
Systems, methods, and storage media for managing encoder instances in a serverless content distribution platform are disclosed. Exemplary implementations may: receive channel information relating to content to be streamed over a networked content distribution platform; associate the channel information with an encoder function that specifies video packaging settings and bitrates to create a stateless event object; repeat the receiving step and the associating step to create multiple stateless event objects; each correspond to one or more channels of streaming content; combine the multiple stateless event objects into a state machine; monitor the state of the state machine; and apply at least one workflow based on the states to manage instances of encoders corresponding to the one or more channels of streaming content.