Task Supervisor Architecture for Scaling Legacy Cloud Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing or modifying services to be horizontally scalable across many simultaneous users in a cloud environment is challenging, requiring additional time and resources, and legacy services often need re-coding to support horizontal scalability and orchestration with other services.
Innovation Solution
A service management platform that subdivides jobs into tasks executed by node servers, using a supervisor environment for communication and a native environment for computations, allowing legacy applications to scale horizontally and manage tasks efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If legacy services are modified to support horizontal scalability and distributed environment execution, then the service can be accessed by many simultaneous users and deployed across multiple servers, but additional development time and resources are required and re-coding is needed
Solution Approach 1:
The service is divided into independent worker processes that can execute in parallel across multiple servers. Each worker process handles individual service requests, allowing the system to scale horizontally by simply adding more worker processes without requiring complex re-coding of the legacy service logic.
Solution Approach 2:
A message queue system acts as an intermediary between the service manager and worker processes. The service manager submits service requests as messages to the queue, and worker processes consume these messages independently. This decoupling allows legacy services to run unchanged while gaining horizontal scalability through the intermediary message queue infrastructure.
2Ease of manufacture
If a service is designed for single user execution, then the service logic is simpler and development is faster, but the service cannot be horizontally scaled to support multiple simultaneous users
Solution Approach 1:
Multiple identical copies of the single-user service logic are created as separate worker processes. Each worker process is an independent copy that can handle service requests autonomously. This allows the simple, well-tested single-user service logic to be replicated across numerous processes, achieving horizontal scalability without modifying the original service code.
Solution Approach 2:
The worker process design creates universal service execution units that can handle any service request assigned to them. Each worker process is a multi-functional unit capable of executing different service types by consuming appropriate messages from the queue, allowing simple service logic to serve multiple purposes and users simultaneously.
Data Source
AI summary
A service management platform can implement functionality for one or more services, each of which can be independently used by a plurality of clients of the services. To activate the functionality of the one or more of the services, a hub server of the service management platform can assign a set of tasks to individual node servers for execution. The hub server can operate in a “supervisor environment” distinct from the processing environment used to execute the computationally intensive portions of the tasks. A task received at a node server can be managed by a supervisor process within the supervisor environment and executed by a native process within a native operating system environment, where the native process executes the computationally intensive calculations of the task and supervisor process provides communications and data transfer between the native process and rest of the service management platform.