Automated Upgrade Subsystem for Distributed Service Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed computer systems, upgrading distributed applications across multiple physical servers is a time-consuming, error-prone, and complex process, especially when manual intervention is required for each server, leading to potential incompletely upgraded systems and complex bookkeeping challenges.
Innovation Solution
An automated upgrade subsystem that facilitates a hierarchical, rolling upgrade of distributed service-based applications, allowing system administrators to initiate a system-wide upgrade with a single input, automatically generating snapshots, detecting issues, and performing a distributed rollback if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual upgrade process is used for each physical server, then system administrator can control each upgrade individually, but the upgrade process becomes extremely time-consuming and error-prone
Solution Approach 1:
The upgrade process is segmented into distinct phases: snapshot creation, upgrade execution, and verification. Each server's upgrade is divided into manageable steps with clear transition points, allowing automated orchestration while maintaining individual control over each server's upgrade journey.
Solution Approach 2:
Snapshots are created as a preliminary action before upgrades begin. This preparatory step establishes a rollback capability in advance, enabling safe automated upgrades without requiring manual intervention during the actual upgrade process.
2Productivity
If automated upgrade process is implemented across distributed computer system, then upgrade time is reduced and manual intervention is minimized, but system complexity and difficulty of managing upgrade status increases
Solution Approach 1:
The upgrade subsystem implements a universal automated upgrade mechanism that handles snapshot creation, upgrade execution, and verification across all servers in the distributed system. This single multi-functional subsystem replaces numerous manual operations and simplifies the overall upgrade management architecture.
Solution Approach 2:
The system continuously monitors upgrade status and provides feedback throughout the upgrade process. This feedback mechanism tracks which servers have been upgraded, which are in progress, and which require attention, automatically managing complexity through real-time status reporting and coordination.
3Reliability
If hierarchical rolling upgrade is performed with snapshots, then rollback capability is ensured for failed upgrades, but additional storage requirements and processing overhead are introduced
Solution Approach 1:
Snapshots are created as a preliminary protective measure before upgrades commence. This advance preparation ensures that if an upgrade fails, the system can automatically rollback to the pre-upgrade state, providing a safety net that enables confident automated upgrading across the distributed system.
Data Source
AI summary
The current document is directed to an automated upgrade subsystem within a distributed computer system that executes a distributed, service-based application comprising multiple local instances of the a distributed, service-based application, each instance including one or more control programs executing on a physical computer-system component of the distributed computer system. In one implementation, the automated upgrade subsystem provides a user interface that allows a system administrator or other user to launch a distributed-system-wide upgrade of a distributed application with a single input to a single input feature on a displayed user interface. The automated upgrade subsystem carries out a hierarchical, rolling upgrade, automatically generates snapshots for each physical computer system prior to launching local upgrade logic on the physical computer systems, detects upgrade problems and, in certain cases, automatically conducts a distributed-computer-system-wide rollback for problems that cannot be automatically or manually resolved.