Multi-Cloud Deployment Controllers for Concurrent Service Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for managing deployment of PaaS applications across multiple service clouds and data centers are inefficient, requiring manual intervention and resulting in slow serial deployments, potential user errors, and lack of audit trails for staging and launching applications.
Innovation Solution
A system and method for staging and deploying applications to multiple service clouds spanning multiple data centers, involving a workstation that receives and stages deployable installations, specifies deployment launch, and uses deployment controllers to select, install, and remap API URLs, while inventorying dependencies and reporting completion, enabling asynchronous and concurrent deployments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual deployment methods are used for PaaS applications across multiple service clouds, then deployment can be performed with simple tools, but deployment speed is slow and user errors occur frequently
Solution Approach 1:
The deployment system performs self-service through automated controllers that independently manage deployment tasks across multiple service clouds. The system automatically stages deployable installations, selects appropriate versions, installs applications, remaps API URLs, and inventories dependencies without requiring manual intervention at each step, thereby eliminating user errors and accelerating deployment.
Solution Approach 2:
The deployment system implements comprehensive feedback mechanisms through audit trails that record all deployment activities. The system provides real-time status reporting on deployment progress, completion confirmation, and error detection, allowing the automated controllers to adjust and correct deployment operations dynamically across multiple service clouds.
2Loss of time
If serial deployment is used for applications across multiple data centers, then deployment process is simple to manage, but total deployment time is excessive
Solution Approach 1:
The deployment process is segmented into distinct independent tasks that can be executed concurrently: staging deployable installations, selecting latest versions, installing applications, remapping API URLs, and inventoring dependencies. Each service cloud receives and executes its own segment of the deployment process independently, enabling parallel execution across multiple data centers and dramatically reducing total deployment time.
Solution Approach 2:
The system performs preliminary actions by pre-staging deployable installations in local storage at each data center before actual deployment is initiated. This preliminary preparation allows the deployment controllers to immediately begin installation and configuration tasks without waiting for file transfers, significantly reducing the critical path of the deployment process.
3Manufacturing precision
If automated deployment system is implemented across multiple service clouds, then deployment consistency is improved, but system complexity increases
Solution Approach 1:
The deployment system implements universal multi-functional controllers that can stage, select, install, remap, and inventory across different service clouds and data centers using a single unified interface. The same automated controller handles diverse deployment tasks (major releases, minor updates, hot fixes) consistently across the entire multi-cloud infrastructure, ensuring deployment precision without requiring separate specialized systems for each function.
4Productivity
If concurrent asynchronous deployment is launched to multiple service clouds, then deployment efficiency is maximized, but coordination complexity increases
Solution Approach 1:
The system employs intermediary deployment controllers that act as mediators between the central deployment initiation and the distributed service clouds. These controllers receive deployment launch specifications, coordinate the asynchronous concurrent execution across multiple service clouds, manage local storage staging, and aggregate completion status back to the initiation point, thereby enabling high-efficiency parallel deployment without overwhelming coordination complexity.
Data Source
AI summary
The technology disclosed describes staging and deploying major releases, updates and hot fixes to service clouds spanning data centers that include hardware stacks. User-specified builds of deployable installations are received as file sets, and the installations are staged to local storage at the data centers. User-specified deployment launch specifications that specify multiple already-staged deployable installations and service clouds to execute a deployment are received; and asynchronous, concurrent deployment of the multiple deployable installations by the service clouds based on the deployment launch specification are launched. Deployment controllers select a latest deployable installation staged to the service clouds, install and start staged applications in the latest deployable installation, remap incoming API URLs from a replaced version to the started staged application in the latest deployable installation, inventory dependencies, post-remapping, on the replaced version and mark the replaced version for deletion if no dependencies remain active; and report completion of deployment.