Storage-Aware Application Scheduling for Stateful Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud native application platforms require manual and complex user intervention for storage-aware scheduling, especially for stateful applications, leading to degraded agility and increased downtime during maintenance and recovery.
Innovation Solution
A method and system for automated application placement management using a storage agent to identify accessible servers or clusters, update data accessibility, and notify the application scheduler, enabling automated maintenance and recovery of stateful applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual storage-aware scheduling rules are configured by users, then application data accessibility is maintained, but operational complexity and time overhead increase significantly
Solution Approach 1:
The storage agent automatically monitors storage volume status and notifies the application scheduler of accessible servers, eliminating the need for users to manually configure and update affinity rules. The system self-manages the scheduling rules based on real-time storage accessibility, reducing operational complexity while maintaining data accessibility.
Solution Approach 2:
The storage agent continuously monitors storage volume status and provides feedback to the application scheduler about which servers can access which volumes. This automated feedback mechanism ensures applications are always scheduled on servers with data accessibility without requiring manual rule updates.
2Reliability
If manual affinity rules are updated whenever storage configuration changes, then application data accessibility is maintained, but agility and response time deteriorate
Solution Approach 1:
The storage agent automatically detects storage configuration changes and updates the application scheduler's knowledge of accessible servers without human intervention. This self-service approach maintains data accessibility while dramatically improving operational agility by eliminating manual rule update cycles.
Solution Approach 2:
The storage agent proactively monitors storage status and prepares accessibility information before applications need to be scheduled or restarted. This preliminary action ensures that when maintenance or failures occur, the system can immediately respond with accurate accessibility information, improving both reliability and agility.
3Loss of time
If storage status monitoring and notification automation is implemented, then downtime is reduced, but system complexity increases
Solution Approach 1:
The storage agent acts as an intermediary component that bridges storage monitoring and application scheduling. It automatically detects storage status changes and communicates accessibility information to the application scheduler, reducing downtime without requiring complex integration between storage and compute layers.
Solution Approach 2:
The storage agent performs multiple functions: monitoring storage volume status, determining server accessibility, and notifying the application scheduler. This multi-functional approach reduces overall system complexity by consolidating automation logic into a single component rather than requiring separate systems for each function.
Data Source
AI summary
A method for application placement management. The method comprising identifying, by a storage agent, a first server from a plurality of servers or a first cluster from a plurality of clusters, the first server or the first cluster can access a first volume through which an application can be executed; identifying, by the storage agent, data associated with the application, wherein the data is stored in the first volume; identifying, by the storage agent, a group of servers from the plurality of servers or a group of clusters from the plurality of clusters having access to the data; updating, by the storage agent, data accessibility associated with each server of the group of servers or each cluster of the group of clusters; and notifying, by the storage agent, the updated data accessibility associated with each server of the group of servers or each cluster of the group of clusters.