Workload Availability Module for Multi-Site Data Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multisite continuous availability solutions are limited by geographic distance and require significant modifications to software and hardware, leading to high implementation costs and unsupported workloads, resulting in potential data loss and inconsistency during site unavailability.
Innovation Solution
A system with a first computing site as an active instance and a second site as a standby instance, using a workload availability module to replicate priority workload data and storage images across multiple sites, ensuring continuous availability and minimal downtime across unlimited distances without modifying existing applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If continuous availability solutions are implemented between sites at unlimited distances, then geographic distance limitation is removed, but when a site becomes unavailable, the remaining site runs in single-site configuration without continuous workload availability
Solution Approach 1:
The patent applies preliminary action by pre-configuring standby instances at secondary sites before primary site failure occurs. These standby instances are pre-initialized with necessary software and infrastructure, so when primary site becomes unavailable, they can immediately activate and maintain continuous workload availability without requiring site-to-site communication or configuration changes.
2Reliability
If traditional disaster recovery solutions are used, then backup capability is provided, but systems require several hours to restart and need significant software and hardware modifications
Solution Approach 1:
The system performs preliminary actions by pre-initializing standby instances with all necessary software applications, operating systems, and supporting infrastructure before failure occurs. This pre-preparation eliminates the need for time-consuming restart procedures and software modifications after primary site failure, reducing recovery time from hours to minutes.
Solution Approach 2:
The patent uses copying by creating and maintaining standby instances that are replicas of primary instances at secondary sites. These copies include all necessary software, data, and configuration, allowing immediate activation without modification or reconfiguration, thus eliminating restart delays and software modification requirements.
3Reliability
If site-level disaster recovery is implemented, then basic backup functionality is achieved, but it operates at site level rather than workload level, limiting flexibility
Solution Approach 1:
The patent applies segmentation by dividing the disaster recovery system into workload-level components rather than site-level components. Each priority workload can be independently replicated to standby instances at secondary sites, allowing selective activation and management of individual workloads. This enables greater flexibility in managing different workloads with different availability requirements.
Solution Approach 2:
The system implements dynamics by allowing flexible configuration and management of standby instances at the workload level. Priority workloads can be dynamically assigned to standby instances at secondary sites, and the system can adaptively manage which workloads are replicated and activated based on availability requirements, providing operational flexibility that site-level solutions cannot achieve.
Data Source
AI summary
A system for maintaining a two-site configuration for continuous availability over long distances may include a first computing site configured to execute a first instance associated with a priority workload, the first instance being designated as an active instance; a second computing site configured to execute a second instance of the priority workload, the second instance being designated as a standby instance; a software replication module configured to replicate a unit of work data associated with the priority workload from a first data object associated with the active instance to a second data object associated with the standby instance, and a hardware replication module configured to replicate an image from a first storage volume to a copy on a second storage volume, wherein the first storage volume is associated with the first computing site, and the second storage volume is associated with a third computing site.