Emulating Stretched Storage via Virtual Device Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity and high cost of implementing stretched storage devices across datacenter sites for enhanced availability and fault recovery solutions make it impractical for vendors and customers to test or develop software, as existing solutions require synchronized disk writes across all locations, introducing significant operational and management challenges.
Innovation Solution
An emulation layer is used to create a virtual stretched storage device by sharing storage devices across datacenter sites, allowing device identifiers to be modified to present a unified view, enabling site preference management and replication without the need for actual stretched storage devices, thus allowing for zero-downtime failover and testing without the full cost of actual stretched storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If actual stretched storage devices are deployed across datacenter sites, then availability and fault recovery are improved, but cost and device complexity increase significantly
Solution Approach 1:
The patent creates a virtual emulation layer that copies the functionality of stretched storage devices without requiring actual physical stretched storage hardware. The emulation layer presents virtual stretched storage devices to management software, allowing failover testing and development without the complexity and cost of real stretched storage infrastructure across datacenter sites.
Solution Approach 2:
The patent introduces an emulation layer as an intermediary between the physical storage infrastructure and the management software. This intermediary layer translates management commands into appropriate storage operations and presents unified device identifiers, enabling availability features to be tested and managed without direct interaction with complex stretched storage hardware.
2Reliability
If actual stretched storage devices are deployed, then fault recovery capability is improved, but the cost becomes prohibitive
Solution Approach 1:
The emulation layer creates virtual representations of stretched storage devices that can be tested and managed without requiring actual stretched storage hardware. This allows fault recovery scenarios to be simulated and tested repeatedly without the need to purchase and deploy expensive stretched storage infrastructure across multiple datacenter sites.
Solution Approach 2:
The patent enables the use of inexpensive virtual storage emulations instead of expensive physical stretched storage devices for testing and development purposes. The virtual storage devices can be created, destroyed, and reconfigured without the financial commitment required for actual stretched storage hardware.
3Reliability
If stretched storage devices are used, then site-level availability is improved, but operational management complexity increases
Solution Approach 1:
The emulation layer serves as an intermediary that simplifies operational management by abstracting the complexity of stretched storage infrastructure. Management software interacts with virtual storage devices through standardized interfaces, eliminating the need to directly manage complex synchronous disk write coordination across geographically distributed sites.
Solution Approach 2:
The patent creates virtual copies of stretched storage devices that replicate their functionality for testing and management purposes. These virtual devices present the same interface and behavior as physical stretched storage, enabling operational procedures to be developed and tested without the burden of managing actual distributed storage infrastructure.
Data Source
AI summary
Exemplary methods, apparatuses, and systems include receiving a device discovery command from a recovery manager running on a management server within a first or second datacenter. In response to the device discovery command, device identifiers for one or more logical storage devices within the first datacenter are requested. In response to the request, a device discovery response including a first identifier is received. The received device discovery response is modified to include the first device identifier as the identifier for both an underlying logical storage device and as an identifier for a peer logical storage device. The modified response is returned to the recovery manager in response to the device discovery command.