Converged Infrastructure Placement Domains for Multi-Tier Application Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The management of converged, hyper-converged, and hybrid cloud information handling systems faces challenges in ensuring proper deployment, operation, and performance optimization, particularly in achieving objectives like high availability and performance due to the fragmented nature of resource acquisition and management in traditional data center models.
Innovation Solution
An infrastructure services manager that utilizes placement domains, including fault and optimization domains, to dynamically manage and monitor resources, ensuring compliance with placement objectives by intelligently placing application tiers within the information handling system, and updating domain information in response to changes in the infrastructure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional silo-like data center architecture is used, then resources are dedicated to specific applications, but capital costs and operating costs increase while utilization decreases
Solution Approach 1:
The patent implements a converged infrastructure where compute, storage, and networking resources are pooled into shared resources that can be dynamically allocated to multiple applications. The infrastructure services manager enables these shared resources to serve multiple purposes and multiple applications simultaneously, replacing the traditional dedicated resource model with a universal resource pool that improves utilization while maintaining application performance requirements.
Solution Approach 2:
The patent introduces dynamic resource allocation through the infrastructure services manager, which can dynamically assign and reassign shared resources to different applications based on demand. This dynamic capability allows the system to adapt resource distribution in real-time, optimizing utilization while ensuring each application receives appropriate resources when needed.
2Reliability
If vertically oriented data centers with dedicated resources are implemented, then application performance is ensured, but capital costs and operating costs increase
Solution Approach 1:
The patent merges previously separate dedicated resources for different applications into a single converged infrastructure. By combining compute, storage, and networking resources into shared pools that serve multiple applications, the system reduces the total quantity of hardware required while maintaining the performance guarantees needed for each application through intelligent resource management.
3Productivity
If horizontal data centers with virtualization are implemented, then costs are reduced and utilization improved, but fragmented resource management occurs
Solution Approach 1:
The patent introduces an infrastructure services manager as an intermediary layer between the shared physical resources and the virtualized application resources. This intermediary manages the complexity of resource allocation, placement, and optimization automatically, allowing the system to maintain simple unified resource pools while delivering complex multi-tenant virtualized environments with proper isolation and performance guarantees.
4Adaptability or versatility
If separate acquisition of processing, storage, and networking resources is performed, then resource specialization is achieved, but interoperability and management efficiency decrease
Solution Approach 1:
The patent segments the infrastructure into distinct functional domains (compute domain, storage domain, networking domain) that can be independently managed and optimized for their specific functions. Simultaneously, these segmented domains are integrated through the infrastructure services manager to form a cohesive converged infrastructure, achieving both specialization and interoperability.
Data Source
AI summary
A system and method access domain information indicating placement domains for an information handling system. The placement domains may include fault domains and optimization domains, wherein an optimization domain includes one or more resources wherein a tier instance for each tier of a multi-tier application service resource can be instantiated such that inter-tier communication is internal to the domain. Tier instances may be placed in accordance with the placement domains to achieve compliance with high availability and performance objectives. Management endpoints corresponding to each resource may be monitored and, responsive to detecting a change in the infrastructure, updated domain information indicative of updated placement domains may be accessed and used to determine whether the placement of the tier instances achieves compliance with the objectives. If the current placement fails to achieve the objectives, the plurality of tier instances may be re-placed in accordance with the updated placement domains.