Cloud Broker Dynamic Resource Allocation Across Federated Infrastructures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customer-driven datacenters face high costs and resource inefficiencies due to the need for maximum throughput, hardware replication for disaster recovery, and frequent upgrades, which can be costly and result in underutilized resources.
Innovation Solution
A cloud broker system that selectively enables access to processing resources across multiple cloud management infrastructures, allowing for dynamic allocation of computing capacity based on demand, geographic location, and priority, thereby leveraging underutilized resources and reducing the need for dedicated datacenters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If datacenters reserve spare capacity and replicate hardware for high availability and disaster recovery, then reliability is improved, but device complexity and cost increase
Solution Approach 1:
The system segments the datacenter infrastructure into multiple independent clouds, each managed by its own cloud management infrastructure. This allows capacity to be divided into discrete units that can be dynamically allocated and shared across multiple organizations, reducing the need for each organization to maintain full redundant hardware for disaster recovery.
Solution Approach 2:
The patent creates a universal cloud broker system that mediates between multiple cloud management infrastructures and end users. This broker enables a single hardware resource to serve multiple functions and multiple customers simultaneously, allowing spare capacity in one cloud to be utilized by another cloud, thereby reducing the need for duplicate hardware replication.
2Productivity
If datacenters upgrade infrastructure to handle maximum capacity requirements, then productivity is improved, but loss of energy and cost increase
Solution Approach 1:
The system implements dynamic capacity allocation where cloud management infrastructures can selectively start and stop services based on actual demand. The cloud broker continuously monitors capacity requirements and reallocates resources dynamically, allowing infrastructure to adapt to varying workload demands rather than maintaining static maximum capacity, thereby reducing energy consumption and upgrade costs.
Solution Approach 2:
The patent changes the operational parameters of datacenter resources by enabling them to operate at variable capacity levels rather than fixed maximum capacity. The cloud broker adjusts resource allocation parameters based on real-time demand, allowing infrastructure to be utilized more efficiently without requiring costly upgrades to handle peak loads.
3Reliability
If datacenters maintain dedicated infrastructure for peak capacity, then reliability is improved, but loss of substance and resource underutilization increase
Solution Approach 1:
The patent merges multiple cloud management infrastructures into a federated system coordinated by a cloud broker. This consolidation allows spare capacity from one infrastructure to be combined with capacity from others, ensuring that dedicated infrastructure for peak capacity is utilized more fully across the federation rather than remaining underutilized in isolated datacenters.
Data Source
AI summary
A cloud broker receives a request for processing resources from one of a plurality of cloud management infrastructures (each of which correspond to one of a plurality of clouds). The cloud broker selectively enables access to the clouds associated with each cloud management infrastructure. The clouds each comprise segregated computing environments having at least one server for executing tasks and a data store for data persistency attached to the at least one server. The cloud management infrastructures selectively starts and stops services executed by the corresponding cloud. After receiving the request, the cloud broker identifies which of the plurality of clouds coupled to the cloud broker has processing capacity to fulfill the request. Subsequently, the cloud broker enables access to at least one of the identified clouds having capacity to fulfill the request. Related apparatus, systems, techniques and articles are also described.