Decoupling Compute and Storage in Cloud HCI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud-based hyper-converged infrastructure (HCI) implementations are limited by fixed form factor hosts with predefined compute-to-storage ratios, leading to suboptimal capacity economics, operational complexities in host maintenance and failure handling, and inflexible scaling of storage relative to compute resources.
Innovation Solution
Decoupling compute and storage resources by provisioning diskless hosts from a public cloud's host platform and attaching network-attached storage volumes from a separate cloud storage platform, allowing for flexible configuration and independent scaling of storage capacity, and implementing intelligent failure handling workflows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed form factor hosts with predefined compute-to-storage ratios are used, then deployment simplicity is maintained, but storage capacity flexibility and capacity economics deteriorate
Solution Approach 1:
The patent segments the host system into compute resources and storage resources that can be independently provisioned and scaled. Compute hosts and storage volumes are separate entities that can be configured independently, allowing customers to optimize storage capacity per host without being constrained by fixed form factors.
Solution Approach 2:
The patent implements dynamic provisioning where storage capacity can be adjusted independently from compute resources. Customers can scale storage up or down based on actual needs without changing compute capacity, enabling flexible adaptation to varying storage requirements.
2Ease of manufacture
If fixed form factor hosts are used, then initial deployment is simplified, but operational flexibility and failure handling efficiency deteriorate
Solution Approach 1:
By separating compute and storage into independent provisionable entities, the system allows flexible operational management. Storage volumes can be detached from failed hosts and attached to new hosts independently, simplifying failure recovery operations.
Solution Approach 2:
The patent enables dynamic changing of storage capacity parameters independent of compute parameters. This allows operational flexibility to adjust storage allocation based on actual needs and failure scenarios without reconfiguring compute resources.
3Device complexity
If storage and compute are coupled on fixed hosts, then resource allocation is straightforward, but scaling efficiency and resource optimization deteriorate
Solution Approach 1:
The patent divides the infrastructure into independent compute hosts and storage volumes that can be scaled separately. This segmentation enables efficient scaling where storage can be increased without proportionally increasing compute resources, optimizing resource utilization and reducing costs.
Solution Approach 2:
Storage volumes become universal resources that can be attached to multiple different compute hosts. This multi-functionality allows the same storage pool to serve various compute workloads, improving overall system efficiency and scaling flexibility.
Data Source
AI summary
Techniques for decoupling compute and storage resources in a hyper-converged infrastructure (HCI) are provided. In one set of embodiments, a control plane of the HCI deployment can provision a host from a host platform of an infrastructure on which the HCI deployment is implemented and can provision one or more storage volumes from a storage platform of the infrastructure, where the storage platform runs on physical server resources in the infrastructure that are separate from the host platform. The control plane can then cause the one or more storage volumes to be network-attached to the host in a manner that enables a hypervisor of the host to make the one or more storage volumes available, as part of a virtual storage pool, to one or more virtual machines in the HCI deployment for data storage.