VDI Storage Allocation via Usage and Penalty Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Virtual Desktop Infrastructure (VDI), thin-provisioned virtual machines (VMs) lead to storage over-commitment issues due to inefficient allocation of resources, where unused storage is not effectively utilized, and newly created VMs are often assigned to datastores that are already over-committed, resulting in potential storage pool fullness risks.
Innovation Solution
A system and method that considers usage values and penalty values for thin-provisioned VMs to calculate capacity values for datastores, penalizing datastores with assigned thin-provisioned VMs and adjusting their selection probability, ensuring more accurate allocation and reducing over-commitment by using a Virtual Desktop Management Server (VDMS) to manage VM provisioning based on storage statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If thin provisioning is used to allocate storage on demand, then storage utilization is improved, but storage over-commitment risk increases
Solution Approach 1:
The system implements feedback mechanisms by continuously monitoring storage consumption of thin-provisioned VMs and using this information to dynamically adjust provisioning decisions. The VDI receives consumption information from datastores and uses it to inform subsequent provisioning choices, creating a closed-loop system that adapts to actual usage patterns and prevents over-commitment.
Solution Approach 2:
The system performs preliminary actions by pre-calculating capacity values that incorporate expected consumption patterns before making provisioning decisions. By estimating future storage needs and incorporating penalty values for recently assigned datastores, the system proactively prevents over-commitment rather than reacting to it after it occurs.
2Productivity
If VMs are assigned to datastores with currently available capacity, then provisioning speed is improved, but storage distribution uniformity deteriorates
Solution Approach 1:
The system changes the parameter used for datastore selection from simple current capacity to a calculated capacity value that incorporates multiple factors including expected consumption and penalty values. This parameter transformation maintains provisioning speed while achieving more uniform storage distribution across datastores.
Solution Approach 2:
The calculated capacity value acts as an intermediary between current capacity and provisioning decisions. Instead of directly using current capacity to make assignment decisions, the system introduces this intermediate calculated metric that balances speed and uniformity concerns.
3Reliability
If storage is dedicated to VMs upon creation, then VM performance reliability is improved, but storage waste increases
Solution Approach 1:
The system transitions from static storage allocation to dynamic thin provisioning where storage capacity adjusts based on actual VM needs. This dynamic approach allows VMs to access storage on demand while the system monitors consumption to ensure overall pool sustainability, resolving the contradiction between guaranteed performance and storage efficiency.
Data Source
AI summary
Systems and methods described herein facilitate provisioning virtual machines (VMs) in a virtual desktop infrastructure (VDI). The VDI includes a virtual desktop management server (VDMS), a VM, and a plurality of datastores. The VDMS includes a management module that is configured to determine a plurality of usage values that are associated with the datastores. The management module is also configured to determine one or more selection penalty values that are associated with one or more thin-provisioned VMs assigned to one or more of the datastores. Further, the management module calculates a plurality of capacity values for the datastores based at least in part on the determined usage values and the determined penalty values such that each of the capacity values corresponds to a separate datastore. Based at least in part on the capacity values, the management module is configured to assign the VM to one of the datastores.