3.1.2 Resource Allocation and Catalogs
The VPC on-demand resource allocation model enables providers to deliver high levels of flexibility in the way resources are allocated through published vApp catalogs in vCloud Director. vApp catalogs further enable providers to publish standard application images and predefined resource profiles that subscribers can customize, based on a given set of application workload requirements.
The following table provides an example of different sizing combinations that can be included in a vApp catalog with the basic service offering.
Note In the following table, vCPU quantity is based on a multiple of 1 GHz. Any quantity of memory or vRAM assigned from
Table 12 is reserved at 75 percent. The provider can give subscribers the capability to select specific quantities of resources, such as vCPU, memory, and storage for a given virtual machine or vApp dynamically, as needed. However, providers must first implement a pricing model commensurate with the range of scale for each resource type.
Table 11. VPC On-Demand Service Offering Catalog Example
Instance Size | vCPU/GHz | Memory (MB) | Storage (GB) | Bandwidth (MBps) | Cost |
Extra Small | 1.0/1 GHz | 500–100,000 | 10–2,000 | Variable | Set by provider |
Small | 1.0/1 GHz | 500–100,000 | 10–2,000 | Variable | Set by provider |
Medium | 2.0/2 GHz | 500–100,000 | 10–2,000 | Variable | Set by provider |
Large | 4.0/4 GHz | 500–100,000 | 10–2,000 | Variable | Set by provider |
Extra-Large | 8.0/8 GHz | 500–100,000 | 10–2,000 | Variable | Set by provider |
The maximum virtual machine instance size is derived from the maximum amount of vCPU and the maximum amount of memory that a physical host has available in the environment. Although the supported ranges for memory and storage in
Table 12 indicate configuration maximums for a vSphere and vCloud Director environment, these ranges differ for different providers, given the variance in hosting architectures and physical infrastructure designs.