Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
VDI Design Guide—Virtual Workstation Interactive Collaboration with NVIDIA Omniverse
Introduction Solution Architecture Validation
Validation
Approach
Methodology
Test results
Summary of findings
Endpoint device
Sizing the Solution Backup and Restore Summary References Appendix A: Definition of Performance Metrics

Methodology

Home > Workload Solutions > Virtual Desktop Infrastructure > Design and Implementation Guides > VDI Design Guide—Virtual Workstation Interactive Collaboration with NVIDIA Omniverse > Methodology

Methodology


  • Testing consisted of around 10 test cases based on the two graphics workloads, as shown in the following table. We carried out a minimum of three tests for each workload. The following table shows a sample summary of results. The results shown are an average taken over the duration of each workload. For example, an average latency of 140 ms means the average latency measured over the duration of a workload.

    Note: nVector Knowledge Worker tests are independent of each other so there is no collaboration or dependency between sessions. In contrast, Maya tests involve collaborative dependence between each session (for example, object movement in Session One is reflected in Session Two and so on).

     Table 7.  Sample summary of test results 
    WorkloadSession One-initiated collaborationSession Two-initiated collaboration

    		Average latency (ms)Average frame rate (FPS)Average image quality (SSIM)Average latency (ms)Average frame rate (FPS)Average image quality (SSIM)
    nVector Knowledge Worker140160.9843160.968
    Maya104.780.97116.55140.971