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
Computer Vision Workflow Enhancement – Data Management with Tiger Surveillance
Overview Solution overview Partner overview Architecture Sizing and scaling guidance Use cases
Baseline and performance
Adding an on-premises remote volume with Surveillance Bridge
Disaster recovery with Surveillance Bridge
Conclusions and references

Baseline and performance

Home > Workload Solutions > Computer Vision > White Papers > Computer Vision Workflow Enhancement – Data Management with Tiger Surveillance > Baseline and performance

Baseline and performance

  • The goal of this testing is to ensure that adding Tiger Surveillance to the platform does not adversely impact the operation of the Milestone Recorders. We gathered resource consumption and performance metrics associated with recording video on the Milestone recording server in the following scenarios and configurations:

    • Baseline, Milestone recorder server without Tiger Surveillance integrated.
    • Disaster Recovery with Tiger Surveillance (vSAN storage and Extended Storage).
    • Disaster Recovery and playback. vSAN storage and Extended Storage while simultaneously with 10% playback running.
    Note: The default number of 70 cameras @ 4 Mb/s were configured on each recorder.

    Resource consumption results

    Table 1. Resource consumption metrics
    Test Type Playback % CPU % Usage Memory % Usage Network MBp/s Disk IOPS
    Baseline 0% 33 25 44 60
    DR - Metadata only 0% 27 23 33 1094
    DR - Full data 0% 80 50 460 633
    DR - Metadata only 10% 40 21 79 477
    DR - Full data 10% 70 46 443 597

    Findings

    • There were no noticeable performance impacts on the baseline operation of Milestone Recorders while integrated with Surveillance Bridge .
    • Resource consumption and performance metrics results functioned as expected. Elevated resource consumption is assumed during data recovery tasks and for full data restoration.
    • Resource consumption during DR meta data recovery is very similar to Baseline metrics with elevated disk IOPS to be expected as meta file stubs are recovered from extended storage.