Discovery resource utilization

  • Release version: Zurich
  • Updated July 31, 2025
  • 2 minutes to read
    • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Discovery resource utilization

    This document outlines the network bandwidth and CPU resource utilization associated with ServiceNow Discovery operations across various operating systems and device types. It provides detailed metrics on data transfer volumes for different discovery scenarios, helping customers understand the impact of Discovery activities on their infrastructure.

    Show full answer Show less

    Bandwidth Consumption Details

    Network traffic generated by Discovery probes and patterns varies significantly by operating system and device type. The provided data breaks down bandwidth consumption into four segments of data flow:

    • MID Server to Instance
    • Instance to MID Server
    • MID Server to Target device
    • Target device to MID Server

    The total bandwidth consumption for typical discovery runs differs, with Windows Server 2012 and EMC Storage devices showing higher bandwidth usage (up to approximately 6.8 MB), while devices such as Cisco UCS Switch and F5 Load Balancer consume less bandwidth.

    For three-tier applications (UI, application, and data layers), initial discovery consumes significantly more bandwidth (up to 17.9 MB total) compared to subsequent discovery runs (around 3.4 MB), reflecting lower data transfer needs once initial data is collected.

    Discovery bandwidth for pattern-based scans over different operating system types also varies, with Windows Server and Solaris devices generally requiring more data transfer during creation and updates, whereas other devices like HP-UX and L3 Switch show lower consumption.

    Note: The measurements are taken with base operating configurations; actual results may vary based on local system settings.

    CPU Usage Considerations

    CPU consumption during Discovery depends on a combination of factors including operating system, hardware chipset, and workload characteristics. Customers should monitor their specific environments to assess the performance impact of Discovery operations. Identifying unique system builds and observing Discovery’s CPU usage can help in optimizing and balancing resource utilization.

    Practical Implications for ServiceNow Customers

    • Understanding bandwidth requirements helps plan network capacity and avoid performance bottlenecks during Discovery runs.
    • Awareness of higher bandwidth consumption during initial discovery supports scheduling and resource allocation strategies.
    • Monitoring CPU impact enables proactive tuning and ensures Discovery operations do not adversely affect system performance.
    • Variation across device types and operating systems underscores the importance of environment-specific testing and performance analysis.

    Standard transactions on Windows and UNIX generate various amounts of network traffic, depending on what is being discovered.

    These tables show the bandwidth consumption for each data flow segment of a typical discovery using probes and patterns over different operating systems.

    Table 1. Discovery Bandwidth Over Different Operating Systems (in Megabytes)
    Device Type MID > Instance Instance > MID MID > Target Target > MID Total
    Windows 2016 0.104966 0.101271 0.77739 2.364353 3.34798
    Windows 2012 0.126327 0.07928 1.177146 3.70751 5.089804
    Windows 2008 0.141816 0.104674 1.032673 3.594784 4.873947
    Windows 10 0.091466 0.075601 0.642313 2.221103 3.030483
    Linux CentOS 0.164232 0.111376 0.148742 0.690117 1.114467
    Mac OSX 0.103707 0.068302 0.021681 0.461365 0.655055
    HP-UX 0.120358 0.106676 0.042669 0.101149 0.370852
    Solaris 0.130551 0.099414 0.060243 0.346605 0.636813
    Cisco UCS Switch 0.029655 0.027465 0.094918 0.097444 0.249492
    F5 Load Balancer 0.043935 0.03689 0.017179 0.012132 0.110136
    A10 Load Balancer 0.046631 0.032266 0.018313 0.03182 0.12903
    EMC Storage 0.4776 0.373828 1.215954 4.741926 6.809308

    The following table shows the bandwidth comparison between an initial discovery for three-tier applications and for each subsequent discovery. Bandwidth is broken up into the three tiers: UI (Apache), application (Websphere), and data (Oracle). This measures the total data transfer for each discovery run once for a device class.

    Table 2. Discovery Bandwidth for Three Tier Applications (in Megabytes)
    Device Type MID > Instance Instance > MID MID > Target Target > MID Total
    Three-tier application - Initial discovery 0.712829 0.678862 7.084678 9.430181 17.90655
    F5 Load Balancer 0.017179 0.012132
    Apache on Linux 0.540161 1.107108
    Websphere on Linux 0.729403 1.165112
    Oracle on Windows 5.797935 7.145829
    Three-tier application - subsequent discovery 0.150882 0.107409 2.536535 0.560122 3.354948
    F5 load balancer 0.001347 0.012132
    Apache on Linux 0.136366 0.79392
    Websphere on Linux 0.341042 0.11365
    Oracle on Windows 2.05778 0.354948

    This table shows discovery of different OS types using patterns. This measures, in megabytes, the total amount of data created and the total amount of data in subsequent scans for each device.

    Table 3. Discovery Bandwidth for Patterns over Different Operating System Types (in Megabytes)
    Device MID > Instance Instance > MID MID > Target Target > MID Total
    Linux Create 0.39 0.486 0.098 0.273 1.247
    Update 0.382 0.499 0.093 0.264 1.238
    Windows Server Create 0.289 0.316 5.628 8.508 14.741
    Update 0.273 0.306 5.621 8.458 14.658
    Solaris Create 1.222 1.4 0.383 0.917 3.922
    Update 1.24 1.42 .399 .675 3.734
    HP-UX Create 0.176 0.222 0.063 0.13 0.591
    Update 0.178 0.247 0.062 0.128 0.615
    Citrix Netscaler Create 0.424 1.919 0.019 0.042 2.404
    Update 0.355 0.619 0.016 0.041 1.031
    F5 Create 0.087 0.135 0.026 0.047 0.295
    Update 0.132 0.171 0.026 0.047 0.376
    L3 Switch Create 0.172 0.125 0.282 0.478 1.057
    Update 0.178 0.126 0.282 0.479 1.065
    Note:
    These measurements were taken with base operating configurations. Your local system results may vary.

    CPU Usage examples

    Examples from CPU Usage will vary among the matrix of thousands of combinations of Operating Systems, chip sets and specific loads for each system.

    Your mix of these variables will determine your unique level of CPU consumption.

    You can identify unique builds using internal templates and discover them by watching the performance impact, or lack of performance impact that your Discovery tool has on your system.