Resource management

Users can review the compute (processing) and memory utilization of ArcGIS Velocity directly in the app. There are two ways to view compute and memory utilization in the Velocity app. On the Home page, under the Resource Utilization section, you can view the summary of overall compute and memory consumption. Alternatively, for detailed information, go to the Utilization tab in the main menu and select the Compute or Memory option to open the Compute Utilization or Memory Utilization page.

Resource utilization dashboard

The Home page in the Velocity app displays a dashboard indicating resource utilization. The amount of compute (processing) and memory used by Velocity processes are shown as percentages.

Home page showing resource utilization

The compute and memory gauges appear green if the percentages are under the alerting thresholds. The values appear in shades of orange and red as the percentage used approaches 100 percent. Administrators receive an alert if use exceeds 75 percent.

Users may receive error messages when starting new feeds or analytics if there is insufficient capacity to run that feed or analytic. A graph displays a history of the compute (processing) and memory utilization with selectable intervals of hour, day, and week. The graph can be used to understand the processing load and memory utilization over time.

Turn off and on the Auto-Refresh option when actively starting and stopping feeds and analytics to see the immediate impacts.

Compute utilization

The Compute Utilization page provides information about the feeds and analytics that are actively running, including ownership, start time, and item type information. Each item's title links to the details page for that item in the Velocity app.

The Email notification settings option can be modified to change the frequency of notification emails sent to organization administrative contacts. You can set notifications for compute usage warnings to be send daily, weekly, or monthly. This feature is only available for compute utilization and not for memory utilization.

The Usage chart displays compute metrics as a total compute consumption by Velocity processes versus the remaining compute capacity. The Usage by type chart shows processing usage for feeds, real-time analytics, stream process, the system control plane, and metrics. Processes external to Velocity also using compute resources are shown as ‘Other’.

The Total velocity counter shows the velocity or the processing speed of feature records in kilobytes per second.

The Compute Utilization page also has a Logs tab administrators can click to navigate to view relevant logs such as compute consumption and warnings. Administrators can still monitor compute activity from the Compute Utilization or the Home page.

Feed and real-time analytic items are shown sorted based on the start time and type. Administrators can stop running items listed in the table.

Compute utilization dashboard

Memory utilization

The Memory Utilization page in the Velocity app provides information on the memory used by Velocity processes relative to the amount of memory available on the host machine.

The Usage chart displays the aggregated current memory use of all Velocity processes versus the remaining system usage. The Usage by type chart breaks down memory consumption by specific process categories such as feeds, streams, real-time analytics, and system-level components such as control pane and metrics.

The Total velocity counter shows the velocity or the processing speed of feature records in kilobytes per second.

The Memory Utilization page also has a Logs tab administrators can click to browse and view relevant logs such as warnings on memory consumption as specified thresholds are exceeded. This helps teams stay proactive in managing the memory health of the system. Administrators can still monitor memory use for Velocity processes directly from the Memory Utilization page or the Home page.

Memory utilization dashboard

Summary

Administrators use the information provided on the feeds, the real-time analytics, and the compute and memory utilizations pages to review the running feeds and analytics as well as the relative size of feature layers consuming available memory.

For details on processing and memory capacity in Velocity, refer to compute and memory capacity.