Home > Edge > Manufacturing Edge > Guides > Dell Validated Design for Manufacturing Edge - Design Guide with Litmus > Leveraging analytics for smart manufacturing operations
Several tools can be utilized at various hierarchical levels to track and analyze system needs and performance.
At the cluster level, performance parameters are leveraged to track compute, latency, and more, down to the physical disk.
The vSphere client portal is part of the VMware suite. It provides out-of-the-box charts to view real-time and historical VM performance of:
There are several ways to collect analytics for overall equipment efficiency, predictive maintenance, and identifying faults.
Complex analytics can be developed using a rich set of tools, including REST API, Python, Java Scripting, and database-level analytics tools like PowerBI, Tableau, and others.
Different ISV applications also offer visualization and analytics using the ISV's native capabilities. They can be leveraged to have better synergies with existing processes for those applications.
When leveraging SDP as the common collaboration platform for data services, an SDP Grafana instance can be used for visualization for data persisted on SDP using the available database sinks in SDP (InfluxDB and TimescaleDB). SDP APIs, such as Pravega Search REST API, also offer UI-based dashboard or external tool development capabilities. External APIs and cloud connectivity can also be leveraged for cross-ISV application visualization and analytics.
Windows PowerShell with VMware PowerCLI integration can be used to collect performance metrics including the SCSI controller, CPU and memory consumption of individual VMs, disk latencies, and throughput information.
At the OS level, many tools are available to track performance metrics. These tools and metrics vary based on OS.
In a Windows systems, the built-in Performance Monitor application provides an easy-to-use graphical user interface (GUI) to configure jobs that can collect specified data on demand. These metrics track compute performance, memory use, and storage use, broken down by disk. These jobs can also be exported as a template to quickly configure multiple machines.
Linux systems offer comparable statistics with select packages. The sysstat package offers system performance tools that are comparable to those in Performance Monitor on Windows, including CPU utilization and disk I/O statistics. This package can be installed on both Debian and CentOS Linux systems, though the installation varies between them.
After installing the sysstat package, execute the iostat
command to output a set of real-time system statistics including CPU use, memory use, and disk use.
Several options can customize the statistics displayed. Details on command options and the collected metrics are available on the iostat man page. Execute the iostat -xtk 1 1800 command to:
These metrics can be easily output to a file for tracking and analysis.