Networking and Connectivity Issues
Your ThingWorx solution performs and manages a significant number of transactions against external systems. The devices transmit large amounts of granular data using the network. This data is stored in a database or is processed by additional systems. ThingWorx may finally get this data from several external systems and display it in the mashups in your solution. This means that the performance of the network backing up all these components is critical for your solution.
Any transmission or connection delay impacts the overall performance of your solution. It is recommended that you monitor your entire network infrastructure for any transmission bottlenecks or other latency issues. Generally, the latency between frequently used components, for example, ThingWorx to database should not be more than 20 ms.
Network issues generally appear in your solution in thread information. For example, the transactions that attempt to establish a connection or transmit data appear as slow. VisualVM and Support subsystem stack traces can identify the threads that are stuck performing network IO or connectivity tasks.
If a network issue is apparent in the thread-level, it is recommended that you capture the network packets during periods of slow performance. Analyze the packet to identify the issues.
The network tools required to analyze network connectivity issues is out of the scope of this guide.
Was this helpful?