General Page
Stable performance is observed with an optimum load of 30 datasources that each has 50 monitored items with minimal processing in the flow.
Notices
The information provided in this report is intended for architects, systems programmers, analysts, and programmers wanting to understand the performance characteristics of IBM App Connect for Manufacturing. The data provided will assist you with sizing solutions. Please note that it is assumed that the reader is familiar with the concepts and operation of IBM App Connect for Manufacturing.
The test results shown in the report should not be used for direct comparisons with what may appear to be similar tests in other performance reports. This is because in many cases the hardware, operating system, underlying platform, and prerequisite software used are different. In many of the tests the user logic is minimal, and the results represent the best throughput that can be achieved for that node type. This should be borne in mind when sizing IBM App Connect for Manufacturing.
References to IBM products or programs do not imply that IBM intends to make these available in all countries in which IBM operates. Information contained in this report has not been submitted to any formal IBM test and is distributed ‘as is’. The use of this information and the implementation of any of the techniques is the responsibility of the customer. Much depends on the ability of the customer to evaluate this data and project the results to their operational environment.
Contents
Evaluation Methodology
The focus of the testing is on the data that is processed by the App Connect for Manufacturing OPC-UA-Input node in a non-HA setup.
The client items are configured such that their values change per second, and with an update interval of 1 second, they are sent for processing to the OPC-UA-Input node. The number of datasources (OPC Servers) and the number of client items that are used in the message flows determine the request-response interactions that take place between the OPC Server and the Integration Server.
This test had 30 datasources with 50 client items each, as the load for testing App Connect for Manufacturing. In this case, on an average data for 1500 items is sent by the OPC UA Simulation Servers per second and processed by the client. The test is run for 3 days continuously without stopping the Integration Server.
Note:
- The message flows devised for this test forward the data from "OPC UA Input node" to the Message Queuing Telemetry Transport (MQTT) nodes without any processing.
- Uses OPC Simulation Servers from various vendors.
- Uses Node.js®-based custom simulation servers (by using node-opcua stack).
- The Node.js®-based simulation servers run in the docker containers.
Metrics Captured
The following metrics are captured during the testing.
- CPU usage by using 'nmon' tool.
- Memory usage by using 'nmon' tool.
- JVM memory by using 'VisualVM' tool.
- Sequences in the output data.
Note: Though the values are for a test run of several minutes, they cannot account for potential differences in the reporting on different setups and operating systems.
Test Environment
The tests is conducted with the IBM App Connect for Manufacturing Version 2.0.0.8, IBM App Connect Enterprise Version 11.0.0.13, and the OPC UA Simulation Servers.
Virtual machine with the following system configurations is used to set up the Integration Server.
- Linux: Operating System Ubuntu 16.04 LTS, 4 Core processor @2.2 GHz 8 GB RAM
Separate VMs are used for running the Integration Server and each OPC UA Simulation Server. MQTT is used for capturing the output. The MQTT Server is also set up on an independent virtual machine (VM). This is to enable the Integration Server to utilize the full capacity available and be least impacted by other processing.
Overview
App Connect for Manufacturing configuration
Following App Connect for Manufacturing properties are used.
| Datasource properties | Values |
|---|---|
|
Message Security Mode |
None |
|
Security Policy |
None |
|
User Authentication Type |
Anonymous |
|
Update Interval |
1000 milliseconds |
|
Server Keep Alive |
20000 milliseconds |
|
Client Keep Alive |
20000 milliseconds |
|
Pub Req Queued Count |
5 |
| Client Item properties | Values |
|---|---|
| Sample Rate | 1000 milliseconds |
| Queue Size | Default |
| YAML properties | Values |
|---|---|
| trustCertificate | true |
| discardData | true |
| isHA | false |
Message Flows
The setup included one message flow per datasource (Total 30 message flows), grouped in five applications.
Each message flow had the following nodes.
- OPC-UA-Input - Establishes session with the OPC UA Simulation Server and receives the Publish response from the OPC UA Simulation Server.
- MQTTPublish - Passes the output received from the OPC-UA-Input node to an MQTT topic.
Test Scenario
The following scenario was included for the test.
Run the message flows continuously for a longer duration without any interruption. Set update interval to 1 second (The OPC UA Simulation Server sends notifications every second for each monitored item.
- Verify whether the client can handle 1500 notification messages per second without any data loss.
- Monitor the CPU and memory usage.
Results
The message flows successfully ran for more than 72 hours without any issues. No unusual memory or CPU usage was observed during the test run and the test was stopped.
The following images show the CPU and memory usage.
Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at Copyright and trademark information.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft and Windows are trademarks of Microsoft Corporation in the United States, other countries, or both.
Other company, product, or service names may be trademarks or service marks of others.
Was this topic helpful?
Document Information
Modified date:
04 April 2022
UID
ibm16446649