How to test a gigabit switch? Network hardware testing equipment

As the core connection device of an enterprise network, a vswitch has the primary performance standard to ensure the speed of the enterprise network. At present, enterprises have launched various Ethernet switches. Many parameter performances are marked by manufacturers themselves. These performance parameters differ from the actual situation of switches.

Hardware Network Tester

The manufacturer must not exaggerate the virtual parameters of a product because the industry has the same equipment for product performance testing. Software testing can be used for general household wireless network devices, but hardware network testing equipment is required for large network devices. IXIA 1600 is a well-known hardware device used for testing Ethernet switches. IXIA 1600 is a load generator and analyzer that can be used for performance testing on multiple network devices, devices that can be tested include vswitches, routers, wired and wireless Modem edge and backbone network devices.

The IXIA performance analysis system can accurately and reliably analyze and detect the performance indicators of various devices that build high-speed data communication networks. The IXIA performance analysis system is widely used in all aspects of device development, production, and quality certification, as well as network benchmark, activation, QoS, and prospective and upgrade tests.
Switch Test
The switch test mainly uses the ScripMate software of the IXIA1600 tester to configure and run various indicator tests. ScriptMate specially designed standard automation scripts for RFC 2544 and RFC 2285, we can easily define various parameters based on our own needs, while generating detailed log files and file describing results.
During the test, all ports of IXIA 1600 allow Adaptive Traffic Control and disable traffic control by default. In this test, three typical frames of 64-byte, 512-byte, and 1518-byte length are considered, unless otherwise specified, all tests are performed in full duplex mode. To ensure the reliability and accuracy of the test conditions, each test is repeated three times. The final result is to take the average value of the three tests.
To help readers better understand the overall performance of the vswitch, The IXIA1600 test instrument is used to test nine main indicators related to the performance of the vswitch.
1. throughput is one of the most important indicators for users to select and measure the performance of a vswitch. The throughput determines the maximum rate of frames sent and received without frame drop. During the test, we performed the task at full load. This test is configured as one-to-one ing.
2. Frame loss rate this test determines that the switch should be forwarded under continuous load, but the percentage of frames that cannot be forwarded due to lack of resources. The frame loss rate can reflect the performance of the vswitch during overload, which is useful for the operation of the vswitch in abnormal conditions such as broadcast storm.
3. Back-to-Back this test considers the number of data frames that can be continuously forwarded by the switch without frame loss. The test of this parameter can reflect the size of the data buffer.
4. delay this indicator determines the time when data packets pass through the vswitch. If the latency is FIFO (First in and First Out), it refers to the time interval from when the device receives the First frame to the input port to when the First frame reaches the end of the output port. Initially, the sending rate is set to the rate obtained in the throughput test. frames are sent within the specified interval. A specific frame is set to a time mark frame. The time tag of the frame is recorded at the time of sending and receiving, and the difference between the two is the delay time.
5. Error frame Filtering this test item determines whether the switch can correctly filter frames of certain error types, such as excessively small frames, ultra-large frames, CRC error frames, Fragment, Alignment errors, and Dribble errors, A small frame is a frame smaller than 64 bytes, including 16, 24, 32, and 63 bytes. A large frame is a frame larger than 1518 bytes, including 1519, 2000, 4000, and 8000 bytes. Fragment indicates a frame with a length less than 64 bytes, and CRC indicates a frame checksum error, dribble frames refer to redundant bytes after the correct CRC Check frame. The switch usually corrects the Dribble frames and forwards them to the correct receiving port, alignment combines a CRC error and a dribble error to indicate an incorrect frame whose frame length is not an integer. This test is configured with One-to-multiple ing.
6. backpressure determines whether the vswitch can avoid packet loss when it blocks external data frames from being sent to the congested port. Some vswitches send back-to-source addresses of blocked signals when the sending or receiving buffer overflows. The vswitch uses X traffic control during full dual-work hours for the same purpose. This test sends data to one port through multiple ports to check whether back pressure is supported. If the port is set to half duplex with back pressure, no frame loss or collision is detected. If the port is set to full-duplex and throttling is set, the throttling frame should be detected. If no back pressure is set, the total number of sent frames is not equal to the number of received frames.
7. Head of Line Blocking and HOL congestion on the wire end. This test determines how congested ports affect the forwarding rate of non-congested ports. During the test, port A and port B are used to send data to port C to form A congested port, while port A also sends data to Port D to form A non-congested port. The result shows the number of received frames, the number of collision frames, and the frame drop rate.
8. Full mesh This test is used to determine the total number of frames that the switch can process when all its ports receive data. Each port of the vswitch sends frames to all other ports in a distributed and cyclic manner while receiving data from other ports at a specific speed. We use the full mesh method to test Gigabit backbone switches for more demanding test environments.
9. Part of the mesh this test tests the maximum capacity of the switch in a more rigorous environment, by sending frames in a mesh form from multiple sending ports to multiple receiving ports for testing. We use this test method to test the Gigabit access switch. Each m is corresponding to 10 MB ports, and the remaining MB ports are used for full mesh testing.