Tech Tips Blog

Network Performance Test Endpoints

When it comes down to it, the most important job of a network is to transport data from one end to the other, as quickly as possible.  If this is accomplished, you will rarely get a complaint.  The EtherScope™ nXG provides a couple of apps to measure this throughput and compare it with a predefined threshold.  These are the Network Performance app and the iPerf app.  Both of these require an endpoint to measure throughput.

Let’s take a look at the endpoints that are available for the Network Performance app and the differences between them.  The three available endpoint types are: Software Reflector, Hardware Reflector, and a Peer.

Software Reflector:

This is a software application that may be downloaded from the NetAlly website and loaded on a Windows 7 and above device.  There is no licensing limit on the number of software reflectors that may be downloaded and deployed within a network.  The software reflector provides an easy means of configuring a device to act as an endpoint for the Network Performance app on the EtherScope™ nXG.

Being a reflector, the software will take any packet received, flip the source and destination MAC and IP address, then send that packet back to the source.  On the EtherScope™ nXG, this will show up as a roundtrip test result.  The number of packets sent by the EtherScope™ nXG will be compared with the number received from the reflector.  While it is possible to determine the number of roundtrip packets lost, it is not possible to know in which direction the packet loss occurred.

The software reflector is not recommended for throughput rates above 100mbps.  Above this rate, it is difficult to determine if any packet loss is the result of the software dropping packets or the network dropping packets.  A great use of the software reflector is as an endpoint for doing pre-deployment testing of networks for technologies such as Voice over IP.

Hardware Reflector:

The hardware reflector operates much the same as the software reflector.  The main difference is that the hardware reflector can reflect packets at full line rate.  The hardware reflector is available on the LinkRunner AT 2000 and the LinkRunner G2.

When these devices are placed in reflector mode, they will reflect network performance packets without packet loss.  This ensures that any packet loss observed is the result of the network dropping packets, not the test instrument.  Both of these devices will reflect traffic at up to 1Gbps.  The use of the hardware reflector provides a means to validate the network is capable of transferring up to 1Gbps, full duplex, from end-to-end, without packet loss.

Peer:

Unlike a reflector, the use of a peer allows packet loss to be tracked in both directions.  If packet loss is observed, it can be determined if that loss occurred on the way to the peer or on the way back.  In addition, peers support asymmetrical data rates.  This allows traffic patterns to be created that are reflective on actual application traffic, such as file transfers.

 

Devices that may be used as peers include OneTouch AT 10G, EtherScope™ nXG, and OptiView® XG.  Each of these devices may be put in peer mode and used as an endpoint for the Network Performance app on the EtherScope™ nXG.  All of these devices are capable of supporting 4 concurrent data streams.  This allows one stream to be configured for high throughput and the others for testing lower rate traffic with QoS implemented.  The results of such a test would help validate that QoS is operating properly, end-to-end.

 

Whether you are using a software reflector, hardware reflector, or peer, the performance test is a great way to ensure your network is capable of transporting application traffic from end-to-end without packet loss.

 

OptiView® is a registered trademark of NETSCOUT Systems, Inc.