How we did it

We tested an HP ProCurve 2824 and 2848 Ethernet switches with software image I.07.31.

We used a Spirent Communications Smartbits 6000 chassis load with Terametrics XD 10/100/1000BaseT cards for all the testing. We used Smartflow Version 3.10 for most tests and TeraRouting Test v3.0 software for all the routing table capacity and convergence tests.

We ran four categories of tests - basic functions, performance, redundancy, and management. The following lists the test methodology by category.

To continue reading, register here and become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

We tested an HP ProCurve 2824 and 2848 Ethernet switches with software image I.07.31.

We used a Spirent Communications Smartbits 6000 chassis load with Terametrics XD 10/100/1000BaseT cards for all the testing. We used Smartflow Version 3.10 for most tests and TeraRouting Test v3.0 software for all the routing table capacity and convergence tests.

We ran four categories of tests - basic functions, performance, redundancy, and management. The following lists the test methodology by category.

Basic Function Tests

Reboot Time
The switch reboot time was measured by stopwatch from initiating a reboot until the console user interface becomes active.

Layer 2 Performance

Layer 2 Throughput
Layer 2 throughput was measured by finding the maximum throughput without loss in a full mesh traffic pattern over all ports of the switch with all ports running in full duplex 1000baseT mode. We ran a packet loss test to determine the switch's packet loss rate from 10% to 100% maximum input load in 10% increments.

Layer 2 Packet Latency
Layer 2 latency was measured at the maximum throughput with no loss in a full mesh traffic pattern over all ports of the switch with all ports running in full duplex 1000baseT mode.

Layer 3 Performance

Layer 3 Throughput
Layer 3 throughput was measured by finding the maximum throughput without loss in a full mesh traffic pattern over all ports of the switch with all ports running in full duplex 1000baseT mode. We ran a packet loss test to determine the switch's packet loss rate from 10% to 100% maximum input load in 10% increments.

Layer 3 Packet Latency
Layer 3 latency was measured at the maximum throughput with no loss in a full mesh traffic pattern over all ports of the switch with all ports running in full duplex 1000baseT mode.

Quality of Service

802.1p VLAN Priority
We tested 802.1p VLAN priority functionality using the following procedure:

1. Configure the Ethernet interfaces for VLAN tagging.
2. Configure test equipment to send traffic from ports 1, 2, 4, 5, 7, 8, 10, 11, 13, 14, 16, 17, 19, 20, 22, and 23 to ports 3, 6, 9, 12, 15, 18, 21, and 24 in a unidirectional mesh arrangement.
3. For each source and destination port pair, send 8 individual flows corresponding to all 8 VLAN priority values.
4. Starting with a traffic load of 40% for each input port, increase load by 10% to 100% wirespeed for each input port.
5. Measure packet loss and latency for each VLAN value at each load value. Based on the relative loss and latency results, determine if the switch VLAN priority functionality is working correctly.

Management

Management Interfaces
We assessed the usability and functionality of the command line interface, and the Web interface.

Port Mirroring
We tested port mirroring by the following procedure.

1. Configure the switch to mirror the transmit and receive of port 1 at port 3.
2. Send bidirectional traffic between ports 1 and 2.
3. Verify that all traffic is seen at port 3.

Read more about lans & wans in Network World's LANs & WANs section.