This week, the Reviewmeister puts a couple of new 10 Gigabit Ethernet switches through their paces - Foundry's FastIron 400 and HP's ProCurve Routing Switch 9300m series.
Since HP buys its 10Gig Ethernet switch from Foundry, let's start with the Foundry switch. In our tests, it topped out at about 8G bit/sec, or about 80% of line rate. Foundry representatives on site for our testing said the switch fabric limited their device to roughly 8G bit/sec, rather than the full 10 Gigabits.
We tested with three different sized packets and in the best case, Foundry moved traffic at 86% of line rate when handling 64-byte frames.
On the other hand, in tests with four interfaces, Foundry's throughput with 256- and 1,518-byte frames was only about 5.5G and 5G bit/sec, respectively.
Delay should be close to nil at 10 Gigabit rates. Consider a hypothetical perfect switch that adds no delay of its own. At 10 Gigabit rates, it would take just 67 nanosec to transmit a 64-byte frame and 1,230 nanosec to transmit a 1,518-byte frame. These numbers are far below the threshold at which the perceived performance of any application would be affected.
In the real world, delays are much higher. With an offered load of 10%, where delay is the result of simple forwarding and no other effect such as queue buildup, we recorded average delay of 4.3 microsec for Foundry's FastIron 400.
Still, Foundry's results were the best of the switches we tested. And that amount of delay is not close to being enough to impact applications. For voice-over-IP or video applications, jitter is even more critical a metric than delay. Our jitter measurements showed that switches with the least delay - from Foundry - also recorded the least amount of jitter.
For the bandwidth-aggregation tests, Foundry's switch came in right up against the 8G bit/sec limit of the switch fabrics. Foundry's results with 256- and 1,518-byte frames were significantly better than in the four-port 10G Ethernet baseline tests.
The switch from Foundry did the best job of keeping average delay and jitter to low levels across all frame lengths. At worst, Foundry's FastIron added average delay of 32.3 microsec with 1,518-byte frames. And while the FastIron's delay is higher than the 7.6 microsec we recorded in the pure 10 Gigabit tests, remember that frames had to cross two chassis and two pairs of interfaces in this configuration, vs. just one chassis and pair of interfaces.
For the full report, go to http://www.nwfusion.com/reviews/2003/020310gbe.html