If you're considering the move to 40/100gigabit Ethernet, here are five things to keep in mind:
1. Check the cables. Both 40- and 100-gigabit Ethernet use different fiber and copper cabling and transceivers than previous versions. Simply attaching higher-speed devices to an existing cabling plant won't work. On the plus side, moving away from link aggregation toward single higher-speed links may actually reduce cabling requirements.
2. Verify device distances. In general, higher-speed Ethernet works over the same distances as previous generations, such as the 100-meter limit across short-reach fiber optics. But as with any deployment of new equipment, it's wise to test whether the new devices will operate without CRC errors across their intended distances.
3. Check network management and security device speeds. Network monitoring and security devices that worked just fine at gigabit or 10-gigabit speeds may have trouble keeping up at 40- or 100-gigabit rates. Check with vendors (and conduct your own tests, if necessary) to determine the highest supported lossless rate for traffic monitoring and traffic capture.
4. Be prepared to drink from a fire hose. At 40- and 100-gigabit rates, monitoring equipment must be prepared to capture and analyze staggering amounts of data. For instance, a stream of 1,518-byte frames at 100Gbps line rate will require nearly 750GB of storage per minute. Make sure monitor ports in switches, along with any attached analyzers, can keep up with much higher traffic rates.
5. Plan a parallel deployment. The simplest way to get started with higher-speed technologies is a parallel rollout alongside existing network operations. If everything works well, a transition to the new links should be smooth.
Read more about lan and wan in Network World's LAN & WAN section.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.