Gigabit Ethernet (GigE) has rapidly gained prominence and acceptance as the next step in the evolution of corporate networks. Relatively low-cost, high-speed, and interoperable with today's de facto standard, 100 Mbps Fast Ethernet, are just a few of GigE's promises. At the same time these manufactures introduced changes to Gigabit Ethernet Maximum Transmission Unit (MTU). The 1500 byte standard MTU for 10 and 100 Mbps networks has...
been replaced with... no standard at all!! Packets on Gigabit Ethernet can be any size that is supported by network manufacturers. They vary from 1500 bytes to over 16000 bytes.
The benefits of jumbo packets (> 1500 bytes) are significant. Today's network hosts are incapable of accessing the full Gigabit capacity with 1500 byte packets - jumbo packets can more than double accessible bandwidth on most machines. And yet there are some hidden perils. In the transition from 10 to 100 Mbps, issues with duplex mismatch and auto-negotiation became a serious headache on almost every network. In a similar way, MTU conflicts may hamper 100 Mbps to Gigabit transitions, and in some cases affecting performance more than duplex mismatches. But also like duplex mismatch, MTU conflicts are entirely avoidable with some careful planning.
This is the fifth in a technical series of white papers from Apparent Networks examining the Perils of the Network. This series explains network idiosyncrasies and degradations and discusses both the benefits and hidden perils of using jumbo packets on Gigabit Ethernet.
Loki Jorgenson, Ph. D., is a Research Director with Apparent Networks, Inc., and has a Ph.D. in computational physics from McGill University. Dr. Loki Jorgenson has been active in computation, physics and mathematics, visualization, and simulation for over 16 years. He has published in areas as diverse as philosophy, graphics, educational technologies, statistical mechanics, logic and number theory. Loki is an Adjunct Professor of Mathematics at Simon Fraser University, where he was previously the Research Manager of the Center for Experimental and Constructive Mathematics. He has headed research in numerous academic projects from high-performance computing to digital publishing, working closely with private sector partners and government. He has also worked as an e-business consultant with Idea Integration in the San Francisco Bay area. At Apparent Networks, Inc., Dr. Jorgenson is responsible for leading research efforts and for creating practical collaborative efforts with academics and other thought leaders.
Dr. Jorgenson welcomes your comments, suggestions and feedback on this white paper, and is available to answer your questions. Feel free to e-mail him here: firstname.lastname@example.org