TechWatch has just published a report on the future of Ethernet, 100G Ethernet and beyond, which explains the background and technical development to the next upgrade of the thirty-year-old family of networking standards, which will be formalised in 2010.
The key thing is that, traditionally, Ethernet has developed in incremental steps of ten. This means that the next step should be from 10G (the current standard) to 100G and indeed, JANET recently undertook a trial of some of the technology that will be used in 100G networking. However, there has been a lack of consensus over whether to move to 100G or a less ambitious target of 40G. The report reviews these debates and explains why the next IEEE standard will be for both 40G and 100G.
As far as institutions are concerned, larger universities and major research centres are likely to be the first to face these questions and there is a lot of money at stake: a typical 224-port switch with 10G on every port is of the order of £250,000 (and of course they’re not bought singly). Newer equipment with 40G or 100G is likely to be more and early adopters often pay over the odds for being at the cutting edge. Taken together with the buying cycles of academia, the message is that network managers have to look ahead to what the demands on the network are likely to be over the next three to five years and plan for the likely levels of traffic in the distribution and core network equipment, allowing for depreciation, in accounting terms, and suitable levels of return on investment.
The TechWatch report explains why there does not appear to be a consensus for a single target and looks at the implications that may have for network managers in HE, particularly in light of JANET’s recent announcement of trials into 100G transmission. It also looks ahead to the development of terabit Ethernet and predicts a period of ‘chaotic’ development before the standards bodies catch up with the new technologies that will inevitably emerge to fill the gap between 100G and 1T Ethernet (the next increment of ten). All this will happen in the context of re-engineering the basic architecture of the Internet and the report argues that layers 1 and 2 of the Internet, which Ethernet handles, will need to be ‘re-virtualised’ to take account of this.