40 Gbs and 100 GbsTechnologies forResearch amp; Education.ppt

All Rights Reserved, ?2007 Fujitsu Network Communications 40 Gb/s and 100 Gb/sTechnologies forResearch Education Networks Tom McDermott Fujitsu July 17, 2007 Agenda Growing Bandwidth Requirements Motivation for 40 Gb/s Motivation for 100 Gb/s Client Interfaces and Network Interfaces Characteristics of optical transmission systems Technical approaches for modulation Some experiments LHC Data Collection LHC currently plans on 10GE clear optical wavelengths for world-wide grid-computing network. When LHC comes on-line, it is predicted to produce 12 to 14 petabytes / year* 14 petabytes per year ? ~ 11.2 * 1016 bits / ? * 107 seconds ~ 3.6 * 109 bits / second Keeps one 10GE busy pretty much full-time forever… But real data collection is usually very bursty in time 40G 100G would offer numerous practical benefits Data set exchange and backup, Load sharing, etc. Motivation for 40 Gb/s Support 40G Packet-Over-SONET (POS) router interfaces. Avoiding inverse-multiplexing or link-aggregation issues. System Capacity Expansion: 4 x 10GE, 4 x OC-192 Existing DWDM Systems support ~ 40 channels in C-band, with 10 Gb/s per wavelength. Typical system has 100 GHz (0.8 nm) spacing between optical channels. Filter bandwidths ~ 65 gHz ( ±0.325 gHz) at half-dB down points 40 Gb/s can ‘fit’ within existing DWDM channel filters and ROADMs if the modulation bandwidth can be constrained. Permits 4x capacity expansion without additional fibers or terminals. Motivation for 100 Gigabit Ethernet Today: Link Aggregation (802.3ag) approach for N x 10 GE. Must not re-order packets within any specific flow. How to identify flows, and force each flow to just one of the multiple 10GE lanes? Today: Equal Cost Multi Path -- A way to (almost) randomly assign each flow to one of the 10GE lanes. Dynamically it breaks: all the big flows fight for the same one lane. Performance degrades to 10GE for milliseconds to seconds. ? Need a better solution. Tomorrow: Use native 100 GE link. No need to ‘