Dr David J Ives di231@cam.ac.uk Electrical Engineering Division, University of Cambridge, Cambridge, UK. 11 May 2016 License: CC BY-SA This data is related to the paper "Using 25GbE Client Rates to Access the Gains of Adaptive Bit- and Code- Rate Networking" accepted for publication in the OSA Journal of Optical Communications and Networking. Refer to the published paper for more physical details of the set up. Files: Jafar-CATT16HD100-G0-028-Sum.csv :- raw data summary from simulation code, congestion aware routing, PM-16QAM modulation, hard decision fixed rate FEC, 100 Gbps demands, launch power spectral density 0.028 mW/GHz Jafar-CATTFRHD100-G0-028-Sum.csv :- raw data summary from simulation code, congestion aware routing, Adapted PM-mQAM (square) modulation, hard decision fixed rate FEC, 100 Gbps demands, launch power spectral density 0.028 mW/GHz Jafar-CATTVRHD25-G0-028-Sum.csv :- raw data summary from simulation code, congestion aware routing, Adapted PM-mQAM (square) modulation, hard decision variable rate FEC, 25 Gbps demands, launch power spectral density 0.028 mW/GHz Jafar-CAVRVCD100L25-Sum.csv :- raw data summary from simulation code, congestion aware routing, Adapted PM-mQAM (square) modulation, hard decision variable rate FEC, 100 Gbps demands, inverse multiplexed across 25 Gbps lanes on same route only, launch power spectral density 0.028 mW/GHz FitX.dat :- contains data used in figure 2 CodeRtae.dat :- contains data used in figure 3 BP4.dat :- contains data used in figures 6 and 7. This is the processed results from the raw data summary files above. GNUPlot.plt - contains the script file used to produce the figures 2,3,6 and 7. Format Details: --------------------------------------------------------------------------------------------------------------------- Jafar-CATT16HD100-G0-028-Sum.csv Jafar-CATTFRHD100-G0-028-Sum.csv Jafar-CATTVRHD25-G0-028-Sum.csv Jafar-CAVRVCD100L25-Sum.csv Each file contains the results of 10000 Monte-Carlo runs, each run occupies 4 rows. Col 1 = row label Col 2 = run number Col 3 to end = data for situation after demand 1 to * added "0Tran" data contains half (one direction of bi-directional demands) the number of active transceivers "Alloc" data contains the running total number of accepted demands "Block" data contains the running total number of blocked demands "Histogram" data contains the number of transceivers operating at a given data rate after all the demands applied. Transceiver rates run 0,5,10,15,...,500 Gbps across columns 3 to 104. This data not used in the publication. --------------------------------------------------------------------------------------------------------------------- FitX.dat Contains data from Numerical interated red circles only. The other lines on figure 2 are generated analytically. Col 1 = span length [km] Col 2 = nonlinear interference factor Xm [mW^-2.span^-1] ---------------------------------------------------------------------------------------------------------------------- CodeRate.dat contains 4 groups of data first group: col 1 = linear SNR of channel col 2 to col 5 = spectral efficiency (b/sym) for ideal HD FEC for modulation formats PM-QPSK, PM-16QAM, PM-64QAM and PM-256QAM. group 2,3 and 4: optimum points for 100, 25 and 10 Gbps client data rate granularity col 1 = linear SNR of channel col 2 = information rate Gbps (there is 5% OH allowed for framing) ---------------------------------------------------------------------------------------------------------------------- BP4.dat Contains groups of processed data one each for the four lines/ four raw results summary files. Within each group: col 1 = average (across 10000 runs) total demand accepted [Tbps], initial steps are 0.2 Tbps since each demand is bi-directional 0.1 Tbps. col 2 = average (across 10000 runs) blocking probability. col 3 = average (across 10000 runs) number of active transceivers. col (4,5), (6,7), (8,9) = (average accepted load and average number transceivers) for 0.1, 1 and 10 % blocking probability.