Oded Trainin, Hadas Yeger, Mark Gustlin

100GE PCS Modeling

IEEE HSSG September 2007

2

How Random is the PCS Data?

The Proposed 100G PCS has the concept of virtual lanes

A 100G stream is scrambled and then distributed and muxed to many lanes

How does the distribution, muxing and skew affect the “randomness” of the streams?

We rely on the scrambling to provide clock transitions for clock recovery, to control baseline wander etc…

Can’t prove or disprove the randomness (at least easily?) So run some simulations

3

How Random is the PCS Data?

Maximum run lengthsMaximum burst of ones or zeros throughout the simulation.

Running disparity(# of ones) – (# of zeros) in a 64,000bits sliding window.

Number of transitions within a windowNo. of transitions in 128 bits sliding window

4

Model Block Diagram - Transmit

Input 100G

Round-Robinof 66b words

between Virtual lanes

VL0 VL1

.

.

.

.

.

.

.VLn

Virtual lanes

Config-urable no. of VLs

Bit mux

Transmit Electrical media -

configurable: - number- latency per

SerDes

Tx GearboxRound-Robin

between electrical

media and between

optical lanes

Optical lanesconfigurable: - number - latency per

lane

*scrambler

Framing Adding

sync bits and

alignment words.

**

****

* - statistics are collected at these points

framing

Scrambler Self synchronous with polynomial X58+X39+1

Random number is picked for seed

Tx EM0

Tx EM9

5

Model Block Diagram – Receive

Rx Gearbox Round-Robin

between optical lanes and

between output electrical media

Receive electrical media -

configurable: - number- latency per

SerDes

Bit demux

VL0VL1.......VLn

Virtual lanes(same no. as input VLs)

-Word delimiter

-Word alignment

-Lane muxing

Output 100G

Verification compare

output with input

*

*

Rx EM0

Rx EM9

6

Scrambler Used

Self synchronous

Polynomial: X58+X39+1 (10GBASE-R)

0 1 38 57in

out

7

Various Input PatternsAll simulation were run with 10 transmit and receive electrical media, 4 optical and 20 Virtual lanes, and with various inputs.Simulations length: 7.5x1010 bits (0.75 second at 100Gbps)Random input data

– Random generated with “Mersenne Twister”.

Idle data input– Repeating 0x1e00000000

All zero inputAll ones inputSquare wave patterns

– SW1: 0b010101010…– SW2: 0b00110011001100…– SW4: 0b0000111100001111…– SW8: 0xFF00FF00…

8

Results: Max Run Length

373736373536343736Scrambler

43

35

38

35

33

40

34

34

43

35

41

37

36

37

39

34

43

35

41

35

38

35

33

40

34

max

4337373740413736Maximum

3231353032333333Rx EM 9

3833323230343233Rx EM 8

3032323132353232Rx EM 7

3133333131333231Rx EM 6

3131353340333336Rx EM 5

3431313431333433Rx EM 4

3133333133323432Rx EM 3

4335323331343431Rx EM 2

3035313132323333Rx EM 1

3537303433413432Rx EM 0

3232353734363232Optical 3

3435333533323634Optical 2

3733343434373335Optical1

3932343333333235Optical 0

3133333133323432Tx EM 9

4335323331343431Tx EM 8

3035313132323333Tx EM 7

3537303433413432Tx EM 6

3231353032333333Tx EM 5

3833323230343233Tx EM 4

3032323132353232Tx EM 3

3133333131333231Tx EM 2

3131353340333336Tx EM 1

3431313431333433Tx EM 0

Sw8Sw4Sw2Sw1All onesAll zeroIdleRandom

9

Results: Max Run Length

Maximum Run Lengths

2527293133353739414345

Rando

m Idle

All zero

All one

s

Sw1

Sw2

Sw4

Sw8

Input Pattern

Max

imum

Run

Len

gth

Tx EM 6Optical 3Rx EM 2ScramblerMax

10

Results: Max Running Disparity

171413861418153017141678122013021352Scrambler1714

1320

1392

1356

1326

1532

1350

1332

1502

1436

1426

1386

1354

1412

1462

1332

1502

1436

1426

1320

1392

1356

1326

1532

1350

max

13861436153017141678139213881412Max

12641296132012441196119212101190Rx EM 9

12781226125412501326139212501232Rx EM 8

11761324135612221250126811941238Rx EM 7

12741290132612361214123012321318Rx EM 6

11581196133611281532118413581328Rx EM 5

12561184120812601350129612281330Rx EM 4

11721252123612481260133211721206Rx EM 3

11521356150212641350117212261252Rx EM 2

12341436118812501268135812301230Rx EM 1

12181246117612601426124413881208Rx EM 0

12261222136012641300129612561386Optical 3

13521354130613001260121612721250Optical 2

12761246125213861274127212421412Optical1

13461330146213601220131012821260Optical 0

11721252123612481260133211721206Tx EM 9

11521356150212641350117212261252Tx EM 8

12341436118812501268135812301230Tx EM 7

12181246117612601426124413881208Tx EM 6

12641296132012441196119212101190Tx EM 5

12781226125412501326139212501232Tx EM 4

11761324135612221250126811941238Tx EM 3

12741290132612361214123012321318Tx EM 2

11581196133611281532118413581328Tx EM 1

12561184120812601350129612281330Tx EM 0

Sw8Sw4Sw2Sw1All onesAll zeroIdleRandom

11

Results: Max Running Disparity

Running Disparity

800900

100011001200130014001500160017001800

Rando

m Idle

All zero

All one

s

Sw1

Sw2

Sw4

Sw8

Input Pattern

Run

ning

Dis

parit

y

Tx EM 4Optical1Rx EM 1ScramblerMax

12

Results: Min Running Disparity

-1674-1430-1616-1410-1492-1234-1674-1478-1400Scrambler-1674

-1390

-1316

-1368

-1532

-1356

-1324

-1386

-1426

-1324

-1372

-1438

-1470

-1322

-1512

-1386

-1426

-1324

-1372

-1390

-1316

-1368

-1532

-1356

-1324min

-1470-1616-1410-1492-1360-1674-1512-1400Min

-1244-1292-1352-1284-1254-1242-1390-1244Rx EM 9

-1226-1218-1316-1244-1278-1230-1204-1272Rx EM 8

-1258-1324-1214-1178-1274-1368-1292-1318Rx EM 7

-1232-1532-1216-1268-1220-1366-1270-1220Rx EM 6

-1196-1210-1226-1322-1216-1356-1260-1252Rx EM 5

-1228-1324-1256-1230-1198-1278-1264-1294Rx EM 4

-1164-1290-1270-1186-1280-1386-1280-1250Rx EM 3

-1230-1172-1164-1426-1262-1234-1248-1224Rx EM 2

-1226-1268-1250-1202-1262-1282-1324-1268Rx EM 1

-1280-1346-1372-1168-1360-1250-1316-1238Rx EM 0

-1228-1438-1242-1316-1274-1304-1268-1278Optical 3

-1470-1316-1274-1286-1284-1262-1264-1310Optical 2

-1252-1322-1310-1272-1234-1280-1262-1288Optical1

-1262-1262-1394-1232-1266-1336-1512-1314Optical 0

-1164-1290-1270-1186-1280-1386-1280-1250Tx EM 9

-1230-1172-1164-1426-1262-1234-1248-1224Tx EM 8

-1226-1268-1250-1202-1262-1282-1324-1268Tx EM 7

-1280-1346-1372-1168-1360-1250-1316-1238Tx EM 6

-1244-1292-1352-1284-1254-1242-1390-1244Tx EM 5

-1226-1218-1316-1244-1278-1230-1204-1272Tx EM 4

-1258-1324-1214-1178-1274-1368-1292-1318Tx EM 3

-1232-1532-1216-1268-1220-1366-1270-1220Tx EM 2

-1196-1210-1226-1322-1216-1356-1260-1252Tx EM 1

-1228-1324-1256-1230-1198-1278-1264-1294Tx EM 0

Sw8Sw4Sw2Sw1All onesAll zeroIdleRandom

13

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106 113 120 127no. of transitions in window

perc

enta

ge

idlesw1randomsw8sw4all oneall zerosw2

Transitions in 128 bits Window – Scrambler Output

14

Transitions in 128 bits Windowoptical and electrical media

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126

no. of transitions in 128b w indow

perc

enta

ge

Electrical

optical

with skew

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121

no. of transitionspe

rcen

tage

electrical

optical

These graphs were very similar

for the various input scenarios.

*

* Skew was added only for input Electrical media.

Average skew is 2.3 bits, maximal skew is 5.6 bits.

15

Stage 2 - Long Simulation, Setup A

Simulations length: 1012 bits, 10 seconds

10 input and output Electrical media, 4 opticalSkew:

None in electrical mediaMax skew in optical: 5 bits (latencies in bit-time: 1, 2, 4, 6)

Input scenarios:Random inputAll ones inputAll ones input, scrambler seed randomized every 6x106 bits (every 60µs, total 185649 times)

16

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125

scrambler

Electrical media

optical media

Random Input (128b window)

1548(-1534)41Max (min)

1514-151039Scrambler

1418-129637Rx EM 9

1466-131636Rx EM 8

1396-135038Rx EM 7

1354-153439Rx EM 6

1314-130835Rx EM 5

1434-132636Rx EM 4

1476-134437Rx EM 3

1412-137235Rx EM 2

1374-138436Rx EM 1

1374-139837Rx EM 0

1402-141841Optical 3

1548-146636Optical 2

1410-144437Optical1

1454-144437Optical 0

1396-135036Tx EM 9

1332-136435Tx EM 8

1312-130835Tx EM 7

1366-141833Tx EM 6

1476-134437Tx EM 5

1348-137235Tx EM 4

1374-138235Tx EM 3

1444-139435Tx EM 2

1418-129638Tx EM 1

1424-144035Tx EM 0

Max running disparity

Min running disparity

Max run length

17

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125

scrambler

Electrical media

optical media

All Ones Input (128b window)

1934(-1460)43Max (min)

1934-128443Scrambler

1400-133439Rx EM 9

1368-137840Rx EM 8

1390-140840Rx EM 7

1318-142837Rx EM 6

1390-130237Rx EM 5

1398-143837Rx EM 4

1494-139438Rx EM 3

1438-134437Rx EM 2

1370-140435Rx EM 1

1398-135039Rx EM 0

1468-136236Optical 3

1482-138440Optical 2

1466-140837Optical1

1502-146040Optical 0

1390-140842Tx EM 9

1336-142237Tx EM 8

1390-129837Tx EM 7

1410-132034Tx EM 6

1494-139437Tx EM 5

1394-132836Tx EM 4

1368-140436Tx EM 3

1428-133435Tx EM 2

1400-133439Tx EM 1

1358-137436Tx EM 0

Max running disparity

Min running disparity

Max run length

18

Stage 3 - Long Simulation, Setup B

Simulations length: 1012 bits, 10 seconds

4 transmit and receive electrical media, 2 opticalPossible future evolution

Skew:None in electrical mediaSkew in optical: 7 bits (latencies in bit-time: 1, 8)

Input scenarios:Random inputAll ones input

19

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126

InEM0

Optic0

OutEM0

Scrambler

ElectricalScrambler

Optic

Random Input

1468(-1602)51Max (min)

1446-160240Scrambler

1414-137242Rx EM 3

1468-142238Rx EM 2

1424-140438Rx EM 1

1400-136836Rx EM 0

1412-142043Optical1

1438-142851Optical 0

1414-137242Tx EM 3

1400-136836Tx EM 2

1424-140438Tx EM 1

1468-142238Tx EM 0

Max running disparity

Min running disparity

Max run length

20

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126

InEM1

Optical1

OutEM2

Scrambler

ElectricalScrambler

Optic

All Ones Input

2008(-1446)51Max (min)

2008-138444Scrambler

1496-144636Rx EM 3

1522-132839Rx EM 2

1494-138637Rx EM 1

1538-140438Rx EM 0

1508-138043Optical1

1624-132851Optical 0

1496-144636Tx EM 3

1538-140438Tx EM 2

1499-138637Tx EM 1

1522-132839Tx EM 0

Max running disparity

Min running disparity

Max run length

21

Conclusion

The multiplexing of the Universal PCS has a negligible effect on the scrambler properties for the various data paths