Getting the Word Out: Membership, Dissemination, and Population Protocols

Slide 1

Slide 1 text

M E M B E R S H I P, D I S S E M I N AT I O N & P O P U L AT I O N P R OTO CO LS G E T T I N G T H E W O R D O U T: SEAN CRIBBS SENIOR PRINCIPAL ENGINEER All photos are my own unless attributed.

Slide 2

Slide 2 text

W H Y B U I L D P E E R -TO - P E E R SYST E M S ?

Slide 3

Slide 3 text

No content

Slide 4

Slide 4 text

Work distribution

Slide 5

Slide 5 text

Work distribution parallelism

Slide 6

Slide 6 text

Work distribution parallelism concurrency

Slide 7

Slide 7 text

Work distribution parallelism concurrency independence

Slide 8

Slide 8 text

Work distribution parallelism concurrency independence Fault Tolerance

Slide 9

Slide 9 text

Work distribution parallelism concurrency independence Fault Tolerance detection

Slide 10

Slide 10 text

Work distribution parallelism concurrency independence Fault Tolerance detection recovery

Slide 11

Slide 11 text

Work distribution parallelism concurrency independence Fault Tolerance detection recovery redundancy

Slide 12

Slide 12 text

Work distribution parallelism concurrency independence Fault Tolerance detection recovery redundancy Locality

Slide 13

Slide 13 text

Work distribution parallelism concurrency independence Fault Tolerance detection recovery redundancy Locality regional presence

Slide 14

Slide 14 text

Work distribution parallelism concurrency independence Fault Tolerance detection recovery redundancy Locality regional presence code-to-data

Slide 15

Slide 15 text

W H Y N OT P E E R -TO - P E E R SYST E M S ?

Slide 16

Slide 16 text

W H Y N OT P E E R -TO - P E E R SYST E M S ? N o w y o u ’v e g o t N p ro b l e m s  

Slide 17

Slide 17 text

W H Y N OT P E E R -TO - P E E R SYST E M S ? N o w y o u ’v e g o t N p ro b l e m s   N = 8 p ro b a b l y

Slide 18

Slide 18 text

W H AT A R E W E B U I L D I N G ?

Slide 19

Slide 19 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S Emoji provided free by Emoji One

Slide 20

Slide 20 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T Emoji provided free by Emoji One

Slide 21

Slide 21 text

Slide 22

Slide 22 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 23

Slide 23 text

Slide 24

Slide 24 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 25

Slide 25 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 26

Slide 26 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 27

Slide 27 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 28

Slide 28 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 29

Slide 29 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” “ P L AT F O R M ” E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 30

Slide 30 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” “ P L AT F O R M ” ➡ Work distribution E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 31

Slide 31 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” “ P L AT F O R M ” ➡ Work distribution ➡ Fault-tolerance E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 32

Slide 32 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” “ P L AT F O R M ” ➡ Work distribution ➡ Fault-tolerance ➡ Locality E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S Emoji provided free by Emoji One

Slide 33

Slide 33 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T SYST E M H E A LT H SYST E M H E A LT H A P P L I CAT I O N M E T R I C S A P P L I CAT I O N M E T R I C S AG E N T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G ” “ P L AT F O R M ” ➡ Work distribution ➡ Fault-tolerance ➡ Locality E XT E R N A L S E R V I C E S E XT E R N A L S E R V I C E S ✓ Peer to Peer! Emoji provided free by Emoji One

Slide 34

Slide 34 text

Slide 35

Slide 35 text

E XT E R N A L S E R V I C E S AG E N T A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T AG E N T “ M U LT I -T E N A N T ” “ M U LT I - R E G I O N ” “ H I G H LY- AVA I L A B L E ” “ R E A L-T I M E ” “ ST R E A M I N G” “ P L AT F O R M ” E XT E R N A L S E R V I C E S Emoji provided free by Emoji One How do cluster nodes ﬁnd each other?

Slide 36

Slide 36 text

Slide 37

Slide 37 text

Slide 38

Slide 38 text

Slide 39

Slide 39 text

Slide 40

Slide 40 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T

Slide 41

Slide 41 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T ➡ Cluster membership and discovery

Slide 42

Slide 42 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T ➡ Cluster membership and discovery ➡ Code and conﬁguration dissemination

Slide 43

Slide 43 text

A R G U S O P E RAT I O N A L V I S I B I L I T Y P R OJ EC T ➡ Cluster membership and discovery ➡ Code and conﬁguration dissemination ➡ Relative and convergent time

Slide 44

Slide 44 text

M E M B E R S H I P P R OTO C O L S

Slide 45

Slide 45 text

W H Y N OT Z O O K E E P E R / C O N S U L / E TC D ? J U ST R U B S O M E C O N S E N S U S O N I T

Slide 46

Slide 46 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S

Slide 47

Slide 47 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness

Slide 48

Slide 48 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness ➡ Balance

Slide 49

Slide 49 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness ➡ Balance ➡ Short path-length

Slide 50

Slide 50 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness ➡ Balance ➡ Short path-length ➡ Low clustering

Slide 51

Slide 51 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness ➡ Balance ➡ Short path-length ➡ Low clustering ➡ Scalability

Slide 52

Slide 52 text

M E M B E R S H I P : D E S I RA B L E P R O P E RT I E S ➡ Connectedness ➡ Balance ➡ Short path-length ➡ Low clustering ➡ Scalability ➡ Accuracy

Slide 53

Slide 53 text

M E M B E R S H I P : “ V I E W ” F L AV O R S Full Partial

Slide 54

Slide 54 text

Full

Slide 55

Slide 55 text

Full ✓ Connectedness

Slide 56

Slide 56 text

Full ✓ Connectedness ✓ Short path-length

Slide 57

Slide 57 text

Full ✓ Connectedness ✓ Short path-length - Accuracy

Slide 58

Slide 58 text

Full ✓ Connectedness ✓ Short path-length - Accuracy - Balance

Slide 59

Slide 59 text

Full ✓ Connectedness ✓ Short path-length - Accuracy - Balance ๏ High Clustering

Slide 60

Slide 60 text

Full ✓ Connectedness ✓ Short path-length - Accuracy - Balance ๏ High Clustering ๏ Low Scalability

Slide 61

Slide 61 text

Partial

Slide 62

Slide 62 text

Partial ✓ Low Clustering

Slide 63

Slide 63 text

Partial ✓ Low Clustering ✓ High scalability

Slide 64

Slide 64 text

Partial ✓ Low Clustering ✓ High scalability - Connectedness

Slide 65

Slide 65 text

Partial ✓ Low Clustering ✓ High scalability - Connectedness - Balance

Slide 66

Slide 66 text

Partial ✓ Low Clustering ✓ High scalability - Connectedness - Balance - Path-length

Slide 67

Slide 67 text

Partial ✓ Low Clustering ✓ High scalability - Connectedness - Balance - Path-length - Accuracy

Slide 68

Slide 68 text

S W I M - 2 0 0 2

Slide 69

Slide 69 text

S W I M - 2 0 0 2 Emoji provided free by Emoji One

Slide 70

Slide 70 text

S W I M - 2 0 0 2 Emoji provided free by Emoji One Heartbeat protocols

Slide 71

Slide 71 text

S W I M - 2 0 0 2 ๏ Quadratic load Emoji provided free by Emoji One Heartbeat protocols

Slide 72

Slide 72 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection Emoji provided free by Emoji One Heartbeat protocols

Slide 73

Slide 73 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times Emoji provided free by Emoji One Heartbeat protocols

Slide 74

Slide 74 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times ๏ False positives Emoji provided free by Emoji One Heartbeat protocols

Slide 75

Slide 75 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times ๏ False positives Emoji provided free by Emoji One Heartbeat protocols SWIM solutions

Slide 76

Slide 76 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times ๏ False positives ➡ Separate membership and failure detection Emoji provided free by Emoji One Heartbeat protocols SWIM solutions

Slide 77

Slide 77 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times ๏ False positives ➡ Separate membership and failure detection ➡ Randomized probing Emoji provided free by Emoji One Heartbeat protocols SWIM solutions

Slide 78

Slide 78 text

S W I M - 2 0 0 2 ๏ Quadratic load ๏ Failure detection ๏ Response times ๏ False positives ➡ Separate membership and failure detection ➡ Randomized probing ➡ Piggyback membership on probes Emoji provided free by Emoji One Heartbeat protocols SWIM solutions

Slide 79

Slide 79 text

S W I M - 2 0 0 2

Slide 80

Slide 80 text

S CA M P - 2 0 0 3

Slide 81

Slide 81 text

S CA M P - 2 0 0 3

Slide 82

Slide 82 text

S CA M P - 2 0 0 3 ๏ Full views limit scalability

Slide 83

Slide 83 text

S CA M P - 2 0 0 3 ๏ Full views limit scalability ➡ Flexible partial-view size, asymmetric

Slide 84

Slide 84 text

S CA M P - 2 0 0 3 ๏ Full views limit scalability ➡ Flexible partial-view size, asymmetric ➡ Reactive view management

Slide 85

Slide 85 text

S CA M P - 2 0 0 3 ๏ Full views limit scalability ➡ Flexible partial-view size, asymmetric ➡ Reactive view management ➡ Join (“subscribe”) via random walk

Slide 86

Slide 86 text

S CA M P - 2 0 0 3 ๏ Full views limit scalability ➡ Flexible partial-view size, asymmetric ➡ Reactive view management ➡ Join (“subscribe”) via random walk ➡ Automatic balancing via indirection and leases

Slide 87

Slide 87 text

S CA M P - 2 0 0 3

Slide 88

Slide 88 text

S CA M P - 2 0 0 3

Slide 89

Slide 89 text

CYC LO N - 2 0 0 5

Slide 90

Slide 90 text

CYC LO N - 2 0 0 5

Slide 91

Slide 91 text

CYC LO N - 2 0 0 5 ๏ Random shuﬄing doesn’t create good balance

Slide 92

Slide 92 text

CYC LO N - 2 0 0 5 ๏ Random shuﬄing doesn’t create good balance ➡ Fixed partial-view size, symmetric

Slide 93

Slide 93 text

CYC LO N - 2 0 0 5 ๏ Random shuﬄing doesn’t create good balance ➡ Fixed partial-view size, symmetric ➡ Cyclic view management

Slide 94

Slide 94 text

CYC LO N - 2 0 0 5 ๏ Random shuﬄing doesn’t create good balance ➡ Fixed partial-view size, symmetric ➡ Cyclic view management ➡ Join via random walk

Slide 95

Slide 95 text

CYC LO N - 2 0 0 5

Slide 96

Slide 96 text

CYC LO N - 2 0 0 5

Slide 97

Slide 97 text

P R O B L E M S W I T H S CA M P & CYC LO N • No failure detectors • SCAMP: asymmetric views ⟹ disconnection • SCAMP: unbounded view size ⟹ imbalance

Slide 98

Slide 98 text

H Y PA R V I E W - 2 0 0 7

Slide 99

Slide 99 text

H Y PA R V I E W - 2 0 0 7

Slide 100

Slide 100 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability

Slide 101

Slide 101 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability ๏ High failure rates decrease quality

Slide 102

Slide 102 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability ๏ High failure rates decrease quality

Slide 103

Slide 103 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability ๏ High failure rates decrease quality ➡ TCP for transport and failure detector

Slide 104

Slide 104 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability ๏ High failure rates decrease quality ➡ TCP for transport and failure detector ➡ Small reactive view (“active”)

Slide 105

Slide 105 text

H Y PA R V I E W - 2 0 0 7 ๏ Fanout is related to reliability ๏ High failure rates decrease quality ➡ TCP for transport and failure detector ➡ Small reactive view (“active”) ➡ Larger cyclic view (“passive”)

Slide 106

Slide 106 text

Slide 107

Slide 107 text

H Y PA R V I E W - 2 0 0 7

Slide 108

Slide 108 text

H Y PA R V I E W - 2 0 0 7 A B C D Passive view   maintenance

Slide 109

Slide 109 text

H Y PA R V I E W - 2 0 0 7

Slide 110

Slide 110 text

W E C H O S E H Y PA R V I E W • Only active view maintenance • Passive view maintains full membership (unbounded) • Later: switch to complete passive maintenance

Slide 111

Slide 111 text

D I S S E M I N AT I O N P R OTO C O L S

Slide 112

Slide 112 text

D I S S E M I N AT I O N : D E S I RA B L E P R O P E RT I E S

Slide 113

Slide 113 text

D I S S E M I N AT I O N : D E S I RA B L E P R O P E RT I E S ➡ Reliability

Slide 114

Slide 114 text

D I S S E M I N AT I O N : D E S I RA B L E P R O P E RT I E S ➡ Reliability ➡ Scalability

Slide 115

Slide 115 text

D I S S E M I N AT I O N : D E S I RA B L E P R O P E RT I E S ➡ Reliability ➡ Scalability ➡ Eﬃciency

Slide 116

Slide 116 text

E P I D E M I C B R OA D CAST ( G O S S I P ) !

Slide 117

Slide 117 text

E P I D E M I C B R OA D CAST ( G O S S I P ) ➡ Send to random peers !

Slide 118

Slide 118 text

E P I D E M I C B R OA D CAST ( G O S S I P ) ➡ Send to random peers ➡ Messages rebroadcast by recipients !

Slide 119

Slide 119 text

E P I D E M I C B R OA D CAST ( G O S S I P ) ➡ Send to random peers ➡ Messages rebroadcast by recipients !

Slide 120

Slide 120 text

E P I D E M I C B R OA D CAST ( G O S S I P ) ➡ Send to random peers ➡ Messages rebroadcast by recipients ๏ High redundancy !

Slide 121

Slide 121 text

E P I D E M I C B R OA D CAST ( G O S S I P ) ➡ Send to random peers ➡ Messages rebroadcast by recipients ๏ High redundancy ๏ Low scalability !

Slide 122

Slide 122 text

I N C R E AS E D E F F I C I E N CY W I T H O U T R E D U C I N G D E L I V E RY G UA R A N T E E S , W E N E E D

Slide 123

Slide 123 text

No content

Slide 124

Slide 124 text

P L U M T R E E - 2 0 0 9 ! !

Slide 125

Slide 125 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N ! A B

Slide 126

Slide 126 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N • All nodes start with full “eager” set ! A B

Slide 127

Slide 127 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N • All nodes start with full “eager” set • Broadcast triggers eager-push ! A B

Slide 128

Slide 128 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N • All nodes start with full “eager” set • Broadcast triggers eager-push • Duplicate messages cause “pruning” (move to “lazy”) ! A B

Slide 129

Slide 129 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N • All nodes start with full “eager” set • Broadcast triggers eager-push • Duplicate messages cause “pruning” (move to “lazy”) ! A B

Slide 130

Slide 130 text

P L U M T R E E - 2 0 0 9 CO N ST R U C T I O N • All nodes start with full “eager” set • Broadcast triggers eager-push • Duplicate messages cause “pruning” (move to “lazy”) • Regular broadcasts proceed with new “eager” sets ! A B

Slide 131

Slide 131 text

P L U M T R E E - 2 0 0 9 R E PA I R ! A B

Slide 132

Slide 132 text

P L U M T R E E - 2 0 0 9 R E PA I R • Lazy-push sends “I Have” messages ! A B

Slide 133

Slide 133 text

P L U M T R E E - 2 0 0 9 R E PA I R • Lazy-push sends “I Have” messages • Timeout triggers “grafting” (move to “eager”) ! A B

Slide 134

Slide 134 text

P L U M T R E E - 2 0 0 9 R E PA I R • Lazy-push sends “I Have” messages • Timeout triggers “grafting” (move to “eager”) ! A B

Slide 135

Slide 135 text

P L U M T R E E - 2 0 0 9 R E PA I R • Lazy-push sends “I Have” messages • Timeout triggers “grafting” (move to “eager”) • Lazy-push batched to reduce overhead ! A B

Slide 136

Slide 136 text

W E C H O S E P L U M T R E E

Slide 137

Slide 137 text

W E C H O S E P L U M T R E E • Good tradeoﬀ between reliability and redundancy

Slide 138

Slide 138 text

W E C H O S E P L U M T R E E • Good tradeoﬀ between reliability and redundancy • Optimizes for lowest-latency paths

Slide 139

Slide 139 text

W E C H O S E P L U M T R E E • Good tradeoﬀ between reliability and redundancy • Optimizes for lowest-latency paths • Existing open-source implementations

Slide 140

Slide 140 text

W E C H O S E P L U M T R E E • Good tradeoﬀ between reliability and redundancy • Optimizes for lowest-latency paths • Existing open-source implementations • Excellent ﬁt with HyParView

Slide 141

Slide 141 text

P O P U L AT I O N P R OTO C O L S

Slide 142

Slide 142 text

R A N D O M I Z E D I N T E R AC T I O N S P O P U L AT I O N P R OTO C O L S U S E

Slide 143

Slide 143 text

No content

Slide 144

Slide 144 text

D I ST R I B U T E D M O N OTO N I C C LO C K S J O N M O O R E Vidcap from StrangeLoop 2015: https://youtu.be/YqNGbvFHoKM

Slide 145

Slide 145 text

D M C P R O B L E M S ๏ “Wacky clock mode” ๏ Hierarchy imbalances load ๏ Long-lived partitions ๏ No convergence proof

Slide 146

Slide 146 text

No content

Slide 147

Slide 147 text

A P P LY I N G D M C • Use existing dissemination with DMC • Transmit clocks along with other messages • Use monotonic clocks as a drift-detection mechanism

Slide 148

Slide 148 text

L E S S O N S L E A R N E D

Slide 149

Slide 149 text

Photo by Chris Meiklejohn Vidcap from RICON West: https://youtu.be/s4cCUTPU8GI Photo by Comcast

Slide 150

Slide 150 text

T H A N K YO U ! @ S E A N C R I B B S