Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ......
Transcript of Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ......
![Page 1: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/1.jpg)
Life in the Fast Lane:
the confluence lens
George Varghese,
Microsoft Research
![Page 2: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/2.jpg)
• I drive fast only when . .
• Only drug I use is . . .
• But I do like to make things run fast
![Page 3: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/3.jpg)
Algorithmics to speed abstractions
Example 1: Virtual Memory:
Abstraction: Illusion of infinite memory
Algorithmics: Paging Algorithms
Example 2: Relational Databases
Abstraction: Operations on Logical tables
Algorithmics: Query Planning
![Page 4: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/4.jpg)
Networking in 1990s
Context: Web exploding, traffic doubling,
address doubling.
Problem: TCP (connected queues) and IP
(datagram) slow, as were routers & servers
Network Algorithmics: techniques to restore
speed of abstractions to that of fiber.
This talk: revisionist history of algorithmics and
the confluence lens
![Page 5: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/5.jpg)
Outline
What is a confluence?
Network Algorithmics viewed from the
lens of confluence
Using confluence in Research
![Page 6: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/6.jpg)
What is a Confluence?
![Page 7: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/7.jpg)
MISSOURI
CONFLUENCE: Where Two Rivers meet
![Page 8: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/8.jpg)
MAIN STREAM
IMPACTING STREAM
NEW STREAM
Inflection Point
Milieu
Change
Transformed Ideas
Confluence Definition for this talk
![Page 9: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/9.jpg)
Realistic Painting
Psychology
Impressionism
Photography
Ideas to Canvas
Thin to thick strokes
Example 1: Impressionism
![Page 10: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/10.jpg)
Algorithms
Probability
R. Algorithms
Crypto
Always to sometimes
Sieve of Eratosthenes to Miller-Rabin
Algorithm TIme on 𝟏𝟎𝟏𝟎𝟎 + 𝟐𝟔𝟕
Miller-Rabin (100 trials) 0.3 seconds
Best Deterministic (AKS) 37 weeks
Example 2: Randomized Algorithms
![Page 11: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/11.jpg)
More Computer Science examples
Distributed Algorithms
Streams: Algorithms, Networks
Inflection Point: Popularity of Internet
Mileu Change: Asynchrony, partial failure
Computational Economics
Streams: Economics, Computer Science
Inflection Point: Internet Auctions
Milieu Change: Large scale, small latency
![Page 12: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/12.jpg)
Why Confluences?
Separate trends from fads
Provide a research theme
Balance desire for beauty and impact
Suggest a new field in making, especially
when the original field has matured
![Page 13: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/13.jpg)
Networking
Learning Theory
Network Learning?
Large network data
Distributed data ?
What concept has changed?
All Interdisciplinary work ≠ confluence
![Page 14: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/14.jpg)
Algorithmics via Confluences
![Page 15: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/15.jpg)
Networking
Architecture
Algorithmics
Cheap Clusters
Machine bus to
Network bus
Net Network
DMA RDMA
Example 1: RDMA [KSL 86]
![Page 16: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/16.jpg)
From RDMA to Fast Servers
Inflection point: Internet heating up (90s)
Fast Buffers (DP 93): Avoid copies without changing protocol 0 copy interfaces
Application Device Channels (DPD 93)
Avoid interrupts VIA standard
Header Prediction (J90) Fast TCP
![Page 17: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/17.jpg)
IP Lookups & Path Compression
1
0
1
0
1
11 1
Benefit: Worst case storage falls from N W to 2 N.
Proof: Adding a new node adds at most 2 trie nodes
![Page 18: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/18.jpg)
Networking
Algorithms
Algorithmics
Traffic, IP v6
Msec to usec
Binary Search On Lengths
P
Prefix 1
Prefix N
O (log N)
Length 0 Length 32
O (log W)
Example 2: IP Lookup [WVTP 97]
![Page 19: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/19.jpg)
Binary Search vs log W prefix match
1* 101*
Length 1 Length 3 Length 2
Day 1: JST, For binary search start in middle
![Page 20: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/20.jpg)
Binary Search vs log W prefix match
1* 101*
Length 1 Length 3 Length 2
10
Day 2: JST, Oh, just add markers
![Page 21: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/21.jpg)
Binary Search vs log W prefix match
1* 101*
Length 1 Length 3 Length 2
10
Day 3, GV, Bug, pre-compute BMP of marker
![Page 22: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/22.jpg)
Crossbars & HOL BLocking
![Page 23: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/23.jpg)
Edge coloring versus PIM
Maximal match in log N steps (AOST93) using randomization (PIM)
Token ring like approach using O(1) steps (M99) Cisco GSR
![Page 24: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/24.jpg)
More Algorithms vs. Algorithmics
Sorting vs Packet Scheduling (SV 96): DRR
avoids sorting, throughput-fair only
Geometry vs ACLs (GM01): Real ACLs have
few regions, decision trees
Bucket Sort vs Timing Wheels (VL97): Empty
bucket overhead OK as OS updates time
![Page 25: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/25.jpg)
Fast Routers common by 2000s
Cisco Cat 6K, GSR, Juniper M40
All the problems (switching, lookups, ACLs,
scheduling) had reasonable hardware
Solutions scaled as link speeds scaled
Would Algorithmics play out by 2000?
![Page 26: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/26.jpg)
Algorithmics
Randomized Algs
Security Algorithmics
Attacks, worms
Within Across packets
Sampling to Sample & Hold
Example 3: Measurement, Security
Randomized algorithms can keep exponentially less space
![Page 27: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/27.jpg)
Heavy Hitters: Sample & Hold
F1 F1 F1 F1 F2
F1 3
F2 1
F1
Uncertainty only at start leads to O(1/M) error vs O (1 /sqrt(M))
First in Gibbons-Mathias 98, with some added twists in EV 02
F3
![Page 28: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/28.jpg)
The NetSift Adventure
Start: Sumeet has idea to automate
signature collection.
Idea: Why not use heavy-hitters on
content hashes to detect worms
Prototype: In a week, Sumeet had his
implementation, detected Kibvu
Realization: NetSift, built a chip -> Cisco.
Transition to Reg Ex obsoleted technology
![Page 29: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/29.jpg)
More streaming networks
Elephant Traps (LWPB07): improves S&H by
evicting low rate flows.
From heavy hitters to flow distribution
(KXSW 04)
More complex security predicates like
Super spreaders (VSGB 05)
![Page 30: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/30.jpg)
Using Confluence in Research
![Page 31: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/31.jpg)
Paris
1860
Monet Renoir
Impressionism
Princeton
1973
Confluence:
Number Theory
with Physics
1. Embrace Collisions
Dyson Montgomery
Rabin MIT, 1975
Miller Randomized
Algorithms
![Page 32: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/32.jpg)
Why Collisions help
Hamming: At first, I ate with the mathematicians . . . I shifted to eating with the physics table
Granovetter 83: Power of Weak Ties. More jobs found from people outside one’s close circle
Outsiders bring new ideas into our closed world.
![Page 33: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/33.jpg)
1. The Procket Collision
NPU pipeline
Crossbar
Result: 2 Port Memories suffice for perfect memory allocation
Source: John Holst of Procket, generalized by Ron & Fan Graham
Memory
![Page 34: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/34.jpg)
Collisions with Events (NPR 14)
Jain-Chiu fairness index
Some preliminary results by Panigrahy et all . .
Income Inequality Networking
![Page 35: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/35.jpg)
Other Networking Confluences
Queuing & Networking (Kleinrock, Lam, Kurose, Towsley)
Economics & Networking (Shenker, Clark)
Network Security (Paxon, Savage, Voelker)
HPC & Networking (Greenberg, Vahdat)
. . . Any others? I must have missed many. Write to me.
![Page 36: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/36.jpg)
Genomics
Computer Systems
?
Cheap sequencing
Fragments mapped
to reference
LZ, SQL SlimGene, GQL
2. Discern Confluences: Genomics
With Christos Kozanitis and Vineet Bafna at UCSD. More work
in Berkeley with Franklin, Haussler, Patterson, Shenker, Stoica,
![Page 37: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/37.jpg)
Picking your confluence
Watch for Trends
Read Trade Rags
Listen to Grapevine
Talk to others (teenagers, kids)
Know your Strengths
Collaborators
Personal skill set
Access to Data (secret weapons)
![Page 38: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/38.jpg)
Sabbatical Join MSR
Peyman
Nick
James
Ratul
Nikolaj
Ming
Example 4: Network Verification
(Victor)
![Page 39: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/39.jpg)
Networking
Prog. Languages
Cloud services
Programs networks
1 Solution to many, SAT to AllSAT
Network Verification as a confluence
Line to rule coverage for testing
![Page 40: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/40.jpg)
Network Verification Opportunities: what are equivalents of static
checks, synthesis, debuggers etc.?
Many groups: Bjorner, Foster, Rexford, Walker, Caesar, Godfrey, McKeown, Millstein, Mahajan, Bjorner, Lam, others?
Confluence: Networks, PL, verification
Data sets: Stanford, Internet 2, Bing, Azure
Invitation: Join the party! Make a difference! MSR is a pretty magical place to do this . . .
![Page 41: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/41.jpg)
3. Seek Coherence in Confluence
Identify recurring themes (principles?)
Move functions in time or space: e.g., pre-computation in prefix search
Relax Specifications: e.g., DRR
Leverage Hardware: e.g., wide words for compressed trie lookups, logic in iSLIP
. . .
Balance innovation with scholarship
![Page 42: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/42.jpg)
Structures to further Coherence
Gather group of PhD students around theme
Organize a workshop
Teach a tutorial
Write a review.
Teach a course
Write a book
![Page 43: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/43.jpg)
Coherence via an Idle Loop
Keep thinking of older problems in
background as one learns new techniques
Synchronize LSPs after partition heals (90s)
Set Difference using IBFs (EGUV 11)
Bridge Learning via sending SYSIDs (90s)
Carousel logging (LMV10)
![Page 44: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/44.jpg)
4. Be contrarian in picking problems
Advice from Towsley, McKeown. My examples:
Need MPLS, route lookups too slow (’94) Fast IP Lookups common today
Earliest deadline scheduling for fairness (‘95) Cheap modification of RR (DRR) suffices
Choose security or performance for firewalls (‘96) Fast packet classification and efficient CAMs
Humans must produce attack signatures (’03) Automated signature extraction.
![Page 45: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/45.jpg)
But balance risk . . .
Analogy from Football: Don’t just throw long
balls, run the football occasionally.
Analogy from Finance: Balance your
portfolio. Buttress your stocks with bonds.
Similarly: keep at least one risky bet but add
safer research. Students need papers!
Confluence Safe work
![Page 46: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/46.jpg)
5. Be congruent -
May the outward man and the inward man be at one.
--- Socrates prayer from Plato’s Phaedrus
Some day you will meet a man who cares for none of
these things. Then you will know how poor you are.
-- Rudyard Kipling in address at McGill University
![Page 47: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/47.jpg)
Ramana Cheenu Cristi Lili Girish
Sumeet Florin Frank Adam Shree
Marcel
Thanks to my students, my fellow confluencers
Rajib Sandeep Christos Terry
Marti
Manmohan
Mahesh
![Page 48: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/48.jpg)
More thanks
Many colleagues but most frequent coauthors:
Subhash Suri: (Algorithmics)
Mike Mitzenmacher (Measurement Algorithmics)
Brad Calder (architecture + networking)
Nick McKeown (network verification)
![Page 49: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/49.jpg)
Algorithmics
Virtualization
Network functions
Moved to Vswitch
Pipelined HW to
Multicore with VMs
Greenberg: Scaling SDN in Public Cloud
Kompella et al: Improving TCP Throughput
More life in the fast lane?
![Page 50: Life in the Fast Lane: the confluence lens · Life in the Fast Lane: the confluence lens ... Maximal match in log N steps (AOST93) using randomization (PIM) Token ring like approach](https://reader034.fdocuments.us/reader034/viewer/2022042122/5e9c4d9402926b5c2f3b856b/html5/thumbnails/50.jpg)
6. Avoid extremes
Influenza, commonly known as "the flu", is an infectious
disease common among mammals. The most common
symptoms are chills and fever.
Confluenza, commonly known as "the conflu", is an infectious
disease unique to researchers. The most common symptoms
is excessive preoccupation with finding confluences in every
aspect of life.
Get your conflu shot today Thank you!
F1