Where%20Does%20Circuit%20Switching%20Make%20Sense%20In%20the%20Internet?

1
Where Does Circuit Switching Make Sense In the
Internet?

• Pablo Molinero-Fernández
• High Performance Networking Group
• Stanford University

2
Outline

• Motivation
• Packet switching
• Description and myths
• Circuit switching
• Where it can be useful
• Research TCPSwitch
• Goals, architecture and QoS
• Conclusion

3
How We Think the Internet Is
4
How the Internet Really Is
SONET/SDH
IP routers
5
What Dictates Internets Performance
Processing power
Link speed
6
Fast Links, Slow Routers
Processing Power
Link Speed (Fiber)
Source SPEC95Int Prof. Miller, Stanford Univ.,
2000
7
Fast Links, Slow Routers
Processing Power
Link Speed (Fiber)
2x / 2 years
2x / 7 months
Source SPEC95Int Prof. Miller, Stanford Univ.,
2000
8
Can We Build All Optical Routers?

• Packet switches require buffering
• We cannot buffer light

We need other techniques
9
Why was Internet Packet-Switched in the First
Place?

• Gallager
• Circuit switching is rarely used for data
  networks, ... because of very inefficient use of
  the links
• Tanenbaum
• For high reliability, ... the Internet was to
  be a datagram subnet, so if some lines and
  routers were destroyed, messages could be ...
  rerouted

10
Statistical Multiplexing
A
11
How Was Internet Used in the 70s and 80s?

• Applics email, news, ftp, telnet
• Peer-to-peer network
• Long lasting and bursty flows (telnet, large
  ftps)
• Interactive applications (telnet) could consume
  pkts as they arrived

Many things have changed since then
12
How Is It Different Today?

• Client-server population
• Traffic dominated by short http flows
• Average 5 s, 10 pkts
• TCP flows
• Connection setup dominated by RTT
• Burst of data dominated by data rate
• Nature of user expectations has changed
• telnet each pkt was useful
• Web only useful when all pkts have arrived

13
How the Internet Was Used Then
T0
T1
Host A
R1
R2
R3
Host B
14
How the Internet Is Used Today
Tresponse
Client A
R1
R2
R3
Server B
15
How Current Internet Works
Servers
Clients
Internet
16
An Example
Circuit sw Packet sw
Bandwidth 1 Mbps 10 Kbps
Average latency 50 sec 100 sec
Worst case latency 100 sec 100 sec
17
Myths about BW Efficiency

• Most networks lightly utilized on average
• Backbone links 10-15
• Private lines 3-5
• Ethernet links extremely low
• Links are not congested

BW efficiency not needed
Source A. Odlyzko and K. Coffman, ATT Research
18
Myths about Robustness

• Link/router failure is rare
• 0.5 reroute prob. of TCP flow
• Rerouting circuits is not hard (if there arent
  too many!)
• 50 ms reroute delay for SONET
• vs. over 30 s for OSPF and BGP
• Current products can reroute 1,000s of STS-1
  circuits

Robustness can also be achieved with circuits
19
If we were to start again with what we know
today, could we use circuit switching?
20
This Is Where We Are
SONET/SDH
IP routers
21
Where We Can Go From Here
PS CS

• Big routers replace circuit switches
• MPLS replaces circuit switches

Circuit and packet switching coexist, as today

• MPlambdaS replaces TDM circuit switches
• Pure circuit switched Internet,

22
All Packet Switched Network Is Unlikely
Decision making frequency (for OC-192) Complexity (for OC-192)
Packet switching Per-packet (32 ns) pkts (10,000) conn (100,000)
Circuit switching Per-connection (50 µs) connections (100,000)
23
All Circuit Switched Network Is Unlikely

• Hard to change all end hosts
• Hard to change mentality of OS and application
  developers
• Too much investment in packet switched LANs

24
Circuits Make Sense in the Core

• Where
• electronics cannot keep up with link speeds
• performance is more important than price
• Broad Picture Is Unlikely to Change
• However, provisioning of circuits is primitive

Lets see how we could change the circuit switches
25
What Other People Are Proposing

• Burst switching
• Mega packets or mini circuits with explicit
  tear-down
• Automatic monitoring of traffic
• Queues, utilization,
• Both cases need new provisioning protocol

26
We Want to Propose a Different Approach

• Expose circuits to IP
• Optimize for the common case
• TCP (90-95 of traffic)
• Data (9 out of every 10 pkts)

TCPSwitch
27
TCPSwitch

• Objectives
• Architecture
• QoS
• Results
• Simulation
• Implementation

28
Objectives of TCPSwitch

• Ability to interoperate with
• Normal IP routers
• Unmodified end-hosts
• Optics and off-the-shelf switches
• QoS and best-effort

29
TCPSwitch Exposes Circuit Switching to IP
TCPSwitches
IP routers
30
TCP Creates a Connection

31
Let TCP Leave State Behind

32
Complexity of Switches

• Ingress
• Like a packet-swed router
• Admission control
• Outgoing port creates circuits
• Core
• Regular TDM/Optical switch
• Egress
• Like core, but with packet reassembly

33
Design Issues

• Layer-4 lookup
• Exact match
• Path rerouting
• SYN-packet not seen
• Hard state vs. soft state

34
Design Issues (2)

• Admission control
• PriorityBW request
• blocking prob. vs dropping prob.
• Circuit establishment/tear-down
• In-band vs out-of-band
• Minimum granularity

35
Simplifies QoS

• In PS
• Very complex schemes for isolation, jitter and
  reserved BW
• In CS
• dedicated BW
• Jitter and isolation fall naturally
• Can mix premium customers (QoS) as well as
  best-effort in a single network

36
Simulation Results

• In ns-2
• Paradox verified using TCP and http traffic
• Avg response 50 and 90 improvement, for some
  blocking prob (0-3)
• Some TCP inefficiencies noticeable with very
  short flows
• Work in progress

37
Implementation Results

• Ingress TCPSwitch
• Linux 2.4 router using netfilter (iptables) and
  iproute2
• Module in user space for proof of concept
• Big penalty for crossing kernel-user boundary
• Data switching similar performance as a router
  with a classifier (e.g. policer)
• Now being ported into kernel

38
Conclusion

• Circuit switching is already there
• CS makes sense in the core
• Electronics can keep up with link speed
• Use of Internet fits well with CS
• TCPSwitch
• Exposes circuits to IP
• Optimized for TCP and data