Transcript Tuesday, April 10, 2007 (QoS)

15-441: Computer Networking
Lecture 18: QoS
Thanks to David Anderson and Srini Seshan
Overview
• Why QOS?
• Integrated services
• RSVP
• Differentiated services
Lecture 21: 2006-11-16
2
Motivation
• Internet currently provides one single class
of “best-effort” service
• No assurances about delivery
• Existing applications are elastic
• Tolerate delays and losses
• Can adapt to congestion
• Future “real-time” applications may be
inelastic
Lecture 21: 2006-11-16
3
Why a New Service Model?
• What is the basic objective of network
design?
• Maximize total bandwidth? Minimize latency?
• Maximize user satisfaction – the total utility
given to users
• What does utility vs. bandwidth look like?
• Must be non-decreasing function
• Shape depends on application
Lecture 21: 2006-11-16
4
Utility Curve Shapes
U
Elastic
U
BW
U
Hard real-time
BW
Delay-adaptive
Stay to the right and you
are fine for all curves
BW
Lecture 21: 2006-11-16
5
Utility curve – Elastic traffic
U
Elastic
Bandwidth
Does equal allocation of
bandwidth maximize total utility?
Lecture 21: 2006-11-16
6
Admission Control
• If U(bandwidth) is concave
 elastic applications
• Incremental utility is decreasing
with increasing bandwidth
• Is always advantageous to have
more flows with lower bandwidth
• No need of admission control;
This is why the Internet works!
Lecture 21: 2006-11-16
U
Elastic
BW
7
Utility Curves – Inelastic traffic
U
Delay-adaptive
U
BW
Hard real-time
BW
Does equal allocation of
bandwidth maximize total utility?
Lecture 21: 2006-11-16
8
Inelastic Applications
• Continuous media applications
• Lower and upper limit on acceptable performance.
• BW below which video and audio are not intelligible
• Internet telephones, teleconferencing with high delay
(200 - 300ms) impair human interaction
• Sometimes called “tolerant real-time” since they can
adapt to the performance of the network
• Hard real-time applications
• Require hard limits on performance
• E.g. control applications
Lecture 21: 2006-11-16
9
Admission Control
• If U is convex  inelastic
applications
• U(number of flows) is no
longer monotonically
increasing
• Need admission control to
maximize total utility
• Admission control  deciding
when adding more people would
reduce overall utility
• Basically avoids overload
Lecture 21: 2006-11-16
U
Delay-adaptive
BW
10
Overview
• Why QOS?
• Integrated services
• RSVP
• Differentiated services
Lecture 21: 2006-11-16
11
Components of Integrated Services
1. Type of commitment
What does the network promise?
2. Packet scheduling
How does the network meet promises?
3. Service interface
How does the application describe what it wants?
4. Establishing the guarantee
How is the promise communicated to/from the
network
How is admission of new applications controlled?
Lecture 21: 2006-11-16
12
Type of Commitments
• Guaranteed service
• For hard real-time applications
• Fixed guarantee, network meets commitment if clients
send at agreed-upon rate
• Predicted service
• For delay-adaptive applications
• Two components
• If conditions do not change, commit to current service
• If conditions change, take steps to deliver consistent
performance (help apps minimize playback delay)
• Implicit assumption – network does not change much over time
• Datagram/best effort service
Lecture 21: 2006-11-16
13
Scheduling for Guaranteed Traffic
• Use token bucket filter to characterize traffic
• Described by rate r and bucket depth b
• Use Weighted Fair-Queueing at the routers
• Parekh’s bound for worst case queuing delay = b/r
Lecture 21: 2006-11-16
14
Token Bucket Filter
Tokens enter bucket
at rate r
Operation:
• If bucket fills, tokens are
discarded
• Sending a packet of size P
Bucket depth b:
capacity of bucket
uses P tokens
• If bucket has P tokens,
packet sent at max rate, else
must wait for tokens to
accumulate
Lecture 21: 2006-11-16
15
Token Bucket Operation
Tokens
Tokens
Tokens
Overflow
Packet
Enough tokens 
packet goes through,
tokens removed
Lecture 21: 2006-11-16
Packet
Not enough tokens
 wait for tokens to
accumulate
16
Token Bucket Characteristics
• On the long run, rate is limited to r
• On the short run, a burst of size b can be
sent
• Amount of traffic entering at interval T is
bounded by:
• Traffic = b + r*T
• Information useful to admission algorithm
Lecture 21: 2006-11-16
17
Token Bucket Specs
BW
2
Flow B
Flow A: r = 1 MBps, B=1 byte
1
Flow A
1
2
3
Flow B: r = 1 MBps, B=1MB
Time
Lecture 21: 2006-11-16
18
Guarantee Proven by Parekh
• Given:
• Flow i shaped with token bucket and leaky bucket rate
control (depth b and rate r)
• Network nodes do WFQ
• Cumulative queuing delay Di suffered by flow i has
upper bound
• Di < b/r, (where r may be much larger than average
rate)
• Assumes that r < link speed at any router
• All sources limiting themselves to r will result in no
network queuing
Lecture 21: 2006-11-16
19
Sharing versus Isolation
•
•
•
•
•
Isolation
• Isolates well-behaved from misbehaving sources
Sharing
• Mixing of different sources in a way beneficial to all
FIFO: sharing
• each traffic source impacts other connections directly
• e.g. malicious user can grab extra bandwidth
• the simplest and most common queueing discipline
• averages out the delay across all flows
Priority queues: one-way sharing
• high-priority traffic sources have impact on lower priority traffic only
• has to be combined with admission control and traffic enforcement to avoid
starvation of low-priority traffic
WFQ: two-way isolation
• provides a guaranteed minimum throughput (and maximum delay)
Lecture 21: 2006-11-16
21
Putting It All Together
• Assume 3 types of traffic: guaranteed, predictive,
best-effort
• Scheduling: use WFQ in routers
• Each guaranteed flow gets its own queue
• All predicted service flows and best effort
aggregates in single separate queue
• Predictive traffic classes
• Worst case delay for classes separated by order of magnitude
• When high priority needs extra bandwidth – steals it from lower
class
• Best effort traffic acts as lowest priority class
Lecture 21: 2006-11-16
22
Service Interfaces
• Guaranteed Traffic
• Host specifies rate to network
• Why not bucket size b?
• If delay not good, ask for higher rate
• Predicted Traffic
• Specifies (r, b) token bucket parameters
• Specifies delay D and loss rate L
• Network assigns priority class
• Policing at edges to drop or tag packets
• Needed to provide isolation – why is this not done for
guaranteed traffic?
• WFQ provides this for guaranteed traffic
Lecture 21: 2006-11-16
23
Overview
• Why QOS?
• Integrated services
• RSVP
• Differentiated services
Lecture 21: 2006-11-16
24
Components of Integrated Services
1. Type of commitment
What does the network promise?
2. Packet scheduling
How does the network meet promises?
3. Service interface
How does the application describe what it wants?
4. Establishing the guarantee
How is the promise communicated
How is admission of new applications controlled?
Lecture 21: 2006-11-16
25
Service Interfaces
• Guaranteed Traffic
• Host specifies rate to network
• Why not bucket size b?
• If delay not good, ask for higher rate
• Predicted Traffic
•
•
•
•
Specifies (r, b) token bucket parameters
Specifies delay D and loss rate L
Network assigns priority class
Policing at edges to drop or tag packets
• Needed to provide isolation – why is this not done for
guaranteed traffic?
• WFQ provides this for guaranteed traffic
Lecture 21: 2006-11-16
26
Resource Reservation Protocol
(RSVP)
• Carries resource requests all
the way through the network
• Main goal: establish “state” in
each of the routers so they
“know” how they should treat
flows.
A
C
• State = packet classifier
parameters, bandwidth
reservation, ..
• At each hop consults admission
control and sets up reservation.
Informs requester if failure
Lecture 21: 2006-11-16
B
D
27
RSVP Motivation
• Resource reservation mechanism
for multi-point applications
• E.g., video or voice conference
• Heterogeneous receivers
• Changing membership
C
B
D
A
I
• Use network efficiently
• Minimize reserved bandwidth
• Share reservations between
receivers
• Limit control overhead (scaling).
• Adapt to routing changes
Lecture 21: 2006-11-16
J
H
E
G
F
28
PATH Messages
• PATH messages carry sender’s Tspec
• Token bucket parameters
• Routers note the direction PATH messages
arrived and set up reverse path to sender
• Receivers send RESV messages that follow
reverse path and setup reservations
• If reservation cannot be made, user gets an
error
Lecture 21: 2006-11-16
29
RESV Messages
•
•
•
•
Forwarded via reverse path of PATH
Queuing delay and bandwidth requirements
Source traffic characteristics (from PATH)
Filter specification
• Which transmissions can use the reserved
resources
• Router performs admission control and
reserves resources
• If request rejected, send error message
Lecture 21: 2006-11-16
30
Path and Reservation Messages
Sender 1
PATH
R
Sender 2
PATH
RESV (merged)
RESV
R
Receiver 1
R
R
Reserved bandwidth is maximum of
what downstream receivers can use
Lecture 21: 2006-11-16
RESV
Receiver 2
31
Soft State
• Periodic PATH and RESV msgs refresh
established reservation state
• Path messages may follow new routes
• Old information times out
• Properties
• Adapts to changes routes and sources
• Recovers from failures
• Cleans up state after receivers drop out
Lecture 21: 2006-11-16
32
Overview
• Why QOS?
• Integrated services
• RSVP
• Differentiated services
Lecture 21: 2006-11-16
33
Differentiated Services:
Motivation and Design
• Edge routers do fine grain
enforcement
• Typically slower links at edge
• E.g. mail sorting in post offices
• Label packets with a type field
Classification
and conditioning
• Uses IP TOS bits
• E.g. a priority stamp
• Core routers process packets
based on packet marking and
defined per hop behavior
• More scalable than IntServ
• No per flow state or signaling
Lecture 21: 2006-11-16
34
Expedited Forwarding PHB
User sends within profile & network commits to delivery with
requested profile
• Strong guarantee
• Possible service: providing a virtual wire
• Admitted based on peak rate
• Rate limiting of EF packets at edges only, using token
bucket to shape transmission
• Simple forwarding: classify packet in one of two queues,
use priority
• EF packets are forwarded with minimal delay and loss (up to the
capacity of the router)
Lecture 21: 2006-11-16
35
Expedited Forwarding Traffic Flow
Company A
Packets in premium
flows have bit set
Premium packet flow
restricted to R bytes/sec
internal
router
host
first hop
router
ISP
edge
router
edge
router
Unmarked
packet flow
Lecture 21: 2006-11-16
36
Assured Forwarding PHB
• AF defines 4 classes
• Strong assurance for traffic within profile & allow source to exceed
profile
• Implement services that differ relative to each other (e.g., gold service,
silver service…)
• Admission based on expected capacity usage profiles
• Within each class, there are three drop priorities
• Traffic unlikely to be dropped if user maintains profile
• User and network agree to some traffic profile
• Edges mark packets up to allowed rate as “in-profile” or high
priority
• Other packets are marked with one of 2 lower “out-of-profile”
priorities
• A congested router drops lower priority packets first
• Implemented using clever queue management (RED with In/Out bit)
Lecture 21: 2006-11-16
37
Edge Router Input Functionality
Traffic
Conditioner 1
Arriving
packet
Traffic
Conditioner N
Packet
classifier
Best effort
Forwarding
engine
classify packets based on packet header
Lecture 21: 2006-11-16
38
Traffic Conditioning
Drop on overflow
Packet
input
Wait for
token
Set EF bit
Packet
output
No token
Packet
input
Test if
token
token
Set AF
“in” bit
Lecture 21: 2006-11-16
Packet
output
39
Router Output Processing
What type?
EF
High-priority Q
Packets out
AF
Low-priority Q
with priority drop
AQM (RIO)
Lecture 21: 2006-11-16
40
Edge Router Policing
AF “in” set
Arriving
packet
Is packet
marked?
Token
available?
no
Clear “in” bit
Forwarding
engine
Not marked
EF set
Token
available?
no
Lecture 21: 2006-11-16
Drop packet
41
Comparison
Best-Effort
Diffserv
Intserv
Service
• Connectivity
• No isolation
• No guarantees
• Per aggregation
isolation
• Per aggregation
guarantee
• Per flow isolation
• Per flow guarantee
Service Scope
• End-to-end
• Domain
• End-to-end
Complexity
• No set-up
• Long term setup
• Per flow setup
Scalability
• Highly scalable
• (nodes maintain
only routing state)
• Scalable (edge
• Not scalable (each
routers maintains
router maintains
per aggregate state; per flow state)
core routers per
class state)
Lecture 21: 2006-11-16
42