Transcript Integrated Service - National Tsing Hua University

Internet QoS
• Goal of QoS architecture:
To provide some level of predictability and
control beyond the current IP best-effort
service.
• The challenge of these IP QoS technologies is :
to provide differentiated delivery services
for individual flows or aggregate
without breaking the Internet in the process.
1
Why use Quality of Service?
• The Internet is being driven by rich media types,
convergence of computer, communications
broadcast, and telephony industries.
• Today’s Internet provides a best-effort service to all
of its application.
– Do not make any promises about the Quality of Service
(QoS)
– The public Internet does not allow delay-sensitive
multimedia applications (ex:IP telephone) to request any
special treatment.
2
QoS technologies
• Resource reservation (integrated services):
– Allocate resources on a per-flow basis
• A “flow” is equivalent to the five-tuple (transport protocol,
source address&port, destination address and port)
• Prioritization (differentiated services):
– Network traffic is classified and apportioned network
resources according to bandwidth management policy
criteria.
3
Integrated Services
• RFC1633
• Network resources are apportioned
according to an application’s QoS request
The Intserv architecture defines two major
classes of service :
– Guaranteed Quality of Service
– Controlled Load Network Service
4
Key features of Intserv
• Resource reservation
• Call setup
• Traffic characterization and specification of the
desired QoS
• Signaling for call setup
• Per-element call admission
5
Resource reservation
• A router is required to maintain enough state to
know what amounts of its resources (buffers, link
bandwidth ) are already committed to on-going
sessions.
6
Call setup
• A session requiring QoS guarantees must first be able
to reserve sufficient resources at each network router
on its source-to-destination path.
• Call setup process requires the participation of each
router on the path.
– Determine the local resources required by the session
– Consider the amounts of its resources that are already
committed to other on-going session
– Determine whether it has sufficient resources to satisfy the
per-hop QoS requirement of the session at this router
without violating QoS
7
Call setup (cont.)
8
Traffic characterization and specification
of the desired QoS
• Rspec (R for reserved)
– Rspec defines the specific QoS being requested by a
connection
• Tspec (T for traffic)
– Tspec characterizes the traffic the sender will be
sending into the network, or the receiver will be
receiving from the network.
9
Signaling for call setup
• A session’s Tspec and Rspec must be carried to the
routers at which resources will be reserved for the
session.
• RSVP protocol is currently envisioned as the
signaling protocol.
10
Per-element call admission
• While receiving the Tspec and Rspec for a session
requesting a QoS guarantee, it can determine
whether or not it can admit the call.
• The call admission decision depends on:
– The traffic specification
– The requested type of service
– The existing resource commitments already made by
the router to on-going sessions
11
Per-element call admission (cont.)
12
Controlled Load Network Service
• RFC2211
• Providing the client data flow with quality of service closely
approximation the QoS that same flow would receive from an
unloaded network element
• Using capacity (admission) control to assure that this service
is received even when the network element is overloaded.
• Applications:
– Adaptive real-time applications
– These applications perform quite well when the network is
unloaded, but rapidly degrade in performance as the
network becomes more loaded
13
Controlled Load Network Service (cont.)
• Controlled-load service simply prioritizes the
packets in the flow, ensuring that they do not wait
too long in router queues as they cross the network.
14
Guaranteed Quality of Service
• RFC2212
• Providing firm bounds on the queueing delays that
a packet will experience in a router.
• Applications:
– Hard real-time applications
– Audio and video playback applications that are
intolerant of late-arriving packets
15
Guaranteed Quality of Service (cont.)
• Leaky Bucket parameters (r,b)
– r :Token bucket rate
– b :Token bucket size
• Tspec:
– p : Peak data rate
– m :Minimum policed unit
– M :Maximum packet size
• Rspec:
– R: Reserved rate ( R>>r)
– S: slack term
(Signify the difference between the desired delay and
16
the delay obtained by using reservation level R)
Guaranteed Quality of Service (cont.)
• Simple Delay bound : b/R
– Request guarantee transmission rate is R
– The amount of traffic generated over interval t
is bounded by rt+b
– The maximum queueing delay experienced by
any packet will be bound by b/R
17
Introduction of RSVP
• Resource ReSerVation Protocol.
• Allows applications running in hosts to reserve
resources in the Internet for their data flows.
• Used by the routers to forward bandwidth
reservation requests.
• RSVP software must be present in the receivers,
sender, and routers.
18
RSVP in Hosts and Routers
HOST
Application
RSVP
process
ROUTER
RSVP
messages
Policy
Control
Routing
Protocol
process
RSVP
process
RSVP
messages
Policy
Control
Data
Classifier
Packet
Scheduler
Admission
Control
Packet
Scheduler
Classifier
Data
Admission
Control
Data
Packet
Scheduler
Data
19
Introduction of RSVP (cont.)
• Two principle characteristics of RSVP
– It provides reservations for bandwidth in multicast
trees(unicast is handled as a special case).
– It is receiver-oriented.
• RSVP reserves resources for only one direction
data streams.
• RSVP is not a routing protocol
– It does not determine the links in which the reservations
are to be made.
– An RSVP daemon consults the local routing databases
to obtain routes.
20
Introduction of RSVP (cont.)
• RSVP depends on an underlying routing
protocol(unicast or multicast) to determine the
routes for the flows
• RSVP is sometimes referred to as a signaling
protocol that allows hosts to establish and teardown reservations for data flows
21
RSVP: multicast- and receiver-oriented.
22
Heterogeneous receivers
• Sender does not have to know the receiving rates
of all receivers.
• It only needs to know the maximum rate of all its
receivers.
• The sender encodes the video or audio into
multiple layers and sends all the layers up to the
maximum rate into multicast tree.
• The receivers pick out the layers that are
appropriate for their receiving rates.
23
Heterogeneous receivers (cont.)
• In order to not excessively waste bandwidth in the
network’s links, the heterogeneous receivers must
communicate to the network the rates they can
handle.
• RSVP gives foremost attention to the issue of
reserving resources for heterogeneous receivers.
24
RSVP Operation Example
Session
(Ipa,PID,Port)
Path message
Resv message
IGMP(1)
IGMP message
Receiver B
Data
Packet
Session
(Ipa,PID,Port) (4)
Resv (3)
IGMP (1)
path (2)
Sender
Resv(3) Receiver A
Merge
point
Session
(Ipa,PID,Port)
25
A Few Simple Examples
An RSVP example
26
An RSVP video conference example
• Each router receives a reservation message from
each of its downstream links in the multicast tree
and sends only one reservation message into its
upstream link.
27
Call Admission
• Whenever a router receives a new reservation
message, it must first determine if its downstream
links on the multicast tree can accommodate the
reservation.
• This admission test is performed whenever a
router receives a reservation message.
• RSVP does not define the admission test, but it
assumes that the routers perform such a test and
that RSVP can interact with the test.
28
Path Messages
• Path messages are another important RSVP
message type.
• Originate at the senders and flow downstream
towards the receivers.
• The principle purpose of the path messages is to
let the routers know on which links they should
forward the reservation messages.
• The path messages also contain a sender Tspec,
which defines the traffic characteristics of the data
stream that the sender will generate.
• Tspec can be used to prevent over reservation.
29
Reservation Styles
• A reservation message specifies whether merging
of reservations from the same session is
permissible.
• A reservation style also specifies from which
senders in a session the receiver desires to receive
data.
• There are currently three reservation styles
– Wildcard-filter style.
– Fixed-filter style.
– Shared-explicit style.
30
Reservation Styles (cont.)
• Wildcard-Filter Style
– It is telling the network that it wants to receive all flows
from all upstream senders in the session and that its
bandwidth reservation is to be shared among the
senders.
• Fixed-Filter Style
– It specifies a list of senders from which it wants to
receive a data flow along with a single bandwidth
reservation. These reservation are distinct, i.e., they are
not to be shared.
31
Reservation Styles (cont.)
• Shared-Explicit Style
– It specifies a list of senders from which it wants to
receive a data flow along with a single bandwidth
reservation. This reservation is to be shared among all
the senders in the list.
32
Reservation Styles (cont.)
• Shared reservations, created by the wildcard-filter
and the shared-explicit styles, are appropriate for a
multicast session whose sources are unlikely to
transmit simultaneously .
• The fixed-filter reservation, which creates distinct
reservations for the flows from different senders,
is appropriate for video teleconferencing.
33
Examples of Reservation Styles
Sample scenario for RSVP reservation styles
Wildcard filter reservations.
34
fixed filter reservations
35
shared-explicit reservations
Soft State
• The reservation in the routers and hosts are
maintained with soft states.
• Each reservation for bandwidth stored in a router
has an associated timer.
• If a receiver desires to maintain a reservation, it
must periodically refresh the reservation by
sending reservation messages.
• A receiver can also change its reservation by
adjusting its reservation in its stream of refresh
messages.
36
Soft State (cont.)
• The senders must also refresh the path state by
periodically sending path messages.
37
Transport of Reservation Messages
• RSVP messages are sent hop-by-hop directly over
IP, thus the RSVP message is placed in the
information field of the IP datagram.
• If an RSVP path or reservation message is lost, a
replacement refresh message should arrive soon.
38
Insufficient Resource
• Because a reservation request that fails an
admission test may embody a number of requests
merged together, a reservation error must be
reported to all the concerned receivers.
• These reservation errors are reported within
ResvError messages, then receivers can reduce
the amount of resource that they request and try
reserving again.
39
Insufficient Resource (cont.)
• The RSVP standard provides mechanisms to allow
the backtracking of the reservations when
insufficient resource are available, but these
mechanisms add significant complexity.
• Killer-reservation problem
– A receiver requests over and over again a large
reservation, each time getting its reservation rejected
due to lack of sufficient resources.
– Because this large reservation may have been merged
with smaller reservations, smaller reservations can not
to be established.
40
Insufficient Resource (cont.)
• Blockade state
– To solve killer-reservation problem , RSVP uses the
ResvError messages to establish additional state in
routers, called blockade state .
– Blockade state in a router modifies the merging
procedure to omit the offending reservation from the
merge, allowing a smaller request to be forwarded and
established.
• Blockade state adds yet further complexity to the
RSVP protocol and its implementation.
41
Disadvantage of RSVP
• Need more memory to record per flow state
information of each node in network.
• RSVP is lack of scalability.
42
Current Internet Services
Source: Chris Metz
43
QoS Approaches
Source: Chris Metz
44
Why Differentiated Services?
• For business purpose:
– high usage utilization of the resources.
• Simper than IntServ(RSVP):
– not keep per-flow state information in each router.
– decrease memory requirements.
• More efficient core routers:
– limited number of service classes.
– simple packet forwarding.
45
Differentiated Services
• DiffServ is an approach to delivering different levels
of service in a scalable way.
• It migrates work loading to the edges and boundaries
of a DS domain.
• Keep the forwarding simple in the core routers.
• It marks packets according to their service
requirement (DS codepoint). Based on the mark, core
routers apply differentiated per-hop forwarding
behavior (PHB).
46
Differentiated Services (cont.)
• Interior nodes in the DS domain only have to deal
with the small number of traffic aggregates rather
than keeping track of every separate traffic flow that
passes through.
47
DiffServ Architecture
Source: Ben Teitelbaum, QBone Architecture
48
DiffServ Architecture (cont.)
• DS domain:
– A DS domain is a set of DS nodes that are with the same
service provisioning policy and set of PHB groups
implemented on each node.
• DS region:
– A DS region is a set of one or more continuous DS domains.
• DS boundary nodes:
– DS boundary nodes interconnect the DS domain to other DS
or non-DS-capable domains.
49
DiffServ Architecture (cont.)
• DS Interior nodes:
– connect to other DS interior or boundary nodes within the
same domain.
• DS ingress nodes:
– responsible for ensuring the traffic entering the DS domain
conforms to any TCA between it and other domain.
• DS egress nodes:
– Perform traffic conditioning functions on traffic forwarded to a
directly connected peering domain, depending on the details
of the TCA between the two domains.
50
DiffServ Components
•
•
•
•
Classifier.
Traffic Conditioner.
Service Level Agreement (SLA).
Traffic Conditioning Agreement (TCA).
51
Classifier
• Behavior Aggregate(BA) classifier:
– BA classifier uses only the DiffServ codepoint(DSCP) in a
packet’s IP header to determine the logical output stream to
which the packet should be directed.
• Multi-Field(MF) classifier:
– MF classifier classifies packets based on one or more fields
in the packet header.
– A common type of MF classifier is a 5-tuple classifier. (src
addr, dest addr, src port, dest port, IP protocol)
52
Traffic conditioner
• Meter:
– Metering is the function of monitoring the arrival times of
packets on a traffic stream and determining the level of
conformance of each packet to a profile.
– Types of meters:
• Average rate meter.
• Exponential weighted moving average meters.
• Token bucket meters.
53
Traffic conditioner (cont.)
• Marker:
– Marker set the DSCP in a packet header.
– Marker may act on unmarked packets or may remark
previously marked packets.
• Shaper:
– Shaper are used to shape traffic to a certain temporal profile.
• Dropper:
– Droppers simply discard packets with no parameters.
54
SLA
• Service Level Agreement(SLA):
– A service contract between a customer and a service
provider that specifies the forwarding service a customer
should receive.
– A SLA may also specify traffic profiles and actions to traffic
streams which are in- or out-of-profile.
• Static SLA:
– norm at the present time.
– first instantiated at the agreed upon service start date and
may periodically be renegotiated.
55
SLA (cont.)
• Dynamic SLA:
– may change as the traffic load fluctuates.
– dynamic SLAs change without human intervention and thus
require an automated agent and protocol.
• Challenging problems for Dynamic SLA :
– Network providers have to balance frequently changing
loads on different routers within the provider network.
– Customer equipments will have to adapt to dynamic SLAs.
– End user applications have to adapt their behavior during a
session.
56
TCA
• Traffic Conditioning Agreement(TCA) specifies
detailed service parameters for each service level:
–
–
–
–
–
Traffic profiles.
Metering rules.
Marking rules.
Discarding rules.
Shaping rules.
57
Traffic Profiles
• A traffic profile specifies the temporal properties of a
traffic stream selected by a classifier.
• In-profile packets may be allowed to enter the DS
domain without further conditioning.
• Out-of-profile packets may be queued until they are
in-profile(shaped), discarded(policed), marked with a
new codepoint(remarked), or forwarded unchanged
while triggering some accounting procedure.
58
Bandwidth Broker
• act the policy and call admission control manager in
each DS domain.
• keep track of current allocation of marked traffic.
• interpret new requests to mark traffic according to
policies and current allocation.
• parcel out marked traffic allocations and set up edge
routers.
59
Bandwidth Broker Architecture
adjacent BB
adjacent BB
Inter-Domain
Interface
application
server
user/
host
network
operator
User/App
Interface
Data
Repository
edge
routers
Policy Manager
Interface
Network Management
Interface
Intra-Domain
Interface
Routing
Information
edge
routers
60
DS Codepoint
backward compatibility
Best-Effort traffic
TOS (RFC 791)
IP precedence (RFC 1349)
0
1
2
3
4
5
6
DSCP
Pool 1
IP precedence
Best-Effort
Default PHB
Expedited
Forwarding
PHB
7
0
CU
1
2
3
4
Precedence
5
6
TOS
7
0
X
X
X
X
X
0
1
1
1
0
0
0
111
Network control
1
1
0
0
0
0
110
Internetwork control
1
0
1
0
0
0
101
Critical
1
0
0
0
0
0
100
Flash override
0
1
1
0
0
0
011
Flash
0
1
0
0
0
0
010
Immediate
0
0
1
0
0
0
001
Priority
0
0
0
0
0
0
000
Routine
High Priority
Class Selector Codepoint
1
0
Low Drop
0
1
0
Medium Drop
0
1
High Drop
0
1
Assured
Forwarding
PHB
1
1
Low Priority
0
0
0
0
0
0
1
1
0
0
0
0
0
1
0
0
1
1
0
1
0
0
1
0
0
0
1
1
1
0
0
Class 1
Class 2
Class 3
Class 4
Low Drop
1
0
0
0
0
0
1
0
1
0
0
0
Medium Drop
1
0
0
0
1
0
1
0
1
0
1
0
High Drop
1
0
0
1
0
0
1
0
1
1
0
0
61
Per-Hop Behavior
• Per-hop Behavior(PHB)
– is a description of the externally observable forwarding
behavior of a DS node applied to a particular DS behavior
aggregate.
– PHBs may be specified in terms of their resource priority
relative to other PHBs, or their relative observable traffic
characteristics.
– PHBs are implemented in nodes by buffer management and
packet scheduling mechanisms.
62
Assured Forwarding PHB Group
• Reference:
– IETF RFC 2597.
• Description:
– AF PHB group is a means for a provider DS domain to offer
different levels of forwarding assurances for IP packets
received from a customer DS domain.
– Four independent forwarding AF classes and with each AF
class, three levels of drop precedence are defined.
– Packets of class x have smaller forwarding time(delay time)
than class y, if x>y.
63
Assured Forwarding PHB Group (cont.)
• Description:
– A packet with drop precedence p must be forwarded with
higher probability than a packet with drop precedence q, if
p<q.
– An IP packet that belongs to an AF class i and has drop
precedence j within is marked with the AFij.
– A DS node must allocate a configurable, minimum amount of
forwarding resources to each implemented AF class.
– An AF class may also be configurable to receive more
forwarding resources than minimum when excess resources
are available either from other AF classes or from other PHB
groups.
64
Assured Forwarding PHB Group (cont.)
• AF PHB recommend codepoint:
AF1
AF2
AF3
AF4
low
010000
011000
100000
101000
mid
010010
011010
100010
101010
high
010100
011100
100100
101100
65
Expedited Forwarding PHB
• Reference:
– IETF RFC 2598.
• Description:
– The EF PHB can be used to build a low loss, low latency,
low jitter, assured bandwidth, end-to-end service through DS
domains.
– The departure rate of the aggregate’s packets from DS
nodes must equal or exceed a configurable rate.
– The EF traffic receives this rate independent of the intensity
of any other traffic attempting to transit the node.
66
Expedited Forwarding PHB (cont.)
– DSCP: Diffserv codepoint
– CU: currently unused
• EF PHB recommend codepoint:
– 101110
67
Future Work
• Dynamic SLA.
• Inter-domain and intra-domain signaling protocol.
• The architectural framework of Interconnecting with
non-DS domains (Combination of DiffServ and
IntServ).
68
RSVP supported products
• Lucent
– Cajun P550 Routing Switch Family.
• Nortel Networks
– Versalar Switch Router 15000.
• Cisco
–
–
–
–
–
Cisco 7500 Family.
Cisco 7200 Family.
Catalyst 5000.
Cisco 6000 Family.
Cisco 8500 Family.
• Alcatel
– PowerRail routing switches.
69
DiffServ supported products
• Lucent
– PacketStar 6400 IP Switch.
• IBM
– IBM 2210 Nways Multiprotocol Router.
• Alcatel
– PowerRail routing switches.
• Cisco
– Cisco 6000 Family.
70
Example: Cisco Catalyst 6000
Family
• Deliver a new line of highperformance, multilayer switching
solutions for campus networks.
• Coupled with the extensive
network services of Cisco IOS to
support RSVP and COPS.
71
• Cisco IOS (Cisco Internetwork Operating System)
Software:
– Specified features include: priority queuing, policy routing,
and weighted fair queue.
– Using IP precedence field.
– The information to determine bandwidth reservation used in
RSVP:
• Mean data rate.
• The largest amount of data the router will keep in queue.
• Minimum QoS.
72
• The feature of Cisco Release 5.3CSX Supervisor
Software:
–
–
–
–
–
–
–
IP and IP Multicast MLS.
IPX MLS.
Multiprotocol routing.
DiffServ packet classification.
Bandwidth policing.
Traffic Scheduling.
COPS/RSVP supported.
73