A survey of multipath routing for traffic engineering - PowerPoint PPT Presentation

1 / 38
About This Presentation
Title:

A survey of multipath routing for traffic engineering

Description:

Optimization of the operational performance of a network ... f{statics, dynamics, CoS, policy..etc} IP/Optical border router-switch with ... – PowerPoint PPT presentation

Number of Views:883
Avg rating:3.0/5.0
Slides: 39
Provided by: vegaI
Category:

less

Transcript and Presenter's Notes

Title: A survey of multipath routing for traffic engineering


1
A survey of multipath routing for traffic
engineering
  • Gyu Myoung Lee
  • Broadband Network Laboratory
  • Tel) (042) 866-6231

2
Contents
  • Introduction
  • Multipath routing fundamentals
  • Multipath routing algorithms
  • Multipath routing in traditional IP network
  • Multipath routing in MPLS networks
  • Multipath routing in optical Internet
  • Summary and conclusions
  • Future work

3
Introduction
  • A survey of multipath routing for traffic
    engineering
  • Traffic Engineering
  • Optimization of the operational performance of a
    network
  • Including load-balancing, constraint-based
    routing, multi-path routing, fast re-routing,
    protection switching
  • Multipath routing
  • Various multipath routing mechanisms for traffic
    engineering
  • solutions for effectively calculating the
    multipaths
  • ways to minimize delay and increase throughput
  • Apply multipath routing MPLS/GMPLS network

4
Multipath routing fundamentals
  • Multipath routing
  • Aggregating the resources of multiple paths and
    reducing the blocking probabilities in QoS
    oriented network
  • Used for providing Quality of Service (QoS)
  • Focus
  • Optimal network utilization and route stability
    taking into consideration link and node failure
  • Objective
  • To dynamically multiplex the incoming traffic
    onto an optimal route ? guaranteeing the required
    level of service
  • Two of the characteristics that can be used for
    determining a path set
  • Path quantity the number of available paths
    between nodes
  • Path independence

5
The benefits of multipath routing
  • Load balancing
  • To make more use of available network resources
    in order to minimize the risk of traffic
    congestion.
  • Quality of Service
  • Several architectures for implementing QoS
  • IntServ per-flow based, not scalable
  • Router require large amount of memory to store
    routing and reservation state and maintain
    consistency
  • DiffServ
  • Not an efficient and scalable solution when link
    failures occur
  • Allow for aggregation of flows and provide a
    scalable solution
  • Quality is guaranteed for individual flows within
    this aggregate

6
Multipath routing algorithms
7
Multipath routing algorithms - 1
  • Equal Cost Multi-path (ECMP)
  • Routing technique for routing packets along
    multiple paths of equal cost using simple round
    robin distribution
  • OSPF-Optimized Multi Path (OMP) use ECMP
  • Multiple Path Algorithm (MPA)
  • Finds a subset of paths that satisfy a condition
    for loop-freeness
  • Can be implemented as an extension to OSPF

8
Multipath routing algorithms - 2
9
Multipath routing algorithms - 3
10
Multipath routing in traditional IP network
11
Multipath routing in traditional IP network
12
Minimum Delay Routing Using Distributed
Computation - 1
A Minimum Delay Routing Algorithm Using
Distributed Computation (1977) multi-path routing
that guarantees minimum average delays
Extensions of the minimum delay algorithm to
handle topological changes
Using A failsafe distributed routing protocol
(1979) Second derivative algorithms for minimum
delay distributed routing in networks
(1984) second derivatives to increase the
convergence rate of the original minimum delay
algorithm Distributed routing with on-line
marginal delay estimation (1990) An improved
technique for measuring the marginal delays
13
Minimum Delay Routing Using Distributed
Computation - 2
Improve the minimum delay algorithm
The modeling of adaptive routing in
data-communication networks (1977) A failsafe
distributed protocol for minimum delay routing
(1981)
A Simple Approximation to Minimum-Delay Routing
(1999) A simpler algorithm can achieve
near-optimal routing On the impact of aggregation
on the performance of traffic aware routing
(2000) The impact of granularity on network and
routing behavior ? A constrained multipath
traffic engineering scheme for MPLS networks
(2002)
14
Linear-programming (LP) problem
  • A linear or piece-wise linear penalty function
  • To minimize the penalty for all links in the
    network
  • Solving the linear-programming (LP) problem ?
    split ratio

Explicit routing algorithms for Internet traffic
engineering (1999) Minimize the maximum of link
utilization
Internet Traffic Engineering by Optimizing OSPF
Weights (2000) Optimal general routing
Internet Traffic Engineering without Full Mesh
Overlaying (2001) Optimal traffic split using LP
15
Packet splitting methods
  • How packets are split on multiple paths
  • Three main approaches RFC2991 (2000)
  • modulo-N hash
  • hash-threshold (most commonly used)
  • highest random weight
  • Performance of hash-threshold RFC2992 (2000)
  • A Traffic Engineering Approach based on
    Minimum-Delay Routing (2000)
  • treat UDP and TCP packets in a different fashion
  • propose a modified version of hash-threshold
    scheme
  • to generate and insert in the packet a random key
    based on the source-destination IP addresses and
    port numbers (at ingress router)

16
Dynamic multipath routing schemes
Dynamic multi-path routing and how it compares
with other dynamic routing algorithms for high
speed wire area networks (1992) The shortest path
is used under light traffic conditions and
multiple paths are utilized as the shortest path
become congested
Load-sensitive routing for long-lived IP flows
(1999) An adaptive flow-level load control scheme
for multipath forwarding (2001) The enhance
routing scheme separating long-lived and
short-lived flows
Dynamic multipath routing (DMPR) an approach to
improve resource utilization in networks for
real-time traffic (2001) DMRP in networks and
switches carrying connection-oriented traffic
(2001) Improve resource utilization of a network
carrying real-time traffic by re-routing on
going calls through shorter routes
17
An adaptive flow-level load control scheme for
multipath forwarding
  • Adaptive multipath load control method
  • Flow classifier
  • Detect long-lived flows through the flow
    characteristics of the duration and the size
  • Dividing flows into long-lived and short-lived
  • Avoid the congestion from the burst transient
    flows
  • The minimal per-flow states at router
  • Maintained by aggregating flows with the
    destination network prefix

18
QoS routing using multipath
QoS routing via multiple paths using bandwidth
reservation (1998) QoS routing via multiple paths
for time constraint (buffer adjustment at the
receiver)
Performance of fast bandwidth reservation with
multipath routing (1998) Analysis of multi-path
routing (1999) The connection establishment time
for multipath reservation is significantly lowered
Impact of resource reservation on the distributed
multi-path quality of service routing scheme
(2001) combines resource reservation with the
ticket-based distributed multi-path QoS routing
Multipath QoS routing with bandwidth guarantee
(2001) The concept of forwarding with routing
marks ? reduce forwarding complexity
19
Multipath routing in MPLS networks
20
Traffic engineering with MPLS
  • The emergence of MPLS
  • Provides basic mechanisms for facilitating
    traffic engineering
  • Explicit routing allows a particular packet
    stream to follow a pre-determined path
  • Constraint-based routing
  • Avoids placing too many LSPs on any link
  • Multiple LSP setup ? traffic split (using
    multipath routing)
  • Administrative policies in online path
    computation
  • Resource color ? to achieve a degree of desired
    LSP placement
  • Network planning
  • The ability to gather per-LSP statistics ?
    reflect the changing traffic distribution
  • The priority feature
  • Offer different grades of service
  • The ability to re-optimize the path of an LSP
  • Rerouting of a LSP (survivability)

21
Multipath routing in MPLS networks
22
Finding traffic split ratio
23
MATE Multipath AIMD
24
MATE mechanism
  • MATE functions in an ingress node

25
Multipath AIMD
  • Distributed Rate Allocation Multipath AIMD
  • Additive-Increase Multiplicative-Decrease (AIMD)
    feedback algorithms
  • both efficient and fair in allocating traffic to
    network paths
  • Binary feedback from LSPs
  • Each LSP j periodically sends messages to all
    sources containing a binary signal fj 0, 1
    indicating its congestion rate
  • Utilization Bj ? fj 1
  • Utilization lt Bj ? fj 0
  • Source adapt rate using AIMD
  • fj 1 ? multiplicative decrease (0 lt kr 1)
  • fj 0 ? additive increase (ka gt 0)

26
Failure recovery (fault tolerant)
  • Traffic engineering using multiple
    multipoint-to-point LSPs (2000)
  • Traffic engineering method using multiple
    multipoint-to-point LSPs
  • Backup routes are used against failures
  • Fault tolerance and load balancing in QoS
    provisioning with multipath MPLS paths (2001)
  • Approaches for fault tolerance and load balancing
    in QoS provisioning using multiple alternative
    paths
  • Searches for maximally disjoint (i.e., minimally
    overlapped) multiple paths
  • The impact of link/node failures becomes
    significantly reduced
  • The use of multiple paths renders QoS services
    more robust in unreliable network conditions
  • All the computed paths must satisfy given
    multiple QoS constraints

27
Multipath routing in optical Internet
28
Multipath routing in optical Internet
  • Wavelength resource management
  • In the data-driven approach to provisioning, the
    boundary routers use traffic measurements to
    autonomously control the number of lightpaths
  • Adaptive Bandwidth Management (UCL, Nortel)
  • Availability of multiple diverse paths provides
    opportunity to improve QoS for streams by dynamic
    load balancing across those paths according to
    the value of the traffic and path
    characteristics.
  • This would improve the utilisation of the network
  • Adaptive multipath load balancing

29
Abstract architecture for IP over Photonic network
IP Layer
Priority
Photonic/IP adapter
IP Border Router
IP Service Controller
IP Service Controller
IP Border Router
Q
Offered rate meter per path per O/P port
Rate control per path per O/P port
IP/Photonic traffic adapter
Provisioning Request
Provisioning Request
R
Bandwidth allocation
O-UNI
Photonic Layer
Photonic N/w Controller
Photonic N/w Controller
Resource availability
Wavelength allocator
IP/ Photonic adapter
Resource Controller
fPolicies, Request
Number of wavelengths
Q
R
Photonic switch
Photonic switch
O/P port wavelength path un/bind/modify
Wavelengths used per O/P port
Notes 1. The IP service and Photonic Network
controller functions are embedded in the
distributable resource Broker 2.
Message passing and Method invocations are local/
remote accordingly
Q Query R Response
30
Adaptive Multipath Load Balancing proposal for
bandwidth management
Port Allocation mapping Function fstatics, dynam
ics, CoS, policy..etc
Flow ID Hash Function f6-tuple.. etc
O/P 1
Path A
I/P 1
Optical Cloud
Path B
Destination
I/P i
O/P n
Diverse and Disjoint paths
IP/Optical border router-switch with multiple
wavelength channels (or MPLS switches with
multiple LSPs)
31
Summary and conclusions
  • Multipath Routing
  • Effectively used for maximum utilization of
    network resources
  • Aggregating the resources of multiple paths
  • Reducing the blocking probabilities
  • Allowing data transfer at higher rate when
    compared to single path
  • Increase the reliability of delivery
  • Various algorithms
  • Effectively calculating the multipaths
  • Minimize delay and increase throughput

32
Future work
  • Traffic Engineering objectives
  • QoS routing, multi-path routing/traffic
    splitting, load balancing, path protection and
    fast re-route
  • TE can be directly addressed by a
    connection-oriented MPLS/GMPLS based approach.
  • Traffic engineering using multipath in
    GMPLS-based optical network
  • Objective
  • Optical resource optimization (load balancing)
  • Reduce blocking probability
  • Fault tolerance
  • Control plane (Routing Signaling)
  • Traffic aggregation and classification
    (flow-level control)
  • Constrained-based multipath explicit route setup

33
References - 1
  • Multipath routing fundamentals
  • 1 D. Awduche et al, Overview and Principles
    of Internet Traffic Engineering,'' IETF Internet
    Draft draft-ietf-tewg-principles-02.txt,
    Work-in-progress, Jan 2002.
  • 2 H. Suzuki and F. A. Tobagi, Fat bandwidth
    reservation scheme with multi-link and multi-path
    routing in ATM networks,'' In Proceedings of IEEE
    INFOOCOM, 1992.
  • 3 I. Castineyra, N. Chiappa, M. Steenstrup,
    The Nimrod Routing Architecture,'' IETF RFC
    1992, August 1996.
  • 4 P. Narvaez, K. Y. Siu, Efficient Algorithms
    for Multi-Path Link State Routing,'' ISCOM'99,
    Kaohsiung, Taiwan, 1999.
  • 5 S. Vutukury and J.J. Garcia-Luna-Aceves, A
    Simple Approximation to Minimum-Delay Routing,''
    SIGCOMM '99, September, 1999.
  • Multipath routing algorithms
  • 6 C. Hendrick, Routing Information Protocol,
    RFC1058, June 1988.
  • 7 R. Albrightson, J.J. Garcia-Luna-Aceves, and
    J. Boyle, EIGRP-A Fast Routing Protcol Based on
    Distance Vectors, Proc. Networld/Interop. 94,
    May 1994.
  • 8 J. Moy, OSPF version 2, RFC2328, 1988.
  • 9 C. Villamizar, OSPF Optimized Multipath
    (OSPF-OMP),'' Expired Internet Draft, 1999.
    Available http//www.ietf.org/proceedings/99mar/I
    -D/draft-ietf-ospf-omp-02.txt
  • 10 P. Narvaez, K. Y. Siu, Efficient
    Algorithms for Multi-Path Link State Routing,''
    ISCOM'99, Kaohsiung, Taiwan, 1999.
  • 11 Hari Shri Palakurthi, Study of Multipath
    Routing for QoS Provisioning, Oct. 2001
  • 12 S. Vutukury and J.J. Garcia-Luna-Aceves,
    MDVA A Distance-Vector Multipath Routing
    Protocol,'' In Proceedings of the INFOCOM, 2001.
  • 13 Srinivas Vutukury, Multipath routing
    mechanisms for traffic engineering and quality of
    service in the Internet, Ph.D thesis, University
    of California, Santa Cruz, March 2001

34
References - 2
  • Multipath routing in traditional IP network
  • 14 R. G. Gallager, A Minimum Delay Routing
    Algorithm Using Distributed Computation,'' IEEE
    Transactions on Communications, Vol. COM-25, No.
    1, January 1977. pp. 73-85.
  • 15 D. P. Bertsekas, E. M. Gafni, and R. G.
    Gallager, Second derivative algorithms for
    minimum delay distributed routing in networks,''
    IEEE Transactions on Communications, vol. 32,
    no. 8, pp. 911-919, August 1984.
  • 16 C. G. Cassandras, M. V. Abidi and D.
    Towsley, Distributed routing with on-line
    marginal delay estimation,'' IEEE Transactions on
    Communications, vol. 38, no. 3, pp. 348-359,
    March 1990.
  • 17 A. Segall, The modeling of adaptive
    routing in data-communication networks,'' IEEE
    Transactions on Communications, vol. 25, no. 1,
    pp. 85-95, January 1977.
  • 18 A. Segall and M. Sidi, A failsafe
    distributed protocol for minimum delay routing,''
    IEEE Transactions on Communications, vol. 29,
    no.5, pp. 689-695, May 1981.
  • 19 S. Vutukury and J.J. Garcia-Luna-Aceves, A
    Simple Approximation to Minimum-Delay Routing,''
    SIGCOMM '99, September, 1999.
  • 20 A. Sridharan, S. Bhattacharyya, C. Diot, R.
    Guerin, J. Jetcheva and N. Taft. On the impact
    of aggregation on the performance of traffic
    aware routing,'' Technical Report, University of
    Pennsylvania, Philadelphia, PA, June 2000.
  • 21 P. M. Merlin and A. Segall, A failsafe
    distributed routing protocol,'' IEEE Transactions
    on Communications, vol. 27, no. 9, pp. 1280-1287,
    September 1979.
  • 22 Y. Wang, and Z. Wang, Explicit routing
    algorithms for Internet traffic engineering,
    ICCCN99
  • 23 B. Fortz, M. Thorup, Internet Traffic
    Engineering by Optimizing OSPF Weights, in
    Proceedings of the INFOCOM 2000, pp. 519-528,
    2000.
  • 24 Z. Wang, Y. Wang, L. Zhang, Internet
    Traffic Engineering without Full Mesh
    Overlaying,'' INFOCOM'01, April 2001.
  • 25 D. Thaler, C. Hopps, Multipath Issues in
    Unicast and Multicast Next-Hop Selection,'' IETF
    RFC 2991, 2000.
  • 26 C. Hopps, Analysis of an Equal-Cost
    Multi-Path Algorithm,'' IETF RFC 2992, 2000.

35
References - 3
  • 27 S. Vutukury and J.J. Garcia-Luna-Aceves,
    A Traffic Engineering Approach based on
    Minimum-Delay Routing,'' Proceedings of IEEE
    ICCCN 2000, Las Vegas, Nevada, USA, October 2000.
  • 28 S. Bahk, M. Zarki, Dynamic multi-path
    routing and how it compares with other dynamic
    routing algorithms for high speed wire area
    networks, ACM computer communication review,
    vol.22, no.4, pp. 54-64, Oct. 1992
  • 29 A. Shaikh, J. Rexford, and K.G. Shin,
    Load-sensitive routing for long-lived IP flows,
    SIGCOMM99, 1999
  • 30 Youngseok Lee and Yanghee Choi, An adaptive
    flow-level load control scheme for multipath
    forwarding, Springer-Verlag, pp 771-779, July
    2001.
  • 31 De, S. Das, S.K., Dynamic multipath routing
    (DMPR) an approach to improve resource
    utilization in networks for real-time traffic,
    Modeling, Analysis and Simulation of Computer and
    Telecommunication Systems, 2001. pp 23-30, 2001
  • 32 De, S. Das, S.K. Tonguz, O, Dynamic
    multipath routing in networks and switches
    carrying connection-oriented traffic, ICC 2001,
    vol. 10, 2001
  • 33 N.S.V. Rao, and S.G. Bastell, QoS routing
    via multiple paths using bandwidth reservation,
    INFOCOM98, 1998
  • 34 T. H. Cheng, J. Chen, Performance of fast
    bandwidth reservation with multipath routing,
    IEEE Proc of Commun. Vol. 145, No. 2, April.
    1998.
  • 35 I. Cidon, R. Rom, and Y. Shavitt, Analysis
    of multi-path routing, IEEE/ACM transaction on
    Networking , vol.7, no 6, pp.885-896, Dec. 1999
  • 36 Yuan Zhong Xin Yuan, Impact of resource
    reservation on the distributed multi-path quality
    of service routing scheme, IWQOS. 2000
  • 37 Yun-Wen Chen Ren-Hung Hwang Ying-Dar Lin,
    Multipath QoS routing with bandwidth guarantee,
    GLOBECOM '01, vol. 4, 2001

36
References - 4
  • Multipath routing in MPLS networks
  • 38 E. C. Rosen, A. Viswanathan, and R. Callon,
    Multiprotocol label switching architecture,
    IETF RFC 3031, Jan. 2001.
  • 39 D. Awduche et al, Requirements for Traffic
    Engineering Over MPLS, IETF RFC 2702, Sept 1999.
  • 40 X. Xiao, A. Hannan, B. Bailey, L. Ni,
    "Traffic Engineering with MPLS in the Internet,"
    IEEE Network magazine, Mar. 2000.
  • 41 C. Liljenstolpe, Offline Traffic
    Engineering in a Large ISP Setting, Internet
    Draft, draft-liljenstolpe-tewg-cwbcp-01.txt,
    2002.
  • 42 V. Sharma et. al., "Framework foyr MPLS
    Based Recovery", Internet Draft,
    draft-ietf-mpls-recovery-frmwrk-06.txt, July
    2002.
  • 43 P. Pan et al, Fast Reroute Techniques in
    RSVP-TE, Internet Draft, draft-pan-rsvp-fastrerou
    te-00.txt, 2002.
  • 44 V. Springer, C. Pierantozzi, and J. Boyle,
    Level3 MPLS Protocol Architecture, expired
    Internet Draft available at, http//www.watersprin
    gs.org/links/mlr/id/draft-springer-te-level3bcp-00
    .txt, 2000.
  • 45 F. Le Faucheur et al, MPLS Support of
    Differentiated Services, RFC3270, May 2002.
  • 46 E. Dinan, D.O.Awduche, and B.Jabbari,
    Analytical framework for dynamic traffic
    partitioning in MPLS networks, ICC2000.
  • 47 Yongho Seok, et al, Dynamic constrained
    multipath routing for MPLS networks, Computer
    Communications and Networks 2001, pp. 348-353,
    Oct. 2001
  • 48 Yongseok Lee, et al, A constrained
    multipath traffic engineering scheme for MPLS
    networks, ICC 2002, April. 2002.
  • 49 A. Elwalid, C.Jin, S. Low, and I. Widjaja,
    MATE MPLS Adaptive Traffic Engineering,
    INFOCOM2001
  • 50 Jianping Wang, Stephen Patek, Haiyong Wang,
    and Jörg Liebeherr, Traffic Engineering with
    AIMD in MPLS Networks, LNCS, May 2002.
  • 51 F. Bonomi and W. Fendick. The Rate-Based
    Flow Control Framework for the Available Bit Rate
    ATM Service. IEEE Network, 9(2)2539,
    March/April 1995.

37
References - 5
  • 52 D. Chiu and R. Jain. Analysis of the
    Increase and Decrease Algorithms for Congestion
    Avoidance in Computer Networks. Computer Networks
    and ISDN Systems, 17114, 1989.
  • 53 R. Jain. Congestion control and traffic
    management in ATM networks Recent advances and a
    survey. Computer Networks and ISDN Systems,
    28(13)17231738, October 1996.
  • 54 V. Jacobson. Congestion avoidance and
    control. In Proceedings of ACM Sigcomm
    88,August, 1988, pages 314329, 1988.
  • 55 R. Jain and K. K. Ramakrishnan. Congestion
    avoidance in computer networks with a
    connectionless network layer Concepts, goals and
    methodology. Proceedings of the
    ComputerNetworking Symposium IEEE Washington,
    DC, pages 134143, 1988.
  • 56 K. K. Ramakrishnan and R. Jain. A Binary
    Feedback Scheme for Congestion Avoidance in
    Computer Networks. ACM Transactions on Computer
    Systems, 8(2)158181, 1990.
  • 57 H. Saito, Y. Miyao, and M. Yoshida, Traffic
    engineering using multiple multipoint-to-point
    LSPs, INFOCOM2000
  • 58 Scott Seongwook Lee, and Mario Gerla, Fault
    tolerance and load balancing in QoS provisioning
    with multipath MPLS paths, Proc of 9th-IWQOS
    (Germany), June 2001
  • Multipath routing in optical Internet
  • 59 Bala Rajagopalan, et al., IP over optical
    networks a framework, IETF internet draft, July
    2002.
  • 60 Radhakrishnam Kadengal, Adaptive bandwidth
    management in IP/photonic networks, presentation
    file available at http//homepage.ntlworld.com/nam
    ita/krishna/rpp.html

38
Any Questions?
  • Thank you for listening!!!
Write a Comment
User Comments (0)
About PowerShow.com