Transcript
8/11/2019 Dm Netedu Presentation Feb07
1/15
DM, February 2007, p.1NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Networking Research:Trends and Issues
Deep MedhiNetworking & Telecommunication Research Lab (NeTReL)
Computer Science & Electrical Engineering DepartmentSchool of Computing & Engineering
University of Missouri-Kansas City, USAhttp://www.csee.umkc.edu/~dmedhi
dmedhi@umkc.edu
D. Medhi, 2007. Permission is granted for this material to be shared for non-commercial, educational purposes.To disseminate otherwise or to republish requires written permission from the author.
DM, February 2007, p.2NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Outline
Review: CSTB 2001 ReportBroad Categories for NetworkingResearch
TrendsSome Historical ExamplesIssues
8/11/2019 Dm Netedu Presentation Feb07
2/15
DM, February 2007, p.3NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
CSTB2001 ReportLooking Over the Fence at Networks: ANeighbor's View of Networking Research By Computer Science & Telecommunications
Board (CSTB), National Research Council ofUS
Published in 2001 by National Academy Press,availablehttp://books.nap.edu/html/looking_over_the_fence/report.pdf
Addresses Three Broad Areas
DM, February 2007, p.4NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
CSTB 2001 report: First Area
Measuring: Understanding the InternetArtifact Challenge of Scale
How to infer based on incomplete knowledge ofconfiguration
How to soundly sample network traffic, and validity ofsampling approach
Measurement Infrastructure Deployment and operational challenges
Nontechnical factors Compose of production commercial systems
E.g., Confidentiality and privacy of data
8/11/2019 Dm Netedu Presentation Feb07
3/15
8/11/2019 Dm Netedu Presentation Feb07
4/15
DM, February 2007, p.7NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Trouble with Success (CSTB 2001Report)
ossification := The process of becoming set in arigidly conventional pattern, as of behavior, habits,or beliefs (American Heritage Dictionary)Intellectual ossification E.g., its not TCP!
Infrastructure ossification What researchers want may not be deployed in a
commercial network (e.g., limited multicast deployment,
QoS)System ossification Research results judged based on how hard it is to deploy on
the Internet
DM, February 2007, p.8NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
CSTB2001 ReportTo Summarize Very well done Over five years old Current Internet view/improvement
Whats Missing? Wireless networking
Optical networking Delivery/Access networks (e.g., Cable/DSL Networks) Interaction of different networks Network security
8/11/2019 Dm Netedu Presentation Feb07
5/15
DM, February 2007, p.9NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Network Research:
Two Broad Categories: Current networks & related research To imagine future networks/services &
related research
Influence of External/Overnight
Factors
DM, February 2007, p.10NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Current Networks-Related Research
How to measure/model behaviorDriven by limitationsHow to improve
Due to scalability issues, operationalproblems Traffic engineering Network reconfigurability Security
8/11/2019 Dm Netedu Presentation Feb07
6/15
DM, February 2007, p.11NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Research Trends: Current Networks-related: A Few Examples
Scalability (in every way thinkable! routingtable lookup, network models, )Internet measurement, sampling Backbone Networks Access Networks
Protocol extensibilityRouting: BGP, inter-network trafficengineeringIntegration of traditional IP networks andPSTN/multimedia services
DM, February 2007, p.12NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
uture Networking or Services:Research, Technology Concepts &Development
Some Recent Examples: (~last 20 years)Networks: ATM, MPLS, Optical networking
Cellular Networks, 3GPP, WiMAXServices: WWW, P2P, IM, text messaging
Associated protocol development, modeling,routing, security, May or may not be commercially successful
8/11/2019 Dm Netedu Presentation Feb07
7/15
DM, February 2007, p.13NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
esearch Trend: utureNetworks/Services and related
researchNetworks/Services: MPLS/GMPLS Virtual Private Networking Sensor Networks Software Radio Service Concept Development
IP Multimedia SubSystem (IMS) Secure Group Communication
Faces issues such as MAC protocol, performance modeling,security as built- in feature, protocol development,performance etc
DM, February 2007, p.14NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
External/Overnight Factor
Something an architecture wasntoriginally intended for, but is forced tohandle
8/11/2019 Dm Netedu Presentation Feb07
8/15
DM, February 2007, p.15NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
TCP Collapse
TCP throughput drop (1986)
Led to new congestion control/timeradjustment approach Key here is on understanding the dynamics
Sliding-window, window-adjustment, timer-
adjustment, sampling,
DM, February 2007, p.16NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Denial of Service (DoS) Attacks
Network functionality allowed the possibilityfor DoS attack First wave: Web-server oriented Next wave: network impact (e.g. code-red virus)
Follow-up Research TCP accept queue, OS implementation etc Source-oriented (stop close to the source) Router-level research: IPprefix lookup
8/11/2019 Dm Netedu Presentation Feb07
9/15
DM, February 2007, p.17NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
IP Network Traffic Engineeringissues for large providers
OSPF/IS-IS: dynamic routing protocolProblem: determine link weights in OSPF/IS-IS networks to load balance a networkEarly-exit/Late-exit routing issues
Router Buffer sizing: Bandwidth-delay product
Consider 40 Gbps link and 250 ms round-trip delay
DM, February 2007, p.18NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Routing Table Growth
CIDRIP address lookup algorithms Need to stay with line rate (approaching
40Gbps)Packet Filtering and classification
8/11/2019 Dm Netedu Presentation Feb07
10/15
DM, February 2007, p.19NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Router Architecture evolution
General purpose computer architectureDifferent architectures for differentpurposes Shared CPU architectures Shared forwarding engine architectures Shared nothing architectures
Clustered architectures
DM, February 2007, p.20NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Router Architectures
8/11/2019 Dm Netedu Presentation Feb07
11/15
DM, February 2007, p.21NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Needs/Issues in NetworkingResearch
Recap: Current networks-related research Future networks/services-related research External factor that drives research (often
for current networks; however, principlesmight be useful for future networks)
Issues: To highlight a few examples
DM, February 2007, p.22NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Broadband Access Technology andService Dynamics
Broadband access Technology (DSL,Cable Modem) designed for webmodel Lot of download bandwidth Not much upstream bandwidth
Challenge faced (last few years) P2P services: Napster, VoIP services
8/11/2019 Dm Netedu Presentation Feb07
12/15
DM, February 2007, p.23NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Broadband Access Technology andService Dynamics (contd)
Can we design an access mechanism that isdynamically adjustable between upstream anddownstream Time-Division Duplexing (TDD)? Others? Require solid knowledge of physical, data link, and
transport layer?Questions: How well will the dynamics work? How does it impact service behavior? Where is the trade-
off? Are there possible pitfalls due to dynamism? And How to
handle them? (recall: lesson learned from Telephone dynamic routing, TCP
congestion, )
DM, February 2007, p.24NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Core Network Interactions: Multi-Layer Networking Dynamics
If IP layer link setting is dynamicallyadjustable to address for failureIf optical networking handles its owndynamic routing (GMPLS) Good for each network, may not be good
together? Can the inter-related network go into tailspin?
How does DoS attack based increase in networktraffic impact dynamic adjustability?
How do we know good from bad traffic (e.g., anew P2P product popularity, or a new DoSattack)?
8/11/2019 Dm Netedu Presentation Feb07
13/15
DM, February 2007, p.25NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Routing table growth
Data source: http://bgp.potaroo.net/
Our recent estimate:N = 105625 + 920.608 (4 Y - 8007)1.52894 , for Y >=2002
DM, February 2007, p.26NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Routing table growth: problemsInternet Architecture Board: Routing andaddressing workshop (reported at IETF67,San Diego, CA, Nov06) TCAM/SRAM for routers slow growth ASIC is already pushing limits
Memory speed improves about 10% per year State growth is super-linear Current trends in the growth are not scalable Use of IP addresses for both ID and Location is a
problem Sort of second CIDR wave, but more difficult
8/11/2019 Dm Netedu Presentation Feb07
14/15
DM, February 2007, p.27NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Future Directions:
Developing disruptive prototypesthat challenge the current InternetHow to do it?Not possible to test over the currentInternet Its a production network!
Retro-fitting is difficult (remainspatch-work problem)
DM, February 2007, p.28NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Steps to Making it happen ..NSFs new research program: FIND (Future Internet Design) Inviting proposals on radical architectural directions
GENI (Global Environment for NetworkingInnovations) http://www.geni.net A large-scale proposed test-bed for experimentation
Experimentation in every layer of the network Customized hardware development Test before deploy time environment Allow many researchers to participate Alliance with European Community
8/11/2019 Dm Netedu Presentation Feb07
15/15
DM, February 2007, p.29NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
How to ensure that all problemsare addressed
Classify problems: Wired/wireless,Core/AccessCore Networks
Some idea because of IETF, NANOG (e.g. BGP problem,addressing growth )
Access Networks Cable/Modem, wireless Campus/Enterprise Networks
Need input from Net@Edu community
Security/Resilience as inherent designparadigm
DM, February 2007, p.30NeTReL, N et w ork i ng & Tel ecommuni cati ons Research
Finally, at a Fundamental Level
Is there a new network communicationmode out there?Consider time spacing:
+ Between Telegraphy and Telephony: ~50years
+ Between Telephony and Internet: ~100 years+ Between Internet and ??
- Will everything now onward be incremental?- How to not get trapped in ossification?
top related