Final Review

1. Final Review • Final Exam: Venue: this classroom Dec 18 (Monday) 6:30pm-8:30pm (upto 9pm) • five big problems, 2 hours to 2 1/2 hours (not more than 3 hours!), similar to Quizzes I/II, similar to sample final exam • short questions, “case study”, problem solving, etc. • “comprehensive”, emphasis on material covered later in the semester • Everything in lecture notes (except “optional” material, e.g., mobility) • Open-book, open-notes, open Internet • concepts, issues, mechanisms/algorithms, problem solving • NO COMMUNICIATION w/ your classmates or others during the exam! • Final conflict: please email us, and let us your availability – we’ll arrange a make-up exam for you! CSci4211: Final Review

2. (Reference) Final Letter Grade Criteria Grades are to help you check how much you have learned, where your weaknesses lie. No competition among members of the class! Everybody have a chance to get A if you work for it Depends on the situations/progress, extra credit would be available • [93 --100] A • [90 -- 93) A- • [85 -- 90) B+ • [81 -- 85) B • [78 -- 81) B- • [73 -- 78) C+ • [68 -- 73) C • [63 -- 68) C- • [58 -- 63) D+ • [50 -- 58) D • [00 -- 50) F The ranges for letter grades only revise down, not up!

3. Questions? CSci4211: Final Review

4. Putting Everything TogetherDid you get the whole picture? CSci4211: Final Review

5. Internet: A Huge Success Story • From the original four-node ARPNET research experiment to today’s global information infrastructure

6. Internet:a huge transformative & disruptive force! What has become of the Internet: Information Service and E-Commerce Platform deliver all kinds of information, news, music, video, shopping web, spotify, iTune, youtube, Netflix, Hulu, … Global Information Repository store and search for all kinds of information google, flickr, dropbox, icloud, … Cyberspace and Virtual Communities keep in touch with friends and strangers email, facebook, twitter, snapchat, … Enormous Super-Computer mobile, cloud computing and services We’re increasingly depending on it!

7. Success of Today’s Internet Today’s Internet can be primarily characterized by its success as a (human-centric, content-oriented) information delivery platform Web access, search engine, e-commerce, social networking, multimedia (music/video) streaming, cloud storage, … users search for and interact with websites (or “content”), or interact with other users; users consume or generate information static vs. dynamic content Rise of web (and HTTP) – coupled with emergence of mobile technologies – led to cloud computing and CDNs Huge data centers with massive compute and storage capacities to store information, process user requests and generate content they desire CDNs with geographically distributed edge servers to “scale out” and facilitate “speedier” information delivery

8. Low-tier High-tier Local Area Wide Area Low Mobility High Mobility More gadgets are plugged in … • smart mobile phones, iPads, e-readers, … • now TVs, thermostats, smart meters, etc., soon toasters, fridges, … • servers, desktops, laptops, … Wireless technologies revolutionizing Internet! • WiFi, bluetooth, 3/4G cellular networks, NFC, RFID, … mobile computing location services Internet of Things (IoT)

9. Within the Internet Core • Large ISPs with large geographical span and • Large content providers with huge data centers • High capacity, dense and rich topology • Cloud Computing/Services and Mobile Computing

10. CDN2 & its servers CDN1 & its servers CP2 data centers Content Distribution Ecosystem • Multiple major entities involved! • content providers (CPs), content distribution networks (CDNs), ISPs and of course, end systems & users • some entities may assume multiple roles • Complex business relationships: sometimes cooperative, but often competitive ISP CP1 data centers ISP ISP users media players

11. Static Content Distribution: YouTube as a Case Study:world’s largest video sharing site • User interaction with content (e.g., search for a video) is separated from video delivery • Employs a combination of various “tricks” and mechanisms to scale with YouTube size & handle video delivery dynamics

12. Static Content Distribution: • Has its own “data center” for certain crucial operations (e.g., user registration, …) • Most web-based user-video interaction, computation/storage operations are cloud-sourced to Amazon AWS • Users need to use MS Silverlight or other players for video streaming • Video delivery was/is partly out/cloud-sourced to 3 CDNs; • but now most utilizes its “own” OpenConnect boxes placed at participating ISPs, forming its own CDN Netflix as a Case Study: “first” large-scale cloud-sourcing success OpenConnect CDN

13. Dynamic Content Distribution • Web search as (dynamic) content delivery – e-commerce, social networking services have similar architectures • response contains both static content (e.g., banner, css files) and dynamically generated search response (dynamic content) • User QoE metric: end-to-end search response time (SRT) • Generic Web Search System Architecture • backend data centers processing search queries & generating responses • front-end edge servers (CDN) handling search query delivery Back-end (BE) Data Centers • Google: deploy its own CDN • Bing: utilized Akamai CDN (it now also builds its own CDN) • Amazon and Facebook have also built its own CDNs Front-End (FE) Servers (Edge Cloud/CDN)

14. Data Center Networks • 10’s to 100’s of thousands of hosts, often closely coupled, in close proximity: • e-business (e.g. Amazon) • content-servers (e.g., YouTube, Akamai, Apple, Microsoft) • search engines, data mining (e.g., Google) • challenges: • multiple applications, each serving massive numbers of clients • managing/balancing load, avoiding processing, networking, data bottlenecks Inside a 40-ft Microsoft container, Chicago data center

15. Data center networks • load balancer: application-layer routing • receives external client requests • directs workload within data center • returns results to external client (hiding data center internals from client) Internet Border router Load balancer Load balancer Access router Tier-1 switches B A C Tier-2 switches TOR switches Server racks 7 6 5 4 8 3 2 1

16. Data center networks • rich interconnection among switches, racks: • increased throughput between racks (multiple routing paths possible) • increased reliability via redundancy Tier-1 switches Tier-2 switches TOR switches Server racks 7 6 5 4 8 3 2 1

17. Facebook Data Center Fabric https://code.facebook.com/posts/360346274145943/introducing-data-center-fabric-the-next-generation-facebook-data-center-network/ CSci4211: Final Review

18. Facebook Data Center Fabric https://code.facebook.com/posts/360346274145943/introducing-data-center-fabric-the-next-generation-facebook-data-center-network/ CSci4211: Final Review

19. Putting Everything TogetherDid you get the whole picture? CSci4211: Final Review

20. Basic concepts in computer networks • packet switching & statistical multiplexing • protocols and layered architecture • fundamental issues in networking • distributed & complex system • addressing, protocols, … • many things can go wrong: error, loss, … • correct operations, efficiency of protocols • Application Layer • application requirements & transport services • client-server vs. peer to peer paradigms • domain name system and DNS (name vs. address) A Quick Review of What We Learned CSci4211: Final Review

21. Transport Layer: basic functions & services • multiplexing and de-multiplexing • UDP:connectionless transport service • src/dst port no.’s, checksum • TCP: connection-oriented, reliable service • TCP segment format, seq./ack. no, “flags” • connection set-up and tear down • reliable data transfer protocols • stop-&-wait, Go-back-N, selective repeat • Network Layer: basic functions & services end-to-end data delivery: addressing, routing & forwarding • network data plane vs. control plane • data plane: layer 3 routers (and also layer 2 switches) • IP addresses: network part (net prefix) vs. host part A Quick Review of What We Learned… CSci4211: Final Review

22. How to obtain an IP address: how does DHCP work? • Network service models: datagram vs. virtual circuit • IP Forwarding: datagram model • forwarding within vs. outside an IP subnet: How does a host know whether a destination is within or outside its subnet? n • within same IP network: • direct forwarding using data link layer • need to know MAC address of destination: ARP! • Outside its own IP network: • forward to its (default) router: • need to know router’s MAC address • router looks up its routing table (using longest prefix matching), and forwards to other routers if necessary; a packet finally reaches its destination host • Understanding interaction with data link layer important! A Quick Review of What We Learned … CSci4211: Final Review

23. IP datagram format • source and destination IP addresses • IP datagram id, offset, length, “fragment flags” • why IP fragmentation may be necessary • link and path MTUs • how fragmentation and reassembly done • how these fields are used • TTL, header checksum, IP options, … • ICMP protocol: • When are ICMP messages generated • What ICMP messages are used for • error/info reporting to source, ICMP redirect, … • Virtual Circuit: how to set up a VC? • incoming and outgoing VCI numbers, input/output ports • MPLS (multi-protocol label switching) A Quick Review of What We Learned … CSci4211: Final Review

24. Network Control Plane: centralized vs. distributed • (Distributed) Network Routing: basic issues • two distributed routing algorithms • link state vs. distance vector • routing information exchanged • how shortest paths computed • how routing tables constructed • count-to-infinity problem in DV • SDN and Centralized Control Plane: Openflow switches and SDN controllers • Routing in Internet • scaling issues and hierarchical routing • inter-domain vs. intra-domain routing • Intra-domain routing protocols: RIP, OSPF • Inter-domain: BGP and policy routing • customer-provider vs. peering relationships A Quick Review of What We Learned … CSci4211: Final Review

25. 5 3 5 2 2 1 3 1 2 1 F D E B C A Routing & Forwarding:Logical View of a Router CSci4211: Final Review

26. Data Link Layer : basic services and functions • data delivery over a link: framing, access control, error checking, … • MAC addresses (typically 48 bits) • flat addressing: hexadecimal notation, 45:AF:00:FF:12:01 • unicast vs. broadcast: how adapter deals with MAC addr.? • Address resolution and ARP • why do we need ARP? (see previous slide) • how does ARP work? • how are ARP messages delivered? • Interaction between IP layer and data link layer! • Broadcast local area network & media access control • Why do we need media access control (MAC?) • shared media: issues and difficulties • addressing (MAC addresses) • Taxonomy of MAC mechanisms A Quick Review of What We Learned … CSci4211: Final Review

27. Data Link Layer …: • Random access control: • ALOHA vs. slotted ALOHA • CSMA vs. CSMA/CD (carrier sensing, collision detection) • Adaptive (on-demand) controlled access: • token passing vs. polling • Efficiency of MAC protocols: light vs. heavy load • Ethernet • CSMA/CD, exponential random back-off • how does it work? basic algorithm • Some important concepts: • collision domain & network diameter • bit time, slot time (512 bit time) • Why does Ethernet have a min. frame size constraint? • Ethernet frame format • 10BaseT, 100BaseT (Fast Ethernet), Gigabit Ethernet A Quick Review of What We Learned … CSci4211: Final Review

28. Data Link Layer …: • 802.11b and Wireless LAN: • key issues & difficulties: • hidden terminal problem, power saving requirement • receiver acknowledgement needed! • how does it work? SIFS < DIFS • CSMA/CA: • how does it work? RTS, CTS, NAV • PPP: point-to-point link layer protocol, byte stuffing • Bridging: connecting multiple LAN segments • basic functions: forwarding/filtering frames • bridge forwarding table & self-learning • looping issue: bride spanning tree • Special Networking Devices we have encountered: • repeaters (hubs), bridges/(layer 2) switches, routers • What are their functions, and how do they work? A Quick Review of What We Learned … CSci4211: Final Review

29. Putting Everything TogetherDid you get the whole picture? CSci4211: Final Review

30. Walk through the whole picture, and do the following “gedanken” experiments, thinking about • operations performed at hosts/servers • source host, destination host (web, mail servers, …) • Internet infrastructure servers (DNS, DHCP, …) • addressing information at each layer • interaction between the layers (e.g.,various protocols used) • operations performed by hubs, bridges/switches, routers • what information maintained by each device, how do they get the information? what actions do they perform? • how host A downloads a web page from web server www.cs.umn.edu? • how host A telnets to host B? • how host C accesses the mail server mail.cs.umn.edu? • how host A downloads a web page from web server www.yahoo.com? • how host A accesses his/her email at his/her yahoo mail account? • how host C sends, say, an instant message, to a user logged on at host X? Putting Everything Together… CSci4211: Final Review

31. Questions? CSci4211: Final Review