Hide Notes 

Slide 1: Overlay Networks (with a focus on Content Distribution Networks) Janardhan R. Iyengar CISC 856 – TCP/IP and Upper Layer Protocols 04/23/2002
Slide 2: What is an Overlay ? What is the topology of this network? WHICH network?? Figure borrowed from www.isi.edu/xbone
Slide 3: Overlay Networks: Overview • Networks built using an existing network as substrate • Also known as Virtual Networks • Most popular overlay – The Internet: Evolved as an overlay on the POTS (Plain Old Telephone System) network • Overlays could consist of routing software installed at selected sites, connected by encapsulation tunnels or direct links
Slide 4: Overlay Networks: Examples • MBone, 6Bone, ABone • RON, VNS • P2P (Napster, FreeNet, Gnutella) • Content Networks - Cooperating Caches - Server Farms - Content Distribution Networks (CDNs)
Slide 5: Example Overlays: (1) MBone • Semi-permanent testbed to carry IP multicast traffic • Routing of IP multicast traffic is not commonly integrated and deployed in production routers on the Internet • Hence, layered on the Internet to support routing of IP multicast packets using tunneling Mbone node Internet router Internet router Mbone node Internet router Internet router Mbone node
Slide 6: Example Overlays: (1) MBone
Slide 7: Example Overlays: (2) 6Bone • 6bone is an IPv6 testbed on the Internet • Intended to eventually subsume the underlying IPv4 network • IPv4 tunnels used to overlay the 6bone • ABone is the Active Networks Backbone, for experimentation in Active networking. Uses tunneling
Slide 8: Example Overlays: (2) 6Bone
Slide 9: Other known Overlays • Resilient Overlay Network (RON): Provides fault tolerance and faster recovery as compared to conventional routing techniques • Virtual Network Service (VNS): Infrastructure for provisioning QoS within Virtual Private Networks • Peer to Peer Networks: Infrastructure for distribution and sharing of files (eg: Napster, Gnutella, Freenet) • Content Networks: - Server Farms, Caching Proxies, Content Distribution Networks (CDNs) - Today, we will try to focus on CDNs - What are the motivations for Content Networks?
Slide 10: Motivations for Content Networks • More hops between client and Web server => more congestion! • Same data flowing repeatedly over links between clients and Web server C1 S C2 C3 C4 - IP router
Slide 11: Motivations for Content Networks (contd.) • Origin server is bottleneck as number of users grows • Flash Crowds (for instance, Sept. 11) • The Content Distribution Problem: Arrange a rendezvous between a content source at the origin server (www.cnn.com) and a content sink (us, as users)
Slide 12: Example content networks: Server Farms • Simple solution to the content distribution problem: deploy a large group of servers www.cnn.com (Copy 1) Request from grad.umd.edu www.cnn.com (Copy 2) Request from ren.cis.udel.edu www.cnn.com (Copy 3) L4-L7 Switch Request from ren.cis.udel.edu Request from grad.umd.edu • Arbitrate client requests to servers using an “intelligent” L4-L7 switch • Pretty widely used today
Slide 13: Example content networks: Caching Proxies ISP Client ren.cis.udel.edu Intercepters TCP port 80 traffic Other traffic Client merlot.cis.u del.edu Internet www.cnn.com Proxy • Majorly motivated by ISP business interests – reduction in bandwidth consumption of ISP from the Internet • Reduced network traffic • Reduced user perceived latency
Slide 14: Consider, On September 11, 2001 Web Server www.cnn.com New Content WTC News! request 1000,000 other hosts old content request 1000,000 other hosts ISP - Congestion / Bottleneck - Caching Proxy User merlot.cis.udel.edu
Slide 15: Problems with discussed approaches: Server farms and Caching proxies • Server farms do nothing about problems due to network congestion, or to improve latency issues due to the network • Caching proxies serve only their clients, not all users on the Internet • Content providers (say, Web servers) cannot rely on existence and correct implementation of caching proxies • Accounting issues with caching proxies. For instance, www.cnn.com needs to know the number of hits to the webpage for advertisements displayed on the webpage
Slide 16: Again, On September 11, 2001 Web Server www.cnn.com WA CA IL MA 1000,000 other users FL NY DE request new content New Content WTC News! MI 1000,000 other users User merlot.cis.udel.edu - Distribution Infrastructure - Surrogate
Slide 17: Web replication - CDNs • Overlay network to distribute content from origin servers to users • Avoids large amounts of same data repeatedly traversing potentially congested links on the Internet • Reduces Web server load • Reduces user perceived latency • Tries to route around congested networks
Slide 18: CDN vs. Caching Proxies • Caches are used by ISPs to reduce bandwidth consumption, CDNs are used by content providers to improve quality of service to end users • Caches are reactive, CDNs are proactive • Caching proxies cater to their users (web clients) and not to content providers (web servers), CDNs cater to the content providers (web servers) and clients • CDNs give control over the content to the content providers, caching proxies do not
Slide 19: CDN Architecture Origin Server CDN Request Routing Infrastructure Distribution and Accounting Infrastructure Surrogate Surrogate Client Client
Slide 20: CDN Components • Content Delivery Infrastructure: Delivering content to clients from surrogates • Request Routing Infrastructure: Steering or directing content request from a client to a suitable surrogate • Distribution Infrastructure: Moving or replicating content from content source (origin server, content provider) to surrogates • Accounting Infrastructure: Logging and reporting of distribution and delivery activities
Slide 21: Server Interaction with CDN 1. Origin server pushes new content to CDN OR CDN pulls content from origin server www.cnn.com Origin Server 1 CDN 2 2. Origin server requests logs and other accounting info from CDN OR CDN provides logs and other accounting info to origin server Distribution Infrastructure Accounting Infrastructure
Slide 22: Client Interaction with CDN 1. Hi! I need www.cnn.com/sept11 1. Go to surrogate delaware.cnn.akamai.com 3. Hi! I need content /sept11 delaware.cnn.akamai.com california.cnn.akamai.com CDN Surrogate (CA) Request Routing Infrastructure Surrogate (DE) Q: How did the CDN choose the Delaware surrogate over the California surrogate ? 1 3 Client 2
Slide 23: Request Routing Techniques • Request routing techniques use a set of metrics to direct users to “best” surrogate • Proprietary, but underlying techniques known: • DNS based request routing • Content Modification (URL rewriting) • Anycast based (how common is anycast?) • URL based request routing • Transport layer request routing • Combination of multiple mechanisms
Slide 24: DNS based Request-Routing • Common due to the ubiquity of DNS as a directory service • Specialized DNS server inserted in DNS resolution process • DNS server is capable of returning a different set of A, NS or CNAME records based on policies/metrics
Slide 25: DNS based Request-Routing Q: How does the Akamai DNS know which surrogate is closest ? Akamai CDN DNS response: A 145.155.10.1 5 delaware.cnn.akamai.com www.cnn.com Akamai DNS DNS query: www.cnn.com california.cnn.akamai.com Surrogate 145.155.10.15 on Surrogate 58.15.100.152 Se ssi DNS query: www.cnn.com merlot.cis.udel. edu 128.4.30.15 DNS response: A 145.155.10.15 local DNS server (louie.udel.edu) 128.4.4.12
Slide 26: DNS based Request-Routing www.cnn.com Akamai CDN M e re su a Akamai DNS Meas u re ts me l nt r su Me re e su a su nt lts Cli to e re to me DNS response DNS query u re N S s ea t D M ien Cl nt D NS Surrogate on Surrogate M ea Se ssi su re me n ts e M u as r s nt e m e DNS query merlot.cis.udel. edu DNS response 128.4.30.15 local DNS server (louie.udel.edu) 128.4.4.12
Slide 27: DNS based Request Routing: Caching www.cnn.com Akamai DNS Requesting DNS - 76.43.32.4 Surrogate - 145.155.10.15 Surrogate 58.15.100.152 Akamai CDN Surrogate 145.155.10.15 Requesting DNS - 76.43.32.4 Available Bandwidth = 5 kbps RTT = 100 ms Requesting DNS - 76.43.32.4 Available Bandwidth = 10 kbps RTT = 10 ms Client 76.43.35.53 Client DNS 76.43.32.4 www.cnn.com A 145.155.10.15 TTL = 10s
Slide 28: DNS based Request Routing Techniques: Discussion • Originator Problem: Client may be far removed from client DNS • Client DNS Masking Problem: Virtually all DNS servers, except for root DNS servers honor requests for recursion Q: Which DNS server resolves pel.cis.udel.edu? Q: Which DNS server performs the last recursion of the DNS request? • Hidden Load Factor: A DNS resolution may result in drastically different load on the selected surrogate – issue in load balancing requests, and predicting load on surrogates
Slide 29: Server Selection Metrics • Network Proximity (Surrogate to Client): - Network hops (traceroute) - Internet mapping services (NetGeo, IDMaps) -… • Surrogate Load: - Number of active TCP connections - HTTP request arrival rate - Other OS metrics -… • Bandwidth Availability •…
Slide 30: Value of a CDN • Scale: Aggregate infrastructure size • Reach: Diversity of content locations (diverse placement of surrogates) • Request routing efficiency, delivery techniques
Slide 31: Content Distribution Internetworking: CDI • Interconnection of content networks – collaboration between caching proxies and CDNs, as well as between individual CDNs • Greater reach, larger scale, higher capacity, increased fault tolerance • A new area, lots of challenges • Basic architecture involves gateways between various content networks
Slide 32: CDI: Architecture CDN1 CDN2 CDN3 CN4 For instance,cache network of some ISPx - Content Peering Gateway
Slide 33: Traditional vs. Overlay Content Networks Traditional networks Content networks Overlay "Content Layer" to enable richer services on top of layer 7 protocols (HTTP, RTSP) • Information processed at layers 1 through 3 of the OSI stack • Units of transported data are frames and packets • Information processed at layers 4 through 7 of the OSI stack • Units of transported data in content networks are images, movies, songs
Slide 34: In Summary • Overlays is a concept which can be used to: - deploy new services on the Internet (Mbone, 6bone, Abone, Peer-to-Peer, Content Networks) - get around problems in the underlying technology (Resilient Overlay Networks) • Further reading - Overlays: - www.savetz.com/mbone/ - www.6bone.net/ - nms.lcs.mit.edu/projects/ron/ - www-2.cs.cmu.edu/~hzhang/VNS/ • Further reading - CDNs: - www.ietf.org/internet-drafts/draft-ietf-cdi-model-01.txt - www.ietf.org/internet-drafts/draft-ietf-cdi-known-request-routing-00.txt - Bunch of papers … send me mail if you are interested • Questions? Answers? Thoughts?
Slide 35: Full-Site vs. Partial-Site Content Delivery • Full-Site delivery is what we have seen so far – entire webpage is delivered from the CDN • Partial-Site delivery delivers only embedded objects (say, only images on the webpage) from the CDN • Embedded object redirection can be done using DNS based request routing or URL rewriting Q: How many TCP connections are needed to do a P-HTTP transfer of a webpage with embedded objects using the above 2 techniques?
Slide 36: CDN with Full-Site Delivery index.html embedded image1.gif image2.gif CDN Surrogate Server Origin Server inde x imag .html, e imag 1.gif, e2.g if Client l x.htm age2.gif nde ET i 1.gif, im G ge ma ET i G
Slide 37: CDN with Partial-Site Delivery index.html embedded image1.gif image2.gif CDN e2.g mag if, i if Surrogate Server GE e1 imag T .g Client ind ex .ht imag e imag 1.gif, e 2. g if GE T ml Origin Server
Slide 38: CDN Types (Skeletal) CDNs Hosting CDN Relaying CDN Partial Site Content Delivery Request Routing Techniques Full Site Content Delivery DNS based URL Rewriting
Slide 39: DNS Outsourcing CDN Client ISP Clients CDN DNS 5 1 6 Client DNS (Local DNS server for client) (DNS server maintained by CDN company) 4 2 3 A or CNAME redirection Customer DNS (DNS containing NS entry for customer site) Content Provider
Slide 40: Tunneling v6 v4 v4 v4 v4 v6 v6 v4 v4 header IP proto = 41 (IPv6) IP proto = 6 (TCP) v6 header IPv6-SDU v6 header IP proto = 6 (TCP) IPv6-SDU
Slide 41: Example Overlays: (1) MBone • IP multicast packets are encapsulated for transmission through tunnels • Tunnel endpoints are typically workstation-class machines with OS support for IP multicast and running the mrouted multicast routing daemon • DVMRP routing algorithm used in the overlay