1. Caching and Content Distribution Networks

2. Web Caching r As an example, we use the web to illustrate caching and other related issues browser Web Proxy cache request response request response Web server browser Web server request response

3. Web Browser Caching r Web browsers have their own caches. When a page is downloaded from a site the web page is put into the browser cache. r This is especially useful in those cases when the back button is pressed.  If a new copy is needed then a “ refresh ” can be done. r No page stays permanently in the cache. There is limited room. m A replacement algorithm is needed to determine which cached page should be purged.

4. Why Web Server Caching r Latency m Reduce latency m Request does not require going to the server m Request is served from the client side which means that network communication is avoided r Reduce traffic

5. Consistency r What if the page changes after saved in the cache? m This means that cached copy is out of date m The copy and the original are not consistent r There are different strategies for dealing with this

6. Web Browser Caching r Client pull m The server provides the content with instructions on when the client should ask for a refreshed copy of the content or if the content should be cached. r Server push m The server transmits page information to the screen. m The browser application displays the information and leaves the connection to the server open. m With an open connection, the server can continue to push updated pages for your screen to display on an ongoing basis. You can close the connection by closing the page. m The server is in control u Browser caches are different from proxy caches (discussed next).

7. Web Caching r Proxy caches (also called proxy server) m Intercepts HTTP requests from client Serves object if in its cache and the date is still valid If not go to object ’ s home server –On behalf of user, gets the object and possibly deposits in its cache before returning to user Usually deployed at edges of a network –Wide area bandwidth savings, improved response time and increased availability of static web-based objects r A browser may have to be configured to point to the proxy server. r Usually a proxy cache is purchased and installed by an organization

8. Web Caching r Not all web pages can be cached m If the Last-Modified tag then page can be cached r Refresh is often done when m There is a request; and m Expiry time has passed

9. Cooperative Caching r Caching infrastructure can have multiple web proxies m Proxies can be arranged in a hierarchy or other structures m Proxies can cooperate with one another Answer client requests Propagate server notifications m Uses a combination of HTTP and ICP (Internet Caching Protocol). ICP can be used by one cache to quickly ask another cache if it has an object. HTTP is used to actually retrieve the object.

10. Problems r Caching proxies do not serve all Internet users r Content providers (say, Web servers) cannot rely on existence and correct implementation of caching proxies. r Accounting issues with caching proxies: m Example: www.cnn.com needs to know the number of hits to the advertisements displayed on the web page.

11. Content Distribution Networks (CDN) r Business Model: A content provider such as www.cnn.com or Yahoo pays a CDN company (such as Akamai) to get its content to the requesting users with short delays. r A CDN provides a mechanism for m Replicating content on multiple servers in the Internet m Providing clients with a means to determine the servers that can deliver the content fastest.

12. Terminology r Content: Any publicly accessible combination of text, images, applets, frames, MP3, video, flash, virtual reality objects, etc. r Content Provider: Any individual, organization, or company that has content that it wishes to make available to users.  Origin Server: Content provider ’ s server, where the content is first uploaded.  Surrogate Server (sometimes called edge server): Content distributor ’ s server, where the replicated content is kept.

13. Players Content Provider H/W and S/W Vendor Content Distributor Hosting Provider Yahoo, MSNBC, CNN CBC Cisco, Oracle- Sun Akamai, Bell Sells servers Send content Install servers

14. CDN Distribution r Content providers are CDN customers Content replication r CDN company installs thousands of servers throughout Internet m In large datacenters m Or, close to users r CDN replicates customers’ content r When provider updates content, CDN updates servers origin server in North America CDN distribution node CDN server in S. America CDN server in Europe CDN server in Asia 14

15. CDN: Functional Components r Distribution Service r Redirection Service r Accounting and Billing system

16. CDN:Distribution Service r The content provider determines which of its objects it wants the CDN to distribute. r The content provider tags and then pushes this content to a CDN node, which in turn replicates and pushes the content to all its CDN servers.

17. CDN: Redirection  When a browser in a user ’ s host is instructed to retrieve a specific object (specified using a URL), how does the browser determine whether it should retrieve the object from the origin server or from one of the CDN servers? r As an example, suppose the hostname of the content provider is www.cnn.com

18. How Akamai Works End-user cnn.com (content provider) DNS root server 12 Nearby Akamai cluster GET index. html 18 http://a.73.g.akamai.net/7/23/ cnn.com/af/cnn.com/foo.jpg HTTP Akamai cluster Akamai global DNS server Akamai regional DNS server

19. CDN: Redirection r Users get an html document from www.cnn.com; this could be index.html r The file index.html uses a modified URL for content that has been replicated. r Example: If the jpeg files are what has been replicated then may be modified as follows: r The browser needs to resolve a73.g.akamai.net hostname for replicated content.

20. CDN: Redirection r What does this mean? m host part: a73.g.akamai.net m Akamai control part: /7/23 m Content URL: /af/foo.jpg

21. CDN: Redirection r DNS is configured so that all queries about g.akamai.net that arrive at a DNS server are sent to an authoritative DNS server for g.akamai.net. r This is referred to as a Akamai DNS server (authoritative DNS server)

22. How Akamai Works End-user cnn.com (content provider)DNS root server 12 Nearby Akamai cluster DNS lookup cache.cnn.com Akamai cluster 3 4 ALIAS: g.akamai.net Akamai global DNS server Akamai regional DNS server

23. CDN: Redirection r DNS is configured so that all queries about g.akamai.net that arrive at a DNS server are sent to an authoritative DNS server for g.akamai.net. This is referred to as a Akamai DNS server (authoritative DNS server) r When the Akamai DNS server receives the query, it extracts the IP address of the requesting browser..

24. P P How Akamai Works End-user cnn.com (content provider) DNS root server 12 Akamai global DNS server Akamai regional DNS server Nearby Akamai cluster Akamai cluster 3 4 6 5 ALIAS a73.g.akamai.net DNS lookup g.akamai.net

25. CDN: Redirection r Based on the IP address and information that it has about the Internet (called a map), the IP address of an Akamai regional server is returned to the requesting browser based on policy m e.g., select the server that is the fewest hops away. r The regional server may choose a surrogate server for content retrieval

26. HTTP How Akamai Works End-user cnn.com (content provider) DNS root server 12 Akamai global DNS server Akamai regional DNS server Nearby Akamai cluster Akamai cluster 3 4 6 5 8 7 DNS a73.g.akamai.net Address 1.2.3.4

27. HTTP How Akamai Works End-user cnn.com (content provider) DNS root server 12 Akamai global DNS server Akamai regional DNS server Nearby Akamai cluster Akamai cluster 3 4 6 5 8 7 9 GET /foo.jpg Host: cache.cnn.com

28. HTTP How Akamai Works End-user cnn.com (content provider) DNS root server 12 Akamai global DNS server Akamai regional DNS server Nearby Akamai cluster Akamai cluster 3 4 6 5 8 7 9 GET /foo.jpg Host: cache.cnn.com 12 11 GET foo.jpg

29. CDN Redirection r The Akamai DNS server IP address is now in the cache of the local DNS server. m This implies that it is not always necessary to go to the root DNS server. r The TTL associated with the IP address of an Akamai server(surrogate) is relatively small. m This is done for performance reasons. r Akamai content distribution servers are caches

30. CDN Redirection r What if content is not there? m If the request content is not found then the surrogate will ask other surrogates within a specified region for information. m If requested information is still not found or is stale, then a request is made to the original web site.

31. CDN Selection r The tricky issue is selecting which local content server to use for a particular request m Want to spread load evenly m Want minimal impact if server is added or removed. r In Akamai, each surrogate server sends measurement results to the Network Operations Communications Center (NOCC). m Measurement results include number of active TCP connections, HTTP request arrival rate, bandwidth availability, etc m This information is used by the Akamai DNS server.

32. Accounting Mechanism r Accounting mechanisms collect and track information related to request routing, distribution and delivery. r Information is gathered in real time and put into log files for each CDN component. r This gets sent to the Network Operations Communications Center (NOCC).

33. Full Site Delivery vs. Partial Site Delivery r Full Site Delivery : All the contents are delivered by the CDN (including HTML, images, and other objects). r Partial Site delivery: Only images, streaming media and other bandwidth intensive objects delivered by the CDN.

34. Current Akamai Customers

35. Summary r We have examined replication and issues related to the design and implementation of a replicated system. r Many choices and tradeoffs to consider