HTTP and the Dynamic Web

1. HTTP and the Dynamic Web

2. How does the Web work? • The canonical example in your Web browser Click here • “here” is a Uniform Resource Locator (URL) http://www-cse.ucsd.edu • It names the location of an object on a server. [courtesy of Geoff Voelker] voelker@cs.ucsd.edu

3. Client Server In Action… • Client uses DNS to resolves name of server (www-cse.ucsd.edu) • Establishes an HTTP connection with the server over TCP/IP • Sends the server the name of the object (null) • Server returns the object http://www-cse.ucsd.edu HTTP [Voelker]

4. Naming and URLs • How should objects be named? • URLs name objects and the virtual locations for those objects. • Location is a DNS name, so there’s two more levels of naming and indirection under there. • Before hypertext we used to worry about access transparency. • Object name interpretation is up to the server, but it’s often a location in the local file tree. • If an object moves, the URL breaks (dangling reference). • Location-independent names seem like the obvious way to go • Why don’t we use them (e.g., URNs)? • How do we make them work, esp. in the face of mobility? [from Voelker, with additions]

5. Protocols • What kind of transport protocol should the Web use? • HTTP 1.0 • One TCP connection/object • Complaints: inefficient, slow, burdensome… • HTTP 1.1 • One TCP connection/many objects (persistent connections) • Solves all problems, right? Huge amount of complexity • Clients, proxies, servers • How do they compare? • Protocol differences [Krishnamurthy99], performance comparison [Nielsen97], effects on servers [Manley97], overhead of TCP connections [Caceres98] • HTTPS: HTTP with encryption [Voelker]

6. HTTP in a Nutshell GET /path/to/file/index.html HTTP/1.0 Content-type: MIME/html, Content-Length: 5000,... Client Server(s) • HTTP supports request/response message exchanges of arbitrary length. • Small number of request types: basically GET and POST, with supplements. • object name, + content for POST • optional query string • optional request headers • Responses are self-typed objects (documents) with various attributes and tags. • optional cookies • optional response headers

7. Scalable Servers • Of course, you are not the only person accessing the server… Server

8. Web Caching • Gee, is there some way to offload those busy servers? • Use caches to exploit reference locality among clients Clients Proxy Cache Servers [Voelker]

9. Caching • How should we build caching systems for the Web? • Seminal paper [Chankhunthod96] • Proxy caches [Duska97] • Akamai hack [Karger99] • Cooperative caching [Tewari99, Fan98, Wolman99] • Popularity distributions [Breslau99] [Voelker]

10. Issues for Web Caching • binding clients to proxies, handling failover • manual configuration, router-based “transparent caching”, WPAD (Web Proxy Automatic Discovery) • proxy may confuse/obscure interactions between server and client • consistency management • At first approximation the Web is a wide-area read-only file service...but it is much more than that. • caching responses vs. caching documents • deltas [Mogul+Bala/Douglis/Misha/others@research.att.com] • prefetching, scale, request routing, scale, performance • Web caching vs. content distribution (e.g., Akamai) • A few weeks from now...

11. HTTP 1.1 • Specification effort started in W3C, finished in IETF....much later. • A number of research works influenced the specification. • HTTP 1.0 shows the importance of careful specification. • performance • persistent connections with pipelining • range requests, incremental update, deltas • caching • cache control headers • negotiation of content attributes and encodings • content attributes vs. transport attributes • transport encodings for transmission through proxies • Trailer header and trailer headers

12. Persistent Connections • There are three key performance reasons for persistent connections: • connection setup overhead • TCP slow start: just do it and get it over with • pipelining as an alternative to multiple connections • And some new complexities resulting from their use, e.g.: • request/response framing and pairing • unexpected connection breakage • Just ask anyone from Akamai... • large numbers of active connections • How long to keep connections around? • These motivations and issues manifest in HTTP, but they are fundamental for request/response messaging over TCP.

13. Cookies • HTTP cookies (RFC2109) have brought us a better Web. • S optionally includes arbitrary state as a cookie in a response. • Cookie is opaque to C, but C saves the cookie. • C sends the saved cookie in future requests to S, and possibly to other servers as well. • Allows stateful servers for sessions, personalized content, etc. • But: cookies raise privacy and security issues. • What did S put in that cookie? Can anyone else see it? How much space does it take up on my disk that I paid soooo much for? • Cookies may allow third parties who are friends of S1,..., SN to observe C’s movements among S1,..., SN. • Unverifiable transactions, e.g., DoubleClick and other ad services.

14. Unverifiable Transactions GET x GET ad Referer mycfo.com • Users may not know that they are interacting with DoubleClick. • Amazon and MyCFO trust DoubleClick, but client is ignorant. • The user visits pages at many sites that reference DoubleClick. • DoubleClick’s cookie allows it to associate all the requests from a given user. • If the browser sends Referer headers, DoubleClick may gather information about all the sites the user visits that reference DoubleClick. ad, cookie c mycfo.com GET y GET ad, cookie c Referer amazon.com/x ad Client doubleclick, akamai, etc. amazon.com

15. Web Cache Consistency • Any caching/replication framework must take steps to ensure that the cache does not deliver old copies of modified objects. • Issues for cache consistency in the Web: • large number of clients/proxies • most static objects don’t change very often • weaker consistency requirements • Stale information might be OK, as long as it is “not too stale”. “Requirements of performance, availability, and disconnected operation require us to relax the goal of semantic transparency.” - HTTP 1.1 specification

16. Cache Expiration and Validation GET x GET x • HTTP 1.0 cache control • Origin server may add a “freshness date” (Expires) response header. • ...or the cache could determine expiration time heuristically. • Proxy must revalidate cache entry if it has expired. • Last-Modified and If-Modified-Since • Whose clock do we use for absolute expiration times? x, Last-Modified m Expires t GET x GET x GET x If-Modified-Since m Proxy Origin Server Clients 304: Not Modified

17. Expiration and Validation in HTTP 1.1 • HTTP 1.1 cache control allows origin server to: • use relative instead of absolute expiration times (max-age); • issue opaque validators (ETag for entity tag) instead of timestamps; • Origin server may specify which of several cached entries to use. GET x GET x x, ETag v max-age t GET x Age < t GET x GET x If-None-Match v Age = 0 Proxy Origin Server Clients 304: Not Modified, ETag v

18. Other 1.1 Cache Control Features • Client may specify that no caching is to occur. • private or no-store • Vary headers allow server to specify that certain request headers must also match if the proxy deems a cached response valid. • language, character set, etc. • Server may specify that a response is not cacheable. • Pragma: no-cache header since HTTP 1.0 • Client may explicitly request the proxy to validate the response. • Pragma: no-cache • Proxy may/should/must tell client the age of a cached response. • Age header • Proxy may/should/must tell client that it could not validate a non-fresh cached response with the origin server. • Warning header

19. The Dynamic Web GET program-name?arg1=x&arg2=y execute program Content-type: MIME/html, Content-Length: 5000,... Client Server(s) • HTTP began as a souped-up FTP that supports hypertext URLs. • Service builders rapidly began using it for dynamically-generated content. • Web servers morphed into Web Application Servers. • Common Gateway Interface (CGI) • Java Servlets and JavaServer Pages (JSP) • Microsoft Active Server Pages (ASP) • Microsoft ASPs are not to be confused with Application Service Providers (ASPs).

20. Multi-tier Services JNDI, JDBC,SQL HTTP RPC, RMI IIOP relational databases HTTP Clients Web application server DCOM, EJB, CORBA, etc. HTML+forms, applets, JavaScript, etc. file servers middle tiers e.g., component “middleware” transaction monitors

21. From Servers to Servlets • Servlets are dynamically loaded Java classes/objects invoked by a Web server to process requests. • Servlets are to servers as applets are to browsers. • Servlet support converts standard Web servers into extensible “Web application servers”. • designed as a Java-based replacement for CGI • Web server acts as a “connection manager” for the service body, which is specified as pluggable servlets. • interface specified by JavaSoft, supported by major servers • Servlets can be used in any kind of server (not just HTTP). • Invocation triggers are defined by server; the servlet does not know or care how it is invoked.

22. Anatomy of a Servlet init(ServletConfig config) String getServletInfo() service(....) destroy() network service (servlet container) GenericServlet (implements) Servlet ServletContext ServletConfig String getServerInfo() Object getAttribute(name) String getMimeType(name) getResource*(name) log(string) String getInitParameter(name) ServletContext getServletContext() Enumeration getInitParameterNames()

23. network service Invoking a Servlet service(ServletRequest, ServletResponse) ??? Servlet ServletInputStream readline(...) ServletResponse ServletRequest setContentType(MIME type) getOutputStream() getContentLength, getContentType, getRemoteAddr, getRemoteHost, getInputStream, getParameter(name), getParameterValues(name), ServletOutputStream print(...) println(...)

24. HTTP Servlets service(...) doGet() doHead() doPost()... GenericServlet HttpServlet ServletRequest ServletResponse HttpServletRequest HttpServletResponse getCookies(), getRemoteUser(), getAuthType(), getHeader(name), getHeaderNames(), HttpSession getSession() addCookie(), setStatus(code, msg), setHeader(name, value), sendRedirect(), encodeUrl()

25. HelloWorld Servlet import java.io.*; import javax.servlet.*; public class HelloWorld extends GenericServlet { public void service(ServletRequest request, ServletResponse response) throws ServletException, IOException { ... } public String getServletInfo() { return “Hello World Servlet"; } }

26. HelloWorld Servlet (continued) public void service(ServletRequest request, ServletResponse response) throws ServletException, IOException { ServletOutputStream output = response.getOutputStream(); String fromWho = request.getParameter(“from"); response.setContentType(“text/html"); if (fromWho == null) { output.println(“<p>Hello world!"); } else { output.println(“<p>Hello world from <em>" + fromWho + “</em>"); } }

27. Example 1: Invoking a Servlet by URL • Most servers allow a servlet to be invoked directly by URL. • client issues HTTP GET • e.g., http://www.yourhost/servlet/HelloWorld • servlet specified by HTTP POST • e.g., with form data <FORM ACTION=“http://yourhost/servlet/HelloWorld" METHOD=“POST"> From : <INPUT TYPE=“TEXT" NAME=“from" SIZE=“20"> <INPUT TYPE=“SUBMIT" VALUE=“Submit"> </FORM> generates a URL-encoded query string, e.g., “<servletURL>?from=me”