Overview on Web Caching COSC 513 Class Presentation

Overview on Web Caching COSC 513 Class Presentation Instructor: Prof. M. Anvari Student name: Wei ID: 103765

Web Caching: A Key Technique for Internet 1. What is Web Cache and what is Web Caching? n Web Cache: sits between server(s) and client(s); n Web Caching: technique to use cache concept on the web. 2. Why use Web Cache? n Reduce latency; n Reduce traffic.

3. Kinds of Web Caches: n Browser Caches: 1) use disk and memory (size 5~50 MB); 2) local to machine: private cache. n Proxy Cache: 1) an application-layer network service; 2) operated on dedicated hardware (5~50 GB hard disk, 64~512 MB RAM); 3) shared by many users.

4. Any difference between proxy and cache? n Proxy refers to an important aspect of design; n Strictly speaking, proxy does not always cache the replies pass through it; n Use term proxy cache to mean Web Cache which is implemented as a HTTP proxy. 5. Advantages of Web Caching: n objects delivered to clients with higher throughput and lower latency; n cache hits reduce WAN traffic;

load on original servers reduced; n isolate end-users from network failure. n 6. Disadvantages of Web Caching: n lose precious access counts when cache hits are served; n manual configuration often required; n cannot be implemented totally transparently; n received stale information; n require additional resources; n single point of failure.

7. Other features of caching: n Depending on specific perspective , the following features might be considered either good or bad: n servers cannot see client IP; n log file data can be used in different way; n related to above issues, cache can be used to block certain requests. 8. Web Caching effectiveness measures: n hit ratio (how about 304 Not Modified); n byte hit ratio (reply but not request);

service time (underestimate at least 2 network RTTs). 9. How Web Cache work? n Have to determine when to serve an object from the cache, if it’s available: n rules are set in protocols (HTTP 1. 0 & 1. 1); n rules set by administrators of cache (either browser user or proxy administrator). n Common rule: n 1) If object’s headers tell cache not to keep the object, it won’t. Also , if no validator is

present, most cache will mark the object uncacheable. 2) If the object is authenticated or secure, it won’t be cached. 3) A cached object is considered fresh (that is, able to be sent to a client without checking with the origin server) If: n it has an expire time or other age-controlling set, and is still within the fresh period. n If a browser cache has already seen the object , and has been set to check once a session. n

If a proxy cache has seen the object recently, and it was modified relatively long ago. n Fresh documents are served from cache directly without checking with origin servers. n 4) If an object is stale, the origin server will be asked to validate the object , or tell the cache whether the copy that it has is still good.

Together, freshness and validation are the most important ways that caches work with content. n Fresh object available instantly from cache; n Validated object will avoid resending if not changed. n

Conclusion As Web service becomes more popular, users are suffering network congestion and server overloading. Web caching is viewed to be one of the efficient techniques to alleviate server bottleneck and network traffic, thereby minimize the user access latency. Note: This presentation only covers a small part of the final report.