摘要:
An improved load balancing method, system, and computer program product handles popular object requests using a front-end cache, and hashing is applied only to the requests in the stream that were not handled by the front-end cache. A cache (e.g., a web proxy cache) is placed in front of a Level 7 switch, such that the cache services the popular requests from the cache based on the content of the request (e.g., based on the portion of an HTTP request following the domain name). The remaining requests are hashed and then routed to the back-end server. This allows the requests that make it past the cache to still be routed to the back-end server and take advantage of the efficiencies provided therefrom.
摘要:
Popular object requests are handled using a front-end cache, and hashing is applied only to the requests in the stream that were not handled by the front-end cache. A cache (e.g., a web proxy cache) is placed in front of a Level 7 switch and services the popular requests from the cache based on the content of the request (e.g., based on the portion of an HTTP request following the domain name). The remaining requests are hashed and then routed to the back-end server. Requests that make it past the cache are still routed to the back-end server. A Level 4 switch is placed in front of a plurality of web proxy caches, each of which are in turn placed in front of a respective Level 7 switch, each of which are connected to a respective server farm, so that incoming web requests are handled on a round robin basis (or other SLB technique) before being sent to the cache.
摘要:
In a computer network system, the caches at individual stations are available to other stations. A central cache directory is maintained at a network server. Each time a station caches a data object received from a remote network, it informs the central cache directory. When a station comes online, it is asked to send a list of the contents of its cache. Whenever a station seeks an object from the remote network, the local network server first checks the central directory cache to see if the request can be satisfied at one of the local stations. Only if it cannot is the requested object retrieved from the remote network.
摘要:
A configuration method includes selecting a design configuration, and modeling the solution configuration having a failure event for the selected solution configuration to determine an amount of down time associated with the failure event. A first penalty cost function is applied the amount of down time to determine a first penalty cost associated with the failure event of the selected solution configuration.
摘要:
An embodiment of a method of designing a storage system begins with a step of assigning decision variables to a range of candidate storage systems. Each of the decision variables identifies a single candidate storage system. The method concludes with a step of employing the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost.