CachePortal II: Acceleration of Very Large Scale Data Center-Hosted Database-driven Web Applications

Wide-area database replication technologies and the availability of data centers allow database copies to be distributed across the network. This requires a complete ecommerce web site suite (i.e. edge caches, Web servers, application servers, and DBMS) to be distributed along with the database replicas. A major advantage of this approach is, like the caches, the possibility of serving dynamic content from a location close to the users, reducing network latency. However, this is achieved at the expense of additional overhead, caused by the need of invalidating dynamic content cached in the edge caches and synchronization of the database replicas in the data center. A typical data center architecture for hosting Web applications requires a complete e-commerce Web site suite (i.e. Web server, application server, and DBMS) to be distributed along with the database replicas. Typically, the WS/AS/DBMS suite is installed in the network to serve non-transaction requests which require accesses to readonly database replicas of the master database at the origin site. In order to distinguish between the asymmetric functionality of master and slave DBMSs, we refer the mirror database in the data center as data cache or DB Cache. DBCache can be a lightweight DBMS without the transaction management system and it may cache only a subset of the tables in the master database. Updates to the database are handled using a master/slave database configuration: all updates and transactions are processed at the master database at the origin site. This architecture has two drawbacks: (1) all requests