concept sharding in category cloud

appears as: sharding
The Cloud at Your Service: The when, how, and why of enterprise cloud computing

This is an excerpt from Manning's book The Cloud at Your Service: The when, how, and why of enterprise cloud computing.

Figure 5.1. Statistics that show why sites such as eBay can’t survive with a centralized database. That database must be partitioned into multiple databases. Partitioning (or sharding) is a critical strategy for scaling.

To create a complex friends page, a user profile page, or a thread discussion page, you must pull together lots of different data from many sources. With sharding, you can’t just issue a query and get back all the data. You have to make individual requests to your data sources, get all the responses, and build the page. Fortunately, because of caching and fast networks, this process is fast enough that your page-load times can be excellent. In the social-networking examples we’ve been using, human response time is a pretty forgiving upper bound.

9.4.9. Designing for scale and sharding

Chapter 5 focused on designing applications that fully utilize the potential scale on the internet as well as applications that suddenly exceed local capacity and those that need sudden expanding storage. The main issue with applications that need to handle scale is that database bottlenecks hinder expansion. The solution is a database-partitioning scheme called sharding. Sharding is a decomposition of a database into multiple smaller units (called shards) that can handle requests individually.