- The Emergence of Web Applications
- Basic Definitions
- The Nature of the Web and Its Challenges
- Performance and Scalability
- The Internet Medium
- Wide Audience
- Always On
- Lack of Complete Control
- Measuring Performance and Scalability
- Measuring Performance
- Beyond Benchmarking
- Measuring Scalability
- Throughput and Price/Performance
- Scalability and Performance Hints
- Think End-to-End
- Scalability Doesn't Equal Performance
I'm not going to beat you over the head with the fact that millions of people use the Internet every day. That is obvious and the increasing numbers are the primary reason that application architects worry about things like scalability in the first place. However, I will inform you of a few things that you may not knowor just may not appreciate, yet.
One is that there is another Internet "audience" to consider, one that is not often addressed. This quieter, hidden, but rapidly growing group of Web clients are better known as "bots." If you are familiar with search engine technology, you already know that search engines use automated softbots to "spider" (recursively traverse) the Web and update search engine indices. This process has been going on since search engines were first deployed; bots are a simple example of one type of information agent.
Today's bots are just the tip of the iceberg. More sophisticated information agents are just around the corner that will allow users to monitor multiple sites continuously and automatically. For example, instead of using the Web interactively to watch and participate in online auctions (like those at eBay and Yahoo), users will configure information agents to watch continuously and bid automatically. This is an inevitable and obvious future direction for the Web: People want to do more than sit around watching their monitors all day, manually hunting for information.
Bots and information agents are particularly fond of things like data feeds, which are information sources that continually change and require monitoring. When the Internet was first being commercialized, it was popular to connect real-time data feeds (such as the newswire services) and build access methods to them via Web applications. This trend shows no sign of slowing; in fact, it threatens to become much greater as Web applications gradually become data feeds in themselves.
I've avoided boring, albeit frightening, statistics about the growing number of human Internet users. Instead, I've reminded you that there are and will be new types of application clients, not just those with two eyes. An increasing number of information agents will automate Web querying and a growing trend will be to treat Web applications like data feeds. In short, the Web's audience is definitely growing, not to mention changing, and so are its demands. What's more, this newer audience is persistent and regular, and does not mind testing the 24x7 feature of the Web and its applications!