Context

System accesses lists of read-only data in a tabular fashion.

Problem

Using objects to represent persistent, tabular, read-only data incurs performance costs for no benefit.

Forces

• The data of multiple business entities are being accessed simultaneously.
• Natural representation format of persistent data is tabular instead of as objects.
• Data access is read-only.
• Efficiency of data access is more important than that the data be entirely current.

Solution

Access persistent data using a data access object (DAO) instead of using enterprise beans. The DAO accesses the data using the mechanism that is most efficient for the data source, and provides the client with a natural, tabular interface. The client avoids the associated overhead of enterprise beans. The client must be tolerant of data that are somewhat stale, since the DAO circumvents the concurrency control that would otherwise be provided by enterprise beans.

Structure

The following class diagram represents the Fast Lane Reader pattern:

Participants & Responsibilities

• Application Component - the component uses the Fast Lane Reader for read-only access to persistent, tabular data
• Fast Lane Reader - a data access object that adapts a persistent data source's API to provide client with tabular data
• Persistent Data Store - any source of persistent data accessible to the Fast Lane Reader; commonly accessed using JDBC^TM, connectors, or proprietary APIs

Strategies

A fast-lane reader is typically implemented as a data access object. The DAO shields clients that use the fast-lane reader from the actual implementation of the resource being accessed.

Consider the following issues when implementing this pattern:

• Nontransactional reads are inappropriate for data that change often. Nontransactional reads are often used to build input to subsequent transactions and for making choices in a user interface. In the sample application, a user creates orders based on catalog items obtained from a nontransactional read (by a fast lane reader). This works properly because catalog information changes infrequently.

  If the catalog data changes frequently, placing orders is more likely to fail, because the catalog items on which the orders depend cannot be found reliably.

• Manage long data lists with the Value List Handler pattern. The pattern breaks up the data into manageable pieces that the user or application can access efficiently.

Consequences

The main benefit of this pattern is faster data retrieval.

This pattern also has the following liabilities:

• Stale data. A fast-lane read access can only produce a snapshot of what's in the database; therefore, data received by the client may be stale.

• Increased design complexity. This pattern introduces another access path to persistent data, and circumvents the enterprise bean model.

Related Patterns

Data Access Object: A DAO facilitates fast-lane read access by encapsulating the logic required to retrieve the data from a resource.

Value List Handler: A value list handler can break up long lists of data retrieved through a fast-lane reader into manageable pieces.

© 2000-2002 Sun Microsystems, Inc. All Rights Reserved.