Databases Reference
In-Depth Information
other software. Second, you do not have to install, configure, and maintain data-
base client software. Third, performance is improved because a database wire pro-
tocol driver does both of the following:
•
Decreases latency by eliminating the processing required in the client soft-
ware and the extra network traffic caused by the client software.
•
Reduces network bandwidth requirements from extra transmissions. That is,
database wire protocol drivers can optimize network traffic because they can
control interaction with TCP.
Regarding Java and ADO.NET, another important advantage exists when
you use a database wire protocol driver/provider: The driver/provider does not
have to make calls to the client software. What does this mean?
•
For Java, this means the driver can use Pure Java code and not make calls to
the native libraries. The Pure Java standard is a set of programs, rules, and
certifications used in the design process to ensure that Java executables live
up to the WORA (write once, run always) principles. A Pure Java program
relies only on the Java Language specifications. Using native methods in a
Java program means that you lose the benefits of the Java runtime, such as
security, platform independence, garbage collection, and easy class loading.
Specific external functionality is provided by the Java core APIs, such as
JDBC.
•
For ADO.NET, this means the provider can use 100% managed code. The
benefit of 100% managed code is that it runs in the Common Language
Runtime (CLR), which provides services such as automatic memory man-
agement and lifetime control of objects, platform neutrality, and cross-lan-
guage integration. Managed code also provides improved versioning
facilities as well as easier deployment. In contrast, unmanaged code, which
includes all code written before the .NET Framework was introduced, does
not run inside the .NET environment and cannot use .NET managed facili-
ties. Performance is decreased because the CLR must perform additional
security checks.
Performance Examples
Figure 3-5 compares the performance of a wire protocol database driver to a
client-based driver. In these examples, each benchmark was run twice using the
same database server, hardware, and operating system. The only variable was the
database driver. The drivers tested were from different vendors. The benchmarks
measured the throughput and scalability of a database application.
