Microsoft SQL Server 2008 R2 810-08234 Benutzerhandbuch
Produktcode
810-08234
146
CHAPTER 8
Complex Event Processing with StreamInsight
Similarly, there are certain types of applications that benefit from the ability to analyze
data as close as possible to the time that the applications capture the data . For example,
companies selling products online often use clickstream analysis to change the page layout
and site navigation and to display targeted advertising while a user remains connected to a
site . Credit card companies monitor transactions for exceptions to normal spending activities
that could indicate fraud .
companies selling products online often use clickstream analysis to change the page layout
and site navigation and to display targeted advertising while a user remains connected to a
site . Credit card companies monitor transactions for exceptions to normal spending activities
that could indicate fraud .
The challenge with CEP arises when you need to process and analyze the data before
you have time to perform ETL activities to move the data into a more traditional analytical
environment, such as a data warehouse . In CEP applications, the value of the information
derived from low-latency processing, defined in milliseconds, can be extremely high. This
value begins to diminish as the data ages . Adding to the challenge is the rate at which source
applications generate data, often tens of thousands of records per second .
environment, such as a data warehouse . In CEP applications, the value of the information
derived from low-latency processing, defined in milliseconds, can be extremely high. This
value begins to diminish as the data ages . Adding to the challenge is the rate at which source
applications generate data, often tens of thousands of records per second .
StreamInsight Highlights
StreamInsight’s CEP server includes a core engine that is built to process high-throughput
data . The engine achieves high performance by executing highly parallel queries and using
in-memory caches to avoid incurring the overhead of storing data for processing . The engine
can handle data that arrives at a steady rate or in intermittent bursts, and can even rearrange
data that arrives out of sequence . Queries can also incorporate nonstreaming data sources,
such as master reference data or historical data maintained in a data warehouse .
data . The engine achieves high performance by executing highly parallel queries and using
in-memory caches to avoid incurring the overhead of storing data for processing . The engine
can handle data that arrives at a steady rate or in intermittent bursts, and can even rearrange
data that arrives out of sequence . Queries can also incorporate nonstreaming data sources,
such as master reference data or historical data maintained in a data warehouse .
You write your CEP applications using a .NET language, such as Visual Basic or C#, for rapid
application development . In your applications, you embed declarative queries using Lan-
guage Integrated Query (LINQ) expressions to process the data for analysis .
guage Integrated Query (LINQ) expressions to process the data for analysis .
StreamInsight also includes other tools for administration and development support . The
CEP server has a management interface and diagnostic views that you can use to develop
applications to monitor StreamInsight . For development support, StreamInsight includes an
event flow debugger that you can use to troubleshoot queries. An example of a situation that
might require troubleshooting is the arrival of a larger number of events than expected .
applications to monitor StreamInsight . For development support, StreamInsight includes an
event flow debugger that you can use to troubleshoot queries. An example of a situation that
might require troubleshooting is the arrival of a larger number of events than expected .
StreamInsight Architecture
As with any new technology, you will find it helpful to have an understanding of the Stream-
Insight architecture before you begin development of your first CEP application. Your ap-
plication must restructure data streams to a format usable by the processing engine . You use
adapters to perform this restructuring before passing the data to queries that run on the CEP
server . The way you choose to develop your application also depends on the deployment
model you use to implement StreamInsight .
Insight architecture before you begin development of your first CEP application. Your ap-
plication must restructure data streams to a format usable by the processing engine . You use
adapters to perform this restructuring before passing the data to queries that run on the CEP
server . The way you choose to develop your application also depends on the deployment
model you use to implement StreamInsight .