Database Reference
In-Depth Information
•
Regression
Supports Linear Regression, Polynomial Regression,
and Transform Regression algorithms that are used for predictive
analytics
•
Associations
Supports association rules for predicting cause-and-
effect relationships
In addition to scoring PMML models, Streams can score a wide range of
algorithms that are exported by the SPSS Modeler Solution Publisher.
Enterprise Class
Many real-time application and parallel processing environments built in the
past have come and gone. What makes Streams so different is its enterprise
class architecture and run-time environment, which are powerful and robust
enough to handle the most demanding streaming workloads; the rich tools
for building, managing, and deploying Streams applications; the hundreds
of built-in analytic operators and functions; and its integration with enter-
prise systems. This is the value that IBM and its research and development
arms bring to the Big Data problem. Although some companies have massive
IT budgets and try to do this themselves, wouldn't it make sense to invest
that money in core competencies and the business?
Large, massively parallel jobs have unique availability requirements because
in a large cluster, there are bound to be failures. The good news is that Streams
has built-in availability characteristics that take this into account. Coupled
with the application monitoring that we describe elsewhere in this topic,
Streams enables you to keep management costs low and the reputation of
your business high. Integration with the rest of your enterprise architecture
is essential to building a holistic solution. It's a recurring theme of this topic:
IBM offers a Big Data platform,
not a Big Data product
. In this section, we cover
some of the hardened enterprise aspects of Streams, including availability
and integration.
High Availability
When you configure your Streams platform, you specify which hosts (servers)
will be part of the Streams instance. You can specify three types of hosts, as
listed next, for each server in your platform.
