Enterprise Service Bus (ESB) - Guide to Cloud Computing for Business and Technology Managers - page 206

heterogeneous set of participating component systems, while sup-
porting complex aggregate (cross component) management use
cases, like dynamic resource provisioning and demand-based rout-
ing, and SLA enforcement in conjunction with policy-based behav-
ior (e.g., the ability to select service providers dynamically based on
the quality of service they offer compared to the business value of
individual transactions).
An additional requirement for a successful ESB implementation
is the ability to monitor the health, capacity, and performance of
services. Monitoring is the ability to track service activities that
take place via the bus and provide visibility into various met-
rics and statistics. Of particular significance is the ability to spot prob-
lems and exceptions in the business processes and move toward
resolving them as soon as they occur. Chapter 18 examines the man-
agement and monitoring of distributed Web services based Cloudware
platforms and applications.
11. Scalability capabilities : With a widely distributed SOA, there will be
the need to scale some of the services or the entire infrastructure
to meet integration demands. For example, transformation ser-
vices are typically very resource intensive and may require mul-
tiple instances across two or more computing nodes. At the same
time, it is necessary to create an infrastructure that can support the
large nodes present in a global service network. The loose-coupled
nature of an SOA requires that the ESB use a decentralized model to
provide a cost-effective solution that promotes flexibility in scaling
any aspect of the integration network. A decentralized architecture
enables independent scalability of individual services as well as the
communications infrastructure itself.
9.5 Leveraging Legacy Assets
There is a fundamental requirement in ESB settings to utilize functional-
ity in existing applications and repurpose it for use in new applications.
Enterprises are still burdened with older-generation operational applica-
tions that were constructed to run on various obsolescent hardware types,
programmed in obsolete languages. Such applications are known as legacy
applications.