Investments in the infrastructure necessary to support short-term or seasonal peak processing may cause unnecessary expenses, as most of such resources will remain idle most of the time. In addition, for many applications the response time is critical to meet the business requirements, which would lead to high

spendings in infrastructure, often making such approaches economically unviable, regardless of resource idleness. In this scenario, it is essential to count on mechanisms capable of performing intelligent allocation of computational resources, preferably according to the demand.

Cloud Computing is an alternative to provide “on demand” computational resources, generating economies in unprecedented scale and almost infinite scalability.

Modern software architectures allow the development of blocks of code that can be easily shared between different applications, centralizing their maintenance, development and infrastructure,
enhancing their reuse, reducing costs and increasing productivity. The aggregation of cloud technology and modern software architecture paradigms, one can encapsulate functionalities and provide a range of services – previously restricted to specific areas – with the scalability of the cloud.

Monte Carlo methods are among the most used and useful computational tools available today, providing efficient and practical algorithms to solve a wide range of scientific and engineering problems. This is a typical example of software tool that when applied to solve complex problems requires powerful computational infrastructure, which often prevents its application. On the other hand, such tool is also easily parallelizable and scalable, which makes it an excellent candidate for be implemented in the cloud.

Monte Carlo Cloud Service Framework (McCloud) provides a generic service implementation of Monte Carlo method, based on Microsoft Windows Azure, to solve a wide range of scientific and engineering problems.