Grid Middleware is a key technology for automating different analysis processes on big data. It is a technology that presents the computing Grid as a single system by hiding administrative and geographic boundaries. I thus provides seamless, secure, and intuitive access by hiding complexities in such a way that its Grid infrastructure appears transparently to its users. This aforementioned example is a specialization of the General term ‘middleware’ known from distributed systems. The goal is to establish a single system view for end-users.
Goals of Middleware
One of the key goals is to connect users and resources. That means access to remote resources, the support for collaboration and to guarantee security. Abstracting from low-level differences (e.g. different schedulers) is important in this context. Another goal is to achieve transparency. The middleware should hide the fact that resources are physically distributed. But choosing the right degree of transparency is often a trade-off decision between functionality and usability. The goal of openness refers to the fact to offer services according to standard rules (e.g., syntax and semantics). Web standards are re-used following the success of the ‘open Internet’. Scalability is another goal that is particularly important since Grids are expected to grow in size and spectrum of services. That means it allows for growth in number of users/resources, geographical extent, number of administrative Domains, and type of Grid resources.
More Information about middleware
We refer to the following video about this subject:
Follow us on Facebook: