Programming Requests/Responses with GreatFree in the Cloud Environment

Programming requests/responses with GreatFree is an efficient programming technique to implement distributed polling in the cloud computing environment. Distributed polling [1] is an indispensable technique to construct distributed systems. When doing that with GreatFree, the procedure becomes straightforward since developers are not required to take care of underlying tough techniques. Using the rich APIs and patterns of DIP (Distributed Infrastructure Patterns) and DOP (Distributed Operation Patterns) supported by GreatFree, the tough programming skills are turned into simplified behaviors, the operation of CPR (Copy-Paste-Replace). 1 GreatFree is a software development environment to construct cloud computing systems in diverse distributed computing environments through programming rather than configuring or scripting. Programming is defined as the procedure to implement a practical software system with essential domain knowledge and required programming skills. In the case of cloud programming, it is necessary for developers to perceive distributed knowledge as well as corresponding programming expertise. Although the overhead is high, it is the unique way to construct creative, adaptive and flexible systems with high-quality for various distributed computing environments. To lower the burden to program distributed systems, GreatFree provides developers with various code-level distributed patterns (DIP and DOP) and rich APIs.

Different from the object-oriented ones [2] which are independent of computing environments and the other distributed ones [3] which should be implemented by developers for specific environments, the code-level patterns in GreatFree specify the generic, mature and executable programs in the relatively fixed forms suitable to most distributed computing environments. For that, developers’ effort is lowered even though they implement a distributed system through programming rather than scripting and configuring.

To be fond of them, developers need to perform the system-level programming initially. That is, they determine the infrastructure of their distributed systems through choosing the most suitable GreatFree DIP initially with respect to their distributed knowledge. And then, they need to accommodate the chosen DIP if it does not fulfill some of the requirements of the particular computing environment exactly with GreatFree DOP and APIs. After that, the system-level programming is completed and it constructs a high quality system foundation for developers to perform the application-level programming using GreatFree DOP and APIs further.

When programming with the operation of CPR, two concepts, i.e., the class reference and the instance reference, need to be identified by developers. The same as most object-oriented programming approaches, GreatFree is implemented in the language of Java [4] and it supports object-oriented programming for sure. The class reference specifies the class that need to be created newly for a new feature. The instance reference specifies that class instances that need to be created newly for a new feature. In the case of programming requests/responses through CPR, one sample request/response should be chosen from DIP to follow. Those references can be retrieved according to the sample. Once if those references are available, developers can create new classes or instances respectively through straightforward operation of CPR in the corresponding patterns of DOP.

The paper is organized in the following sections. Section 1 gives a brief introduction to the technique of programming requests/responses with GreatFree. Section 2 presents the related work to implement distributed polling in distributed computing environments and makes a rough comparison. Section 3 talks about the procedures to program cloud systems with GreatFree. Section 4 explains the details to program requests/responses with GreatFree through one case. Section 5 talks about the evaluation environment of the programming technique and discusses the potential future work of the technique.

Since distributed polling is one of the fundamental functions for any cloud systems, it is required to implement it with efficient approaches. Nowadays, there are three categories of solutions to the issue, including the traditional languages solutions, the framework-based solutions [5] and the programming-oriented solutions [6] [7].

To adapt to the requirements of distributed computing environments, many new APIs are proposed to assist developers to do that. Traditional languages are defined as those programming languages that are originally established on the synchronous and standalone computing environment. In the obsolete environment, programming languages aim to solve the problems which can be processed in a sequential fashion within a single computer. That happens since the underlying CPU is designed in such a way as well as the computing resources are limited in terms of the processing power and

This content is AI-processed based on open access ArXiv data.