Enhancing MIDAS towards a transparent interoperability between SaaS and DaaS

Abstract

Over the years, Software as a Service (SaaS) has become a common delivery model for many applications. In cloud applications, a huge volume and variety of data can be generated and they can be available for consumption by DaaS (Data as a Service). For this, the data provided by DaaS can be stored in a non-structured (e.g. text), semi-structured (e.g. XML, JSON) or structured format (e.g. Relational Database). However, the access of that kind of DaaS, in a transparent manner, needs substantial efforts due to the lack of interoperability between SaaS and DaaS. In this paper, we propose a new enhanced version of MIDAS, middleware to provide seamlessly and independently interoperability between SaaS and DaaS. First, this new version of MIDAS allows both semi-structure and structure data format from SaaS. It mediates queries from NoSQL (e.g. MongoDB) and SQL (MySQL) databases. Secondly, it was enhanced with Join operations, both in SQL and in NOSQL statements. And lastly, other formats were added for the DaaS to fit SaaS requests, such as JSON, XML, and CSV formats. To evaluate this new version of our middleware, we provide three types of experiments to cover critical issues such as execution time, the overhead of our approach, and scalability of MIDAS. Our results show the effectiveness of our approach to tackling interoperability issues in cloud computing environments.

Defining and Providing Pragmatic Interoperability: The MIDAS Middleware Case

“…In the third study, we develop a prototype of the pragmatic middleware for SaaS and DaaS: MIDAS 3.0. Similar to the previous versions [Marinho et al 2016, Vieira et al 2017, Ribeiro et al 2018, Ribeiro et al 2019b, Mane et al 2020], MI-DAS 3.0 intermediates communication between SaaS applications and heterogeneous data sources (e.g., DaaS and DBaaS) independently of API. We incorporated our framework into MIDAS to discuss and present a middleware version for pragmatic interoperability.…”

Transparent Interoperability Middleware between Data and Service Cloud Layers

“…Despite the fact that our prior approach (MIDAS 1.6) (Vieira et al, 2017), it had some limitations: (i) Each DaaS must be manually inserted and updated; (ii) DBaaS is not provided; and (iii) data were returned to SaaS only in JSON format.…”

“…The following components were included or modified to meet the goals of our current version: Data Mapping, Query Builder, Result Formatter, and Crawler. The other components, Query Decomposer and DIS, both works similarly to our previous version (Vieira et al, 2017).…”