This research aims to support software engineers in their first attempt at a conversion of relational databases to a database technology recognized as better suited to cloud computing (NoSQL), which in this particular case study is HBase. To do this, an initial research activity aims to understand and clarify the notions of essential and optional characteristics in the cloud computing definition. This discussion is considered by taking the perspective of this research (i.e. a database conversion from RDB to NoSQL). Subsequently, the main contribution of this research explores the possibility of extracting database conversion rules.
Two experiments were carried out to identify an initial list of conversion rules:
1. The first experiment consisted of asking participants (engineers from École de technologie supérieure) to convert a particular schema, without the use of a guide or conversion rules. Several questionnaires were completed by the participants during the experiment. The main goal of this experiment was to evaluate the need for conversion rules to help software engineers in the conversion of relational schemas to non-relational schemas.
2. The second experiment consisted of dividing the participants into sub-groups tasked to convert a single RDB relation (i.e. one-to-one) to a possible schema in HBase. The goal was to test the results of a basic relationship compared with all the schematic conversion possibilities in HBase, a particular non-relational database, and then extract the conversion rules. This experiment would show why one particular schema is better than another and would allow for the identification of conversion rules based on facts (e.g., performance measurement of a query on each of the resulting schemas).
The main contributions of this thesis are:
1. A list of rules for converting relational database schemas to the schema of a non-relational database:
a. Rules on data proximity;
b. Rules on column families;
c. Rules on data quantity;
d. Rules on access patterns.
2. Clarification to the term cloud computing is also proposed by introducing the notion of mandatory and optional characteristics excluded from the most popular definition.
| Date | 1 Feb 2020 |
|---|
| Original language | American English |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Alain April (Supervisor) |
|---|
Ouanouki, R. (Author),
April (Supervisor),
1 Feb 2020Student thesis: Doctoral thesis › Doctorate in Engineering: Engineering