This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
|
teaching:infoh415 [2020/10/04 12:10] ezimanyi [Project] |
teaching:infoh415 [2020/10/05 18:20] ezimanyi [Topics for the current academic year] |
||
|---|---|---|---|
| Line 141: | Line 141: | ||
| It is important to understand that the objective of the project is NOT about developing an application with GUI. The objective is to benchmark the proposed tool in relation to the database requirements of your application. Therefore, it is necessary to determine the set of queries and updates that your application requires and do a benchmark with, e.g., 1K, 10K, 100K, and 1M "objects" (rows, documents, nodes, etc. depending on the technology used) to determine if the tool shows a linear or exponential behavior. As usual when performing benchmarks, the queries and updates are executed n times (e.g., 6 times where the first execution is not considered because it is different from the others since the cache structures must be filled) and the average of the execution times is computed. A comparison with traditional relational technology must be provided to show that the chosen tool is THE technology of choice for your application, better than all other alternatives, and that it will perform correctly when the system is deployed at full scale. | It is important to understand that the objective of the project is NOT about developing an application with GUI. The objective is to benchmark the proposed tool in relation to the database requirements of your application. Therefore, it is necessary to determine the set of queries and updates that your application requires and do a benchmark with, e.g., 1K, 10K, 100K, and 1M "objects" (rows, documents, nodes, etc. depending on the technology used) to determine if the tool shows a linear or exponential behavior. As usual when performing benchmarks, the queries and updates are executed n times (e.g., 6 times where the first execution is not considered because it is different from the others since the cache structures must be filled) and the average of the execution times is computed. A comparison with traditional relational technology must be provided to show that the chosen tool is THE technology of choice for your application, better than all other alternatives, and that it will perform correctly when the system is deployed at full scale. | ||
| - | The choice of topic and the application must be made in agreement with the lecturer. The topic should not be included in the program of the Master in Computer Science and Engineering. The project will be presented to the lecturer and the fellow students at the end of the semester. This presentation will be supported by a slideshow. A written report containing the contents of the presentation is also required. The presentation and written report will explain the possibilities offered by the database management system chosen and give a general description of the application implemented. | + | The choice of topic and the application must be made in agreement with the lecturer. The topic should not be included in the program of the Master in Computer Science and Engineering. The project will be presented to the lecturer and the fellow students at the end of the semester. This presentation will be supported by a slideshow. A written report containing the contents of the presentation is also required. The presentation and the report will (1) explain the foundations of the technology chosen, (2) explain how these foundations are implemented by the database management system chosen and (3) illustrate all these concepts with the application implemented. |
| The evaluation of the project focuses on the following criteria: | The evaluation of the project focuses on the following criteria: | ||
| Line 184: | Line 184: | ||
| * Document stores and MongoDB: Beata Janiak and Astrid Asoumoy | * Document stores and MongoDB: Beata Janiak and Astrid Asoumoy | ||
| * Key-value databases and BerkeleyDB: Mohamed Amchemer and <TBD> | * Key-value databases and BerkeleyDB: Mohamed Amchemer and <TBD> | ||
| + | * Key-value databases and DynamoDB: Abdeslam Bakkali Taheri and <TBD> | ||
| * Multimedia databases and Oracle: Fan Chen and Noëmie Muller | * Multimedia databases and Oracle: Fan Chen and Noëmie Muller | ||
| * Object-oriented databases and Perst: Nicolas Boucher and Romain Perret | * Object-oriented databases and Perst: Nicolas Boucher and Romain Perret | ||