SDS - mazmaz2k/Minimal-balanced-node-separator GitHub Wiki
According to the use cases and the user stories we described on the SRS page before, we'll build a guideline scheme for our project.
This Document will describe the system architecture.
- Preparation for the realization of our software.
- Help us to split the tasks of the implementation between team members.
- Checking correctness of software design.
- Save coding cumbersome
- Realizing preparation clearly code
- Focus on the developments we should think about in advance, before we start coding.
- In this document you will find several diagrams: Deployment diagram, CRC Cards& Class diagram and sequence diagrams.
Capability | Capability Description | Risk Level | |
---|---|---|---|
1 | Implementation platform | The software will be implemented in Java in the eclipse environment | Low |
2 | Mode of operation | The law department will be fully synchronized with the graph department | Medium |
3 | Software division | The software will be divided into two different directions of one work of the theory rules and the other of a graph representing the theory | Critical |
4 | Functional | The data structure of the graph /rules will be dynamic | Medium |
5 | Mode of operation | The graph's data structure will run the algorithm for finding the DFS binding component, and will find the largest. | Critical |
6 | Technical specification | The data structure of the Graph/Rules will be built according to the data structure of the rules Format of body-> head |
Critical |
7 | Test | We will run tests on existing models so that there seems to be an improvement in running times. | Critical |
8 | Functional | The graph data class will return up to three vertices for the set of rules to place in these variables. | Medium |
(Each diagram should come with a description section, discussing the various considerations leading to the suggested design)




- All system’s components are based on Java , as required.
- Eclipse provides simple management interface which does not require any knowledge in software development or programming languages.
- Security is not necessary in our software.
Task | Time To Complete (Days) | |
---|---|---|
1 | Test software with simple set of rules | 1 |
2 | Test app with various set of rules | 3 |
3 | Testing the application with the mentors cnf file | 4 |
4 | Bug Fix | 4 |
5 | Present the final project to the mentors | 1 |
Risk | Severity | Response | |
---|---|---|---|
1 | Delay of development processes in building data structures. | Medium | Using existing data structures on the Internet. |
2 | New Team | Low | Diversity will help us overcome problems |
3 | The algorithm for decomposing the largest bonding component will not break the binding element equally | Low | We'll try to break down with a few vertices. We will still improve the running time after decomposition. |
4 | There is not a large set of rules that we know the inefficient running time for his solution | Medium | Resolve with set of existing laws that we already have. |
5 | Lack of experience with the framework | High | Get assistance from the lecturer and from the Internet. |
6 | The synchronization of the graph class and the structure of the data structure of the rules will not work |
High | The work on building the departments will be done in full coordination. |
7 | Choosing the working environment and programming language is not suitable for running a large set of rules |
Low | Enough that the theory will work then the code can be passed to a variety of more efficient programming languages. |
8 | Trying to recover by log file will prove to be ineffective | Medium | we will find more elegant level of code is found for the memory of deleted vertices and rules. |
9 | If no minimum model exists after decomposing the binding element | Low | If there is no minimal model, we will use the regular model.. |
10 | Our software functionality will be rejected by the mentors | Medium | We will talk about the design on the meetings and keep in touch on regular basis. |