Iteration 4: Design
The ISS analysis tool is a web application that provides analysts at Barrios, an aerospace company, with logistics optimization, usage forecasting, and prediction accuracy analysis surrounding consumable delivery to the International Space Station (ISS).
- Provide a clean, usable, intuitive interface for Barrios employees to upload user-generated data and view analysis results.
- Provide accurate and useful analyses, including supply-chain optimization, predictive modeling, and prediction accuracy analysis.
- ISS: International Space Station
- IMS: Inventory Management System - Barrios' ISS inventory management datastore
- ORM: Object Relational Mapper - A framework-specific set of classes and methods for interacting with a database
- ASGI: Asynchronous Server Gateway Interface - An asynchronous calling convention for Python web servers
-
DOM: The Document Object Model - A data structure that represents an
HTMLdocument.
- Emmett: A "full-stack" web framework written in Python
- HTMX: A client-side JavaScript library that enhances HTML for easy communication with the server
- Hypermedia APIs vs. Data APIs
- PostgreSQL: A relational SQL database
- Pandas: A Python library for exploring and manipulating tabular data
- DuckDB: An analytical database system that allows developers to run SQL queries directly on flat CSV files
- Prophet: A forecasting algorithm
The ISS analysis tool will give Barrios analysts the ability to quickly transform user-provided data into three types of analysis: prior prediction accuracy assessment, logistics optimization, and predictive modeling.
The current architecture is based on a Model-View-Controller architecture. HTTP requests to the server will be handled by Controllers, which will coordinate with Models to generate or persist analyses based on how the user interacts with Views. In its current state, the system handles basic file uploads and the persistence of user-provided data to
The proposed system has not changed since it was proposed in Iteration 3, with the exception that the proposed system might use DuckDB for performant, on-the-fly analysis of user-provided CSV files. The system will also use an HTTP server framework named Emmett to facilitate a basic Model-View-Controller application structure. With this structure, controllers will respond to HTTP requests sent from a browser-based HTML interface augmented with the HTMX library and coordinate the creation of optimization analysis and predictive modeling using either the Pandas data library or DuckDB and the Prophet algorithm. The system will also facilitate uploading and storing client-generated data using PostgreSQL.
Auth & User
Flight Plans
Inventory Management System Overview
Inventory Management Detail
Other Tables
Access matrix
| Role | Analyses | View Files | Upload Data | Manage Users |
|---|---|---|---|---|
| Viewer | Read | |||
| Analyst | Read/Write/Execute | Read | Read/Write/Execute | |
| Admin | Read/Write/Execute | Read | Read/Write/Execute | Read/Write |
Global software control for this system will be handled with a server-client architecture implemented using the Emmett web framework. This will allow the server to respond to HTTP requests from a browser with HTML templates. These templates will be augmented using the HTMX library.
Requests from the browser will be handled asynchronously by the server, allowing the system to respond to a larger number of requests than a synchronous server. This is made possible because the Emmett framework follows the ASGI, or Asynchronous Server Gateway Interface convention. This convention allows web server software to respond to numerous concurrent requests without the need to enlist additional processes, which will save expensive computing resources.
Synchronization between the browser and server will be handled like traditional HTTP servers. Rather than issuing responses using an intermediate language like JSON, server endpoints will respond with HTML. This will be possible by using HTMX, a small JavaScript library that makes asynchronous calls to a web server and replaces appropriate sections of the DOM with the server's response. This removes the need for a client-side JavaScript framework and reduces the overall complexity of the application by ensuring all application state remains on the server.
Startup The server software will run on what is essentially an infinite loop, waiting for HTTP requests. The true system, however, will start up when a client makes a request. If that request is targeted at a valid route, the request will be processed by a controller. If not, the system will return a 404 error.
Errors All exceptions will be collected in an error log for additional analysis by developers. If the exception is severe enough that it fails to process a client request properly, it will be returned to the client as an error message with an HTTP status code.
Shut down When a client ends a session, the system will ensure all active database connections used in client requests are closed and that any ongoing transactions are committed to the database.
Controllers
Models
Analyzers and Analysis
ISS: International Space Station
IMS: Inventory Management System - Barrios' ISS inventory management datastore
ORM: Object Relational Mapper - A framework-specific set of classes and methods for interacting with a database
ASGI: Asynchronous Server Gateway Interface - An asynchronous calling convention for Python web servers
HTTP: Hypertext Transfer Protocol - The messaging and networking protocol used to request and deliver hypermedia on the web
HTML: Hypertext Markup Language - A language used to define hypertext documents for use by a web browser
DOM: The Document Object Model - A data structure representing an HTML document and all of the elements contained therein.
PostgreSQL: A popular relational database
