Telemetry depot and web server for Open MCT, developed for the HYPSO student satellite project. The project report documenting the design and implementation of MCT Depot is available here.
MCT Depot inputs and outputs
.svg)
Early high-level system block diagram
.svg)
System data flow diagram (level 0)
Class diagram
The main step here is cloning or downloading the contents of this repository, and running npm install inside the folder for it. npm run start will start the full MCT Depot server, while npm run test can be used to run the test suites for it while doing development.
MCT Depot is fairly I/O-bound, as it does little processing after telemetry samples have been stored in its database but often needs to supply large amounts of data when Open MCT is rendering complex combined views, or views with large amounts of historical data. It is thus recommended to keep the SQLite database on a fairly fast drive (such as an SSD), with at least a few gigabytes of free space set aside per month of telemetry data it should handle (although this depends heavily on the detail level of the telemetry data stored).
As for non-I/O requirements, it should run alright on most things; 512 MB of free memory should in most instances be enough for the application itself, along with two or more CPU threads offering alright single-threaded performance. Ubuntu Server or a similar flavour of Linux is recommended for hosting it, but it should run fine on any operating system that can host Node (including Windows).
- 512 MB or more of free memory
- 1 GB or more of free space
- 1 GB or more of free memory
- 10 GB or more of free space on a fast drive, with 2-3 GB extra per month of telemetry samples depending on size per sample and sample rate
- Two or more available CPU threads with good single-threaded performance