2020 Volume 15 Pages 1-17
We have upgraded the JVO ALMA WebQL web service - available through the JVO ALMA FITS archive - to include legacy data from other telescopes, for example Nobeyama NRO45M in Japan. The updated server software has been renamed FITSWebQL. In addition, a standalone desktop version, independent of the Japanese Virtual Observatory, that runs on local end-user computers (laptops, desktops) and supports Linux, macOS and Windows 10 Linux Subsystem (Bash on Windows) is also available for download from https://github.com/jvo203/fits_web_ql.
The FITSWebQL server enables viewing of even over 100 GB-large FITS files in a web browser running on a PC with a limited amount of RAM. Users can interactively zoom-in to selected areas of interest with the corresponding frequency spectrum being calculated on the server in near real-time. Hence at a glance users gain access to large easy-to-see images as well as the corresponding frequency spectra that are visible both at the same time. The client (a browser) is a JavaScript application built on WebSockets, HTML5, WebGL, SVG and WebAssembly.
The new FITSWebQL lays foundations towards supporting an interactive preview of terabyte-class FITS files. In addition to handling larger files, FITSWebQL also improves support for real-time spectrum updates through tackling the latency problem in two ways. First, it introduces an adaptive frame rate control: monitoring the network latency and local web browser responsiveness, and reducing the frames-per-second (FPS) rate as and when necessary. Secondly, our software tracks in real time end user's mouse movements with the Kalman Filter, which is then used to predict the future target mouse position after taking into account the network latency and server-side computation time. Hence we speculatively deliver ahead-of-time real-time spectrum data for positions where the user is likely to be looking at. The new version also allows users to view multiple FITS files simultaneously either in a special RGB composite mode (presently limited to NRO45M FUGIN only), where each dataset is assigned one RGB channel to form a single colour image, or as separate image tiles. Spectra from multiple FITS cubes are shown together too.
The paper describes the main features of FITSWebQL and its technical architecture. We also touch on some of the recent developments, such as a switch from C/C++ to Rust (see https://www.rust-lang.org/) for improved stability, better memory management and “fearless concurrency”, displaying FITS data cubes in the form of interactive on-demand video streams in a web browser or efforts to incorporate machine learning for an improved user experience.
Since version 3 FITSWebQL has had a robust choice of different colourmaps, automatic integration with the Splatalogue molecular database, fixed and variable (auto-scaled) Y-Axis, synthesised beam overlay, manual reference frequency/source velocity corrections as well as a 3D viewing mode and client-side contouring handled in JavaScript.
