Handling large volumes of diverse data

Visual Weather efficiently ingests, processes, and visualises large volumes of diverse data, from observation reports and model data to satellite and radar imagery. With advanced data handling technologies, it supports high-resolution models, multi-member ensembles, and custom formats. Offering flexible integration via FTP, HTTP, and more, Visual Weather ensures seamless processing, display, and delivery of data through multiple channels, all while optimising performance and user experience.

Visual Weather ingests and displays all your data!

Visual Weather can receive, store, and process diverse data types from various sources, such as observation data (SYNOP, METAR, AWS), model data (including ensemble models), satellite and radar imagery, lightning data, wind profiler readings, and more. Advanced technologies ensure that you can efficiently handle and display a growing volume of high-resolution model data, multi-member ensemble models, and satellite data. Additionally, Visual Weather produces output in various formats and delivers it through multiple channels.

Visual Weather contains several technologies and techniques to improve data handling with a focus on ease of use, bandwidth friendliness and display without waiting:

Switching Subsystem

Visual Weather embeds a simplified switching subsystem stemming from the Moving Weather. The switching subsystem brings versatility and flexibility in the number and variety of data sources to be ingested and visualised. Visual Weather can directly ingest data from FTP/FTPS, SFTP, SCP, HTTP/HTTPS or WMO TCP Socket.
It understands various data formats according to the WMO Manual on Codes (WMO Publication No. 306), incl. binary codes (GRIB, GRIB2, BUFR).

Custom Formats

Visual Weather supports the extensibility of its data processing part on several levels:

Pre-processing support in the switching subsystem allows you to adjust or convert data on input to a format that Visual Weather understands
Geospatial and Temporal Entity (GATE) is Visual Weather's data access feature. It allows users to transform and classify custom data for processing in Visual Weather's data access computational part, known as the Mathematical Kernel.
Python API allows custom decoding and visualisation plugins to be written.

Hybrid Database

A hybrid database is a unique client-side caching database designed to improve data transmission in high-latency networks with bandwidth efficiency. The minimal number of interactions between client and server ensures good horizontal scalability and enables the use of hybrid clients across WAN networks with limited bandwidth and high latency. The hybrid database features so-called subscriptions to trigger automatic data synchronisation, avoiding delays in user data access. Locally downloaded data and indices allow the application to run even in offline mode, although only a limited data subset is available.

Super Cache

Super Cache improves the display of NWP-processed data (for instance, satellites, meteograms, model ascents) by persistently storing decoded cached content on Visual Weather clients or servers. Storing cached content allows applications to access decoded data without repeatedly loading and decoding them from the database.