Gaia is an approved ESA cornerstone project, currently scheduled for launch in late 2011. Gaia will provide photometric, positional, spectroscopic and radial velocity measurements with the accuracies needed to produce a stereoscopic and kinematic census of about one billion stars in our Galaxy and throughout the Local Group, addressing its core science goals to quantify the formation and assembly history of a large spiral galaxy, the Milky Way. Gaia will achieve this by obtaining a six-dimensional (spatial & kinematic) phase-space map of the Galaxy, complemented by an optimised high-spatial resolution multi-colour photometric survey, and the largest stellar spectroscopic and radial velocity surveys ever made. The Gaia data set will be constructed from 2 × 10 12 observations (image CCD transits), whose analysis is a very complex task, involving both real-time (this proposal) and end-of-mission data products. This paper describes the UK Gaia Data Flow System activities as part of the emerging European wide Gaia Data Processing system. We describe the data processing challenges that need to be overcome to meet the heavy demands placed by Gaia. We note the construction processes required to handle the photometric reduction of the data from Gaia's 100+ focal plane CCDs, the pipeline needed to support the 'science alerts' and epoch photometry handling, and the spectroscopic processing system. We note the system software and hardware architecture, and how the data products will be generated to ensure compliance with emerging VO standards.