NASDAQ Market Disclosures

Similarly, when id3as was approached to optimize the NASDAQ market disclosure workflow (Thomson Reuters prior to acquisition by NASDAQ), we initially saw a very similar workflow to the early Prime Minister's Questions model. Partly because these workflows had evolved organically from the initial delivery of a live stream and had followed a more traditional broadcast architecture - where large video files would be moved around to be “treated” (processed) by rare and expensive hardware - their workflow also suffered considerable delays between the end of an event and the availability of the catch-up offering.

Our approach was to centralize the delivery of the high bandwidth into a cloud model and to move the capability to video. The software capability - particularly with our highly virtualized model - was comparatively small, and because it was not tied to a physical unit, it was much faster to bring many encoders to the virtual end point of the source video in the cloud, perform all the transcoding into the 38+ formats that NASDAQ deliver, to provide a web-based UI (see below) to top-and-tail the archive - which itself simply provided an edit decision list (EDL)

NASDAQ web-based EDL editor and “cutting interface''

Figure 3.4 NASDAQ web-based EDL editor and “cutting interface''.

to mark up all the various archives, and to cut and top-and-tail the various archives instantly. The final archives are then initially made available on an internal point of origin within the cloud, from where the content delivery network can “pull” and deliver them. The NASDAQ turnaround for completed archives is now a few seconds from the end of the event. This transformed the speed at which their catch-up service could be offered, and has been a benchmark for the financial disclosure market ever since (Figure 3.4).

< Prev   CONTENTS   Source   Next >