things and scrap them if they don’t work easily or reduce capital expenditure. Note that it’s not all roses in the cloud; it can be expensive in certain areas. This must be managed.” NINE BILLION TELEMETRY DATA POINTS “People can get caught up in not looking at the total cost of ownership (TCO),” continues Brooks. “What is the initial cost? What is the implementation cost? What’s the operating cost? “There are a number of areas ripe for savings; for instance, the use of advanced analytics and machine learning. Look at just one area – root cause analysis (RCA). This is a big expense for broadcast channels. On average, it costs about $500,000 a year in RCA per broadcast channel. That’s a big number. “We’re doing multi-object correlation analytics, analysing over nine billion telemetry data points every day across all our customers. If there is an incident, meaning you might have unrecovered packets, one of your redundant streams is showing errors or is offline, we will produce that RCA by observing all the objects, correlating and determining what the incident cause was. This displays all the graphs and logs in one place, all time-synchronised for that incident. These things
save you much more than you would ever spend on another solution. “You can be blind when trying to find a fault if something goes wrong or have visibility of everything across the entire process. If there’s an issue, the system will inform you – finding where and what it is. Or if there’s a potential issue, it will use machine-learning anomaly detection to predict them before they happen. “We support 18 protocols, so if you want to use something else, you can. The Zixi protocol is proprietary, but it’s the only service with ARQ encryption and forward error correction. We have by far the lowest latency, since you can’t get to sub- second latency without forward error correction. “Because we have been doing this for so long, we also make it easy to manage large organisations with few people, and learn quickly. Operations and engineering costs are considerable; we’re focused on getting those costs down,” concludes Brooks. V17 of the Zixi core software drastically improves processing capacity, requiring 80% less computing while also reducing transport stream egress cost by up to 50%. This lets clients maximise resources, minimise expenses and deliver broadcast-quality video reliably – with profound impact on the TCO of live video streaming implementations.
@feedzinesocial
Powered by FlippingBook