How does transmission technology work?

‘Pieces’ of video

When it comes to video content, whether it is a live broadcast or a pre -recorded video, there are two main challenges to address. First, video data has a large size, which makes it a long time to transmit from the source to devices such as televisions, computers, tablets and smartphones.

Second, the transmission must be adaptive to accommodate the differences in user devices and internet capabilities. For example, spectators with lower resolution or slower Internet speeds must still be able to watch a given video, although in lower quality, while those with higher resolution screens and faster connections enjoy the best possible quality.

To address these challenges, video suppliers implement a series of optimizations. The first step involves fragmenting videos in smaller pieces, commonly known as “pieces.” These fragments are subjected to a process called “Coding and Compression”, which optimizes the video for different resolutions and bits rates to adapt to various network devices and conditions.

When a user requests a video on request, the system dynamically selects the appropriate flow of fragments according to the user’s device capabilities, such as screen resolution and current Internet speed. The video player in the user’s device assembles and reproduces these pieces in sequence to create a perfect visualization experience.

For users with slower Internet connections, the system offers lower quality fragments to guarantee reproduction without problems. That is why you can notice a drop in video quality when your connection speed is reduced. Similarly, if the video stops during playback, it is generally due to the fact that your player is waiting to cushion additional fragments of the supplier.

Video transmissions reach users at different levels of quality depending on the user device and the Internet connection. Chetan Jaiswal

Deal with distance and congestion

The delivery of large -scale video content, whether pre -recorded or live, raises a significant challenge when the immense number of videos consumed worldwide is extrapolated. Transmission services such as YouTube, Hulu and Netflix organize huge content libraries on request, while simultaneously managing innumerable live broadcasts worldwide.

A seemingly simple approach to deliver video content would imply building a massive data center to store all videos and related content, then transmit them to users around the world through the Internet. However, this method should not be favor because it comes with significant challenges.

An important problem is geographical latency, where a user’s location in relation to the data center affects the delay he experiences. For example, if a data center is located in Virginia, a user in Washington, DC, would experience a minimum delay, while a user in Australia would face much longer delays due to the greatest distance and the need for data to cross multiple interconnected networks. This added travel time slows the delivery of content.

Another problem is the congestion of the network. As more users worldwide connect to the central data center, interconnection networks become increasingly occupied, resulting in frustrating delays and video buffer storage. In addition, when the same video is simultaneously sent to multiple users, duplicate data traveling through the same Internet links waste bandwidth and further congests the network.

A centralized data center also creates a single failure point. If the data center experiences an interruption, no user can access its content, which leads to a complete service interruption.

Content delivery networks

To address these challenges, most content suppliers trust content delivery networks. These networks distribute content through dispersed points globally, which are groups of servers that store copies of high demand content locally. This approach significantly reduces latency and improves reliability.

The suppliers of the content delivery network, such as Akamai and Edgio, implement two main strategies for the implementation of presence points.

The first is “Enter deep“Approach, where thousands of smaller presence knit nodes closely closer to users, often within the Internet service providers networks. This guarantees minimal latency when the content is closer as close as possible to the end user.

This diagram, with the backbone of the Internet at the top and users at the bottom, shows the ‘deep’ approach to place the ‘deep’ content delivery servers on the network, near the users. Chetan Jaiswal

The second strategy is “Bring home“, Which implies the implementation of hundreds of groups of larger presence points in strategic locations, generally where the ISPs are interconnected: Internet exchange points. Although these groups are further from users than in the Enter Deep approach, they have a larger capacity, which allows them to handle higher traffic volumes efficiently.

This diagram, with the backbone of the Internet at the top and users at the bottom, shows the ‘bring home’ approach to place content delivery servers between regional internet service providers and the trunk network. Chetan Jaiswal

Infrastructure for a connected world

Both strategies aim to optimize the transmission of videos reducing delays, minimizing bandwidth waste and guaranteeing a perfect visualization experience for users around the world.

The rapid expansion of the Internet and the increase in video transmission, both live and request, have transformed how the video content is given to users worldwide. However, the challenges of handling massive amounts of video data, reducing geographical latency and accommodating different user devices and Internet speeds require sophisticated solutions.

Content delivery networks have become a cornerstone of modern transmission, allowing efficient and reliable video delivery. This infrastructure supports the growing demand for high quality videos and highlights the innovative approaches necessary to meet the expectations of a connected world.

By Cetan Jaiswal, associate professor of computer science, University of Quinnipiac. This article is published again from the conversation under a Creative Commons license. Read the Original article.