High Performance Computing: A new direction to Media & Entertainment Industry | HCL Blogs

High Performance Computing: A new direction to Media & Entertainment Industry

High Performance Computing: A new direction to Media & Entertainment Industry
April 19, 2018

In today’s fast-moving world, consumers are simply switching media consumption from ‘analog’ to ‘digital.’ The greatest change is that the people who used to consume news, entertainment, and other information via printed periodicals, television sets, and radio sets are now consuming the same via personal computers, tablet computers, and smartphones. Advancements in mobile, video, and wireless technologies have ignited an explosion in the growth of streaming services.

As per PWC Media and Entertainment worldwide revenues are expected to rise at a CAGR of 4.4% over the next five years, from $1.7 trillion in 2015 to $2.1 trillion in 2020.

Thenextweb.com states that internet users have grown by around 82% (1.7 billion people) since January 2012. More than 1.3 billion people have started using social media – that’s a rise of 88% in just five years, and equates to more than eight new users every second. More than 864 million people have started using social platforms via a mobile device in the past 24 months at a rate of almost 14 new users every second.

The way a product becomes mature in the market and gets attention from the consumer has changed tremendously due to the advent of mobile applications and digitization. For example: Airlines took 68 years to gain 50 million users whereas a mobile app, Pokémon Go, took only nine days to cross the 50-million-user mark, as stated on Steemit.com.

In the M&E industry, various software (used for 3D modelling and animation, web and graphic design, VFX and gaming, filmmaking and more) demand devices with powerful processors with compatible RAM and ROM. Applications like video rendering and high resolution image editing processes require high network performance, fast storage, large amounts of memory, very high computing capabilities, or a combination of all of these, which we call High Performance Compute (HPC).

During the rendering process, the render engine calculates/computes details like ray trace settings, resolution settings, texture, lights, etc., using the scene descriptions and utilizes 98-100% of the resource in the compute nodes – like CPU and memory. Rendering takes a long time on normal computers or workstations, whereas HPC devices expedite the rendering process and completes the same task in few hours which would have taken days. HPC uses a ‘divide and conquer’ approach that leverages a parallel-processing architecture, where very large analytical problems are parsed into many smaller processing jobs and run on individual compute nodes simultaneously.

In general, HPC refers to -

The practice of aggregating computing power in a way that delivers much higher performance than one could get out of a typical desktop computer or workstation to solve large problems in science engineering or business. HPC often refers to the individual computers in a cluster as nodes. For a small business, HPC could have as few as four nodes or 16 cores. A common cluster size in many businesses is between 16 and 64 nodes, or from 64 to 256 cores.

A high performance computer overcomes the limitation of resources by aggregating the computation power of several units.

A high performance computer ensures that the individual nodes can work together to solve a problem larger than any one computer can easily solve. And, just like people, the nodes need to be able to talk to one another in order to work meaningfully together.

Just like your desktop or laptop, your HPC cluster won’t run without a software. Two of the most popular choices in HPC are Linux (in many varieties) and Windows. Your choice of operating system should really be driven by the kind of applications you need to run on your high performance computer.

HPC is a specialized discipline which uses big computers to simulate real-life processes. HPC services come as a boon to those industries which demand enormous amount of computing resources, and M&E is one of them. Earlier, HPC technology would cost fortunes to an enterprise, but with the advent of new Intel chipsets and the overall economies of scale in the IT sector, HPC is a commercially viable solution, which will prove to be a boon for industries like M&E which await these changes.

The HPC market has the potential for an exciting future. Now all that remains for businesses is to create the right blend of talent and technology to turn that prospect into a commercial reality. The HPC industry will survive and prosper only if it can find mass markets for more modestly parallel systems. The future will certainly be interesting and the challenges for the parallel software industry are significant – but potentially offer greater rewards for those companies who can get it right.