All Blog Posts’s Vision for a More Powerful Kind of Cloud Computing

Avi Barliya
Avi Barliya

Consider the fact that more than 3.3 billion people or about 35% of the world’s population now have smartphones. Then, think about all those smartphones combined with the infinite number of laptops, desktops, IoT devices, servers, minicomputers, and mainframes that are in use today. That’s a tremendous amount of compute power. 

Given how prolific computers are today, it’s nearly impossible to imagine how anyone could think that the world would only need the availability of five computers. Yet in 1943, that’s exactly what Thomas Watson, the president of IBM, thought would be necessary. 

Photo of Thomas Watson and Quote

A New Vision for Cloud Computing

As the founder of, I envision a world of computing that offers a new way of realizing Watson’s projection. Rather than 5 computers spread across the world, companies would rely on a small number of powerful computing systems to run their workloads. I see something more akin to a complex organism rather than a single cell. 

Right now, there is a massive amount of unused compute power sitting idle while their owners are sleeping or doing something else. I believe that when viewed as a collective resource, these individual devices can produce a new and improved type of computing experience. 

What if there was a way organizations could capitalize on the spare compute power of all those devices? This is exactly what is building. In fact, it is so much more. Individual devices may be online all the time, but most are rarely running continuously at peak capacity. There is a great deal of underutilized computing capacity that exists today.

It’s one of the areas where is focusing. We are developing a computing platform for tapping into excess compute capacity to run enterprise workloads. It’s the next step in the evolution of how users experience computing.

This image is conveying serverless computing with multiple devices.

Cloud Computing with

What cloud computing users often need, especially enterprise customers, is a vendor-neutral, serverless computing platform. This is what provides. Specifically, lets companies run code on a compute platform as if it were running on a local desktop. As an added benefit, if an application needs to be containerized, it will be. The platform runs on bare metal, as well as on AWS. Support for Microsoft Azure and the Google Cloud Platform is on the horizon. 

You may be asking “just how does it work?” Agents get installed on each computing device across a decentralized network. Then the Smart Scheduler directs compute jobs to the decentralized network of compute nodes that, together, will provide the necessary compute power, as the image below conveys.

Image of How Disco Smart Scheduler Works is changing the way businesses run workloads, both on-premises and in the cloud, with a platform that hides the underlying details of hardware and cloud services and provides a common way to utilize compute resources wherever they are available. The future of will include accessing those billions of underutilized smartphones, tablets, and more.

Applying the model

The distributed and decentralized computing approach by is powerful. Let’s take a look at a specific use case – intelligent video processing at scale. A video app needs to process videos on-demand. They average 3,000 video uploads per day. The videos are about 30 seconds in length which equates to 375 hours of required compute time to process all 3,000 videos at a speed of 2 frames per second. That’s potentially an improvement in speed of 100x! is an ideal solution for compressing the amount of time it takes to process each video. Once a customer splits a video into slices and sends it to, the system will optimize the processing of those slices of videos in parallel across multiple CPUs and GPUs. Once the processing is complete, the video slices will be saved to a zip file and returned to the customer.

Validation of Video Processing Hours Mentioned in Text

That’s just one example of a use case. There are more, such as processing the massive volumes of data generated from self-driven cars, producing on-demand analytics to assess the in-store experience of customers, movie CGI, 3D rendering, and volumetric videos for AR/VR/XR.

It’s an exciting time as we witness the acceleration of intelligent compute processing power to levels that could not have been imagined by Thomas Watson. Stay tuned as transforms the landscape of compute to further the evolution that began 50 years ago when compute was first introduced.

Learn more about and Avi’s vision for the future of cloud computing in this video.


Avi Barliya
Avi Barliya

Most recent articles

How Does Work? provides an easy and cost-effective way to use multiple machines to make compute jobs faster. Whether you need to…

Yuval Greenfield
Yuval Greenfield’s Vision for a More Powerful Kind of Cloud Computing

Consider the fact that more than 3.3 billion people or about 35% of the world’s population now have smartphones. Then,…

Avi Barliya
Avi Barliya

What Technology Tools Can Help Data Scientists?

To say the field of Data Science is exploding is truly an understatement. It only takes one statistic to be…

esther carlson
esther carlson

Serverless Computing vs. Cloud Computing: What’s the Difference?

When IT pros first see or hear the phrase serverless computing, their minds go in multiple directions. Is it true…

esther carlson
esther carlson

What is Serverless Computing?

How many developers would prefer spending more of their time building new products that create value rather than configuring and…

esther carlson
esther carlson

What is DevOps and Why You Should Care

Customers have set the bar high when it comes to their expectations for speed and responsiveness, on-demand availability, easy access…

esther carlson
esther carlson