The cutting-edge computer architecture that’s changing the AI game (VB Live)

Presented by SambaNova

To stay on top of AI innovation, it’s time to upgrade from multicore architecture. Join this VB Live event to learn how cutting-edge computer architecture can unlock new AI capabilities, from common use cases to real-world case studies and more.

Register here for free.

AI and machine learning demand new approaches to computer architecture — but, of course, there are more factors. Large amounts of data, the arrival of industry-standard frameworks such as TensorFlow and PyTorch, and the death of Moore’s Law, are all signs that it’s time for the next generation of computing systems. And it’s one of the biggest transitions that the computer industry has seen since the changes demanded by the Internet and online connectivity.

The new wave of computer architecture is being driven by three main issues. First, data centers are growing larger, the amount of data that needs to be processed is growing exponentially, and compute is getting more expensive, which means companies need new more effective, powerful, and efficient architectures for data processing.

The second is the difficulty — in time, expense, and resources — of turning that massive amount of data into actual value for a business. The companies that manage this transmutation will have a dramatic competitive edge over the ones that are falling behind.

Third, applications are evolving in sophistication and ability, and companies want the computing architecture that allows them to take advantage of these new possibilities.

It’s important to consider that AI apps are largely developed outside of the IT organization, so the ‘persona’ that develops the AI/ML apps is often different than the IT person who will provide infrastructure services to support those apps. Given this, the AI apps owner may care more about how the model performs and less about the IT & compute components (whereas the IT person providing services and support to the AI/ML app owner will care about the compute infrastructure). This creates an interesting dynamic in the way that AI innovation is developing.

In short, it’s about ease of development, ease of deployment, and ease of creating value, faster. It’s about enabling companies to do things that they can’t do today.

For the last thirty years, IT leaders have been focused on optimizing for instructions and operations. But right now data and how it flows through a system is driving performance, and hardware and software has been evolving to support it. However, while most of the architectures available today can handle current AI and deep learning functionality, the issue is about future-proofing. Computing is changing fast, and organizations need to be prepared to take advantage of the advances that continue to emerge.

At the core, computer chips are still the foundation, but to function in the new world of computing, the entire system needs to integrate across several layers of new hardware and software, end to end, from the algorithms to the silicon. And it needs to be available to all companies managing the shift into a new kind of computing.

Computing has been limited by the standardization of x86 and the GPU over the past ten to twenty years. The hardware portion has been commoditized, and then the software standardized, so computer software and hardware architecture has had little space to grow or innovate. New, purpose-built architecture can expand the horizons of capability for the future of machine learning, AI, and development, freeing users, developers, and applications from the constraints of legacy architectures.

To learn more about the ways advanced, modern architecture can accelerate everything from recommendation engines, NLP model deployment, computer vision, and more, plus the factors companies need to consider when future-proofing their IT centers and more, don’t miss this VB Live event.

Don’t miss out!

Register here for free.

You’ll learn:

  • Why multicore architecture is on its last legs, and how new, advanced computer architectures are changing the game
  • How to implement state-of-the-art converged training and inference solutions
  • New ways to accelerate data analytics and scientific computing applications in the same accelerator

Speakers:

  • Marshall Choy, VP of Product, SambaNova Systems

 More speakers to be announced soon!

Source: Read Full Article