Our Technology

Better AI Starts Here

The demand for affordable and scalable computational capacity is increasing at a rapid pace. This trend is largely being driven by efforts to introduce Artificial Intelligence (AI) into virtually every aspect of modern life and fueled by advancements in the increasingly complex Machine Learning (ML) algorithms that support these initiatives.

Here’s the Challenge: AI and Machine Learning innovation has quickly outpaced the ability of traditional CPU and GPU platforms to deliver efficient, affordable and scalable resources. In addition to energy and resource implications, current platform deficiencies often result in workload processing delays and long network latencies that prevent these breakthrough applications from becoming truly agile, scalable and responsive.

Vaerus Technology is the Answer: Vaerus is focused on accelerating AI adoption through our innovative platforms that are optimized for Machine Learning workloads. The Vaerus Intelligent Computing Experience (VICE)™ platform is designed to provide scalable, on-demand capacity, while eliminating performance bottlenecks and minimizing energy usage. Our mission is to help unlock the potential of AI and Machine Learning — thus, enhancing lives while also being mindful of precious resources.

Double Rack Image

THE VAERUS INTELLIGENT COMPUTING EXPERIENCE (VICE)™ ENABLES EFFICIENT, SECURE AND RELIABLE ON-DEMAND SCALING OF COMPUTATIONAL AND DATA MANAGEMENT RESOURCES

VICE™ Platform Advantages

Low Latency

On Demand Scalability

Application Optimized

Highly Secure

Zero Downtime

Scalability is at the Core of Our Technology

To meet the massive increase in demand for AI, technology must scale efficiently, and adhere to strict latency and performance requirements…. all while keeping costs and power consumption in check.

The innovative Vaerus Intelligent Computing Experience (VICE)™ platform is designed from the ground up to be scalable. We’ve also built it with an eye towards providing customers with superior results and a world class experience. Our platforms are developed using customized Programmable Accelerators (FPGAs) and Application-Specific Integrated Circuits (ASICs) that are optimized for high-bandwidth workloads, eliminating the performance bottlenecks associated with traditional computational platforms. This approach provides ~10-20X gain in performance as well as price/performance versus a traditional GPU.

Efficient scaling of the VICE platform is enabled through our patent pending Vaerus Warp-Speed (VWS)™ interconnect technology. This technology is enabled through a revolutionary 2D network-on-chip (2D NoC) together with an array of Machine Learning processors (MLPs) optimized for high-bandwidth and AI/Machine Learning workloads, high-bandwidth GDDR6 interfaces, 400G Ethernet and PCI Express Gen5 ports — all interconnected to deliver ASIC-class performance while retaining the full programmability of FPGAs.

While our focus is on delivering a best-in-class computational and data management platform, we also use our extensive application acceleration expertise to help partners optimize their applications and workloads for our revolutionary platforms.

VAERUS WARP SPEED (VWS)™ TECHNOLOGY ENABLES EFFICIENT SCALING OF THE VICE™ PLATFORM TO ACCELERATE AI/ML WORKLOADS

INTELLIGENT COMPUTING AS A SERVICE

Better AI Starts Here