Latency Zero: The Hardware Pact That Promises to Re-Wire AI's Reality

Three of the world’s biggest technology providers have combined to offer a private AI solution that enables businesses to train AI models in scalable, cost-efficient public and private clouds while ensuring enhanced control, security, and low-latency deployment on-premises.

The partnership between Equinix, Dell, and NVIDIA will enable rack-scale AI deployments with advanced storage and networking performance.

The partnership includes the hardware and the data preparation required to execute AI projects within the Equinix data center environment of more than 260 centers worldwide. 

The goal is to demystify AI so that it becomes another workload, albeit high performance and the partnership is a solution response to the issues many organizations are having with AI deployments.

Many are challenged in leveraging data effectively to feed AI models in a neutral, cloud-adjacent platform where customers can securely and cost-effectively connect to public clouds, colocation facilities, and their own private cloud and on-premises infrastructure.

“Our collaboration with Dell Technologies and NVIDIA enables enterprises to harness the power of generative AI while maintaining control over their data and supporting their corporate sustainability goals,” said Lisa Miller, senior vice president for platform alliances at Equinix.

“This solution is designed to support the most demanding AI workloads and help ensure our customers can innovate and drive outcomes.”

Healthcare case studies

The Dell AI Factory with NVIDIA has helped customers in industries such as healthcare deliver tangible results.

Streamlining radiology reports has allowed for more focus on patient care, in manufacturing where predictive maintenance and quality control has enhanced operational efficiency.

In Hong Kong, Equinix has also announced a new collaboration with AiHPC to launch an AI healthcare platform, Orchestration AI (OrchAI), designed to enhance digital capabilities in the healthcare sector.

“With the performance, flexibility, and scalability of the new GPU cluster, Continental improved AI training time by 70%.”

The collaboration will see OrchAI set up on Platform Equinix, fostering advancements in healthcare and life sciences research and development, clinical trials, and the application of advanced biomedical technologies within Hong Kong.

The initiative aims to attract innovative enterprises and research organizations from around the globe to establish operations in the city.

As medical research institutions increasingly invest in AI and high-performance computing (HPC) infrastructure, complex analyses often encounter obstacles related to computational power and data exchange. 

Dr. Sam Chu, founder of AiHPC, highlighted the need for new healthcare solutions given Hong Kong's demographic challenges and aspirations to become an international hub. “Our objective of establishing Hong Kong's first integrated AI and HPC ecosystem is to streamline academic research, healthcare implementation, and cross-border collaboration between Hong Kong, the Greater Bay Area, and the world,” he said.

Autonomous vehicles

Equinix has also worked with other partners, such as IBM, to assist organizations deliver value from AI projects.

In the development of autonomous vehicles, solutions provider Continental’s Advanced Driver Assistance Systems (ADAS) team needed to process more than 150 terabytes (TB) of data to inform design decisions that would increase connected and autonomous vehicle safety. 

The team first wanted to pull image and sensor data from vehicles in different geographies including Europe, America, and Asia but a critical priority was then to store and process it in a connected central repository quickly accessible by hundreds of engineers worldwide.

Continental leveraged carbon-neutral AI infrastructure such as Platform Equinix to build and interconnect its AI-driven NVIDIA graphics processing unit cluster and IBM Elastic Storage System 3000—reducing AI training time to augment safety standards from weeks to days.

This approach, featuring real-time data access for its scalable, future-proof AI, boosted performance to accelerate deployment, enable more experiments with secure data protection, and enhance data privacy controls.

“With the performance, flexibility, and scalability of the new GPU cluster, Continental improved AI training time by 70%,” said Balázs Lóránd, head of AI Development Centre at Budapest, Continental AG, Business Area Autonomous Mobility.

Image credit: iStockphoto/panumas nikomkai