In at present’s fast-paced digital panorama, the flexibility to harness the ability of synthetic intelligence (AI) is crucial. At Cisco, we’re empowering organizations worldwide to seamlessly combine AI into their enterprise operations, making certain they continue to be on the forefront of innovation and safety.  Our collaboration with NVIDIA will mark a big milestone on this journey, introducing a groundbreaking new AI infrastructure that places safety and suppleness at its core.

A Strategic Leap Ahead 

As we introduced earlier this yr, we’re working with NVIDIA to ship a strategic leap ahead in AI infrastructure. Collectively, we plan to remodel the best way corporations deploy, function and safe AI Factories—information facilities meticulously designed to energy AI workloads. Our co-optimized structure will allow organizations to tailor their infrastructure exactly to their AI wants, with out compromising on operational simplicity or safety. 

Crucial Flexibility and Agility 

Our partnership with NVIDIA is predicted to supply organizations unmatched flexibility and agility, embedding safety from the appliance and the workload to the underlying infrastructure. Whether or not you’re at first of your AI journey or seeking to scale present infrastructure, Cisco Safe AI Manufacturing unit with NVIDIA will present scalable, high-performance options that adapt to your distinctive necessities. With customizable modular elements and confirmed reference architectures, we intend to empower you to design and deploy AI infrastructure that aligns seamlessly with your enterprise objectives. 

The vertically built-in deployment choice can be primarily based on Cisco Nexus Hyperfabric AI, and can embody: 

• Cisco 6000 Collection Switches, utilizing Cisco Silicon One and cloud-managed by the Cisco Nexus Hyperfabric AI controller. 
• Accelerated compute with Cisco UCS C885A M8, primarily based on the NVIDIA HGX platform with NVIDIA H200 GPUs 
• NVIDIA BlueField-3 DPUs and SuperNICs 
• NVIDIA AI Enterprise software program platform 
• VAST Information Storage 
• Cisco Optics 

The modular deployment choice will characteristic customizable elements, together with:   

• Accelerated compute with Cisco UCS C885A M8 and C845A M8 servers, primarily based on the NVIDIA HGX and MGX platforms with NVIDIA H200 GPUs, and managed by Cisco Intersight  
• NVIDIA BlueField-3 DPUs and SuperNICs 
• Superior networking with Cisco Nexus 9000 Collection switches utilizing Cisco Silicon One and managed by Cisco Nexus Dashboard  
• Future Cisco switches with NVIDIA Spectrum-X, managed by Nexus Dashboard 
• NVIDIA AI Enterprise software program platform 
• Excessive-performance storage from NVIDIA-Licensed™ companions Pure Storage, Hitachi Vantara, NetApp, and VAST Information 
• Open Supply containerization and automation options from Purple Hat to optimize AI and containerized workloads 
• Cisco Optics 

To scale back AI funding dangers, Cisco will provide options validated and primarily based on NVIDIA Enterprise Reference Architectures enabling quicker, versatile, and scalable deployments with high-performance AI infrastructure. Cisco Safe AI Manufacturing unit with NVIDIA helps enterprise prospects to operationalize and safe a sturdy AI infrastructure for information engineering, AI coaching and mannequin customization, AI pipeline safety, AI inferencing, and compliance for AI fashions with use circumstances throughout industries together with public sector, manufacturing, finance, healthcare, and retail.   

Safety at Each Layer  

AI infrastructure is inherently advanced, and securing it requires a brand new, holistic method. That’s why our structure embeds safety at each layer of the AI stack. With options like Cisco Isovalent, Cisco Safe Firewall, Cisco Hypershield  and Cisco AI Protection, we mechanically broaden and adapt safety measures as infrastructure modifications, offering safety in opposition to evolving threats.  This additionally contains the alternative to combine with NVIDIA BlueField-3 DPUs to enhance safety throughout the stack with real-time AI workload risk detection with NVIDIA DOCA AppShield. 

A Unified Imaginative and prescient for the Future 

Our journey with NVIDIA in direction of a validated and unified structure is shifting swiftly. We’re proud to be working towards options that cater to the evolving calls for of AI factories, making certain you’ve the appropriate instruments to thrive within the AI period. Our versatile deployment fashions, together with the ready-to-deploy Cisco Safe AI Manufacturing unit with Nexus Hyperfabric AI or modular options, set a brand new normal for consistency and innovation. 

 In accordance with our latest State of AI Safety report, organizations that strategically deal with each infrastructure and safety challenges concurrently are extra agile, scale quicker, and derive enterprise worth faster. By integrating accelerated compute, high-performance networking, software program, storage, and industry-leading safety, we allow you to unlock the total potential of AI, driving transformative outcomes for your enterprise. 

The subsequent era of worldwide connectivity with 5G Superior and 6G networks is being developed now, and it’s vital that the {industry} work collectively to foster innovation. Cisco is main the cost in growing cellular core, vital infrastructure, and safety expertise that’s already accelerating AI innovation. By working with NVIDIA and the broader ecosystem to construct AI for Wi-fi into next-gen networks, we are going to allow dramatic enhancements in efficiency, reliability, and safety that can profit customers and enterprise prospects, whereas bringing the world nearer collectively.   

At Cisco, we consider within the energy of collaboration and innovation to create a extra related and inclusive future for all. Our increasing partnership with NVIDIA can be a testomony to this dedication, bringing collectively the very best of our capabilities to revolutionize AI infrastructure. As we proceed to push the boundaries of what’s potential, we invite you to affix us on this thrilling journey, embracing the potential of AI to remodel your group and the world. Are you prepared?

Congratulations to the groups working arduous to make this a actuality.

Collectively, we’re constructing a safer, agile, and revolutionary tomorrow. 

Share:

Hypershield makes use of AI to dynamically refine safety insurance policies primarily based on software id and conduct. It automates coverage creation, optimization, and enforcement throughout workloads. As well as, Hypershield guarantees to let organizations autonomously phase their networks when threats are an issue, achieve exploit safety with out having to patch or revamp firewalls, and robotically improve software program with out interrupting computing assets.

AI Protection discovers the assorted fashions being utilized in a buyer’s AI growth and makes use of 4 options to assist prospects implement AI safety: AI entry, AI cloud visibility, AI mannequin and software validation, and AI runtime safety. AI entry presents visibility into who needs or has use of an AI software after which it controls entry to guard and implement data-loss prevention and mitigate potential threats. AI cloud visibility robotically uncovers AI belongings comprising custom-built AI functions throughout your distributed atmosphere, together with unsanctioned AI workloads. This supplies a single-pane-of-glass view of AI stock, Cisco mentioned.

“AI infrastructure is inherently complicated, and securing it requires a brand new, holistic strategy. That’s why our structure embeds safety at each layer of the AI stack,” Patel wrote in a weblog publish concerning the information. “With options like Cisco Isovalent, Cisco Safe Firewall, Cisco Hypershield and Cisco AI Protection, we robotically broaden and adapt safety measures as infrastructure modifications, offering safety towards evolving threats. This additionally consists of the alternative to combine with NVIDIA BlueField-3 DPUs to enhance safety throughout the stack with real-time AI workload risk detection with NVIDIA DOCA AppShield.”