The system is designed to decrease latency in AI purposes by pairing Nvidia’s accelerated computing and software program with Cisco’s high-performance Ethernet networking
In sum – what to know:
Cisco expands Safe AI Manufacturing unit with Nvidia – New infrastructure provides VAST Information integration to hurry up retrieval-augmented era (RAG) pipelines and help agentic AI at enterprise scale.
Decrease latency for AI workloads – AI PODs cut back RAG pipeline response instances from minutes to seconds, serving to brokers entry knowledge quicker and ship extra related, real-time enterprise insights.
Deal with safety and governance – The system consists of role-based entry, audit readiness, and observability by means of Splunk, aiming to safeguard enterprise knowledge whereas enabling AI adoption.
U.S. networking firm Cisco has introduced an enlargement of its Safe AI Manufacturing unit with Nvidia, introducing new infrastructure capabilities designed to help enterprise adoption of agentic AI.
In a launch, Cisco famous that the brand new options give attention to accelerating retrieval-augmented era (RAG) pipelines, enabling quicker and safer knowledge entry for AI brokers at scale.
The corporate stated its AI PODs — infrastructure modules that mix compute, storage, and networking — are actually obtainable with VAST Information’s InsightEngine. This integration is constructed on the Nvidia AI Information Platform reference design, permitting organizations to arrange uncooked knowledge into usable AI datasets, the agency stated.
Cisco famous that these PODs are powered by its UCS servers geared up with Nvidia RTX PRO 6000 Blackwell Server Version GPUs, among the many first to undertake the Nvidia AI Information Platform reference structure.
The agency defined that the system is designed to decrease latency in AI purposes by pairing Nvidia’s accelerated computing and software program with Cisco’s high-performance Ethernet networking. The Safe AI Manufacturing unit framework additionally integrates with Splunk for observability and Cisco AI Protection for safety guardrails, it added.
Jeremy Foster, senior vp and common supervisor of Cisco Compute, stated: “Agentic AI has the potential to unlock the worth of AI for enterprises around the globe. Transferring past chatbots to brokers that may assist clear up true enterprise challenges is revolutionary, however provided that enterprises can successfully leverage the precise knowledge on the proper instances. Cisco, Nvidia and VAST are working collectively to offer prospects a easy path to unlocking the worth of their knowledge. We’re designing the structure for a way the enterprise will construct the following era of AI factories.”
Justin Boitano, vp of Enterprise AI at Nvidia, added: “The following wave of agentic AI will probably be fueled by enterprise knowledge, enabling brokers to faucet into enterprise data throughout inference for exact, up-to-date insights. Bringing collectively Cisco Safe AI Manufacturing unit with Nvidia and VAST InsightEngine creates an built-in platform for operating highly effective AI brokers at scale.”
John Mao, vp of strategic alliances at VAST Information, stated: “By integrating the VAST Information InsightEngine into the Cisco Safe AI Manufacturing unit with Nvidia, we’re giving enterprises the primary built-in design for RAG acceleration at scale. This collaboration with Cisco and Nvidia represents a serious milestone within the evolution of enterprise AI.”
In response to Cisco, the brand new design addresses enterprise considerations round knowledge entry, governance, and efficiency. By decreasing RAG pipeline latency from minutes to seconds, the system goals to supply near-real-time AI responses. It additionally helps a number of brokers and workloads concurrently, whereas incorporating role-based entry controls and audit readiness to assist organizations defend delicate info as they develop AI use circumstances, the agency added.
In June, Cisco launched a variety of information middle and networking improvements aimed toward serving to organizations scale their synthetic intelligence (AI) infrastructure.
The corporate famous that the brand new capabilities had been unveiled following a powerful third quarter during which the networking agency exceeded its annual $1 billion AI infrastructure order goal from hyperscalers.