Enterprise networks are grappling with significant challenges as organizations increasingly adopt AI strategies that demand more from their infrastructure. Traditional desk-to-cloud traffic patterns are insufficient to support the needs of agentic AI, a point underscored by Cisco’s VP of Networking Sales for EMEA, Enrico Mercadante, during a discussion at Cisco Live Amsterdam. Mercadante emphasized that refreshing network infrastructure is now a strategic imperative for companies looking to effectively transform through AI.
Agentic AI, which relies on enterprise data rather than solely on internet data, places three unique demands on network infrastructure. First, organizations need to connect a wider array of devices, such as IoT gadgets, sensors, and operational technology that may currently be isolated. Second, the shift in data movement patterns means enterprises are facing high-volume data flows across their entire operations. For instance, industries like healthcare and logistics are now processing large amounts of video data or real-time sensor information from manufacturing plants. This requires networks to facilitate the seamless transfer of this data to AI models.
Thirdly, security needs escalate in complexity. Organizations must now consider how to verify the identity of sensors connecting to the network and what data an AI agent should have access to. Questions also arise regarding how to authenticate the AI agents themselves. These complexities strain existing security architectures that were not designed with AI integration in mind.
As companies plan for future network refresh cycles, they must also consider the impending threats posed by quantum computing within the next five years. Newer network devices from Cisco incorporate features such as quantum-secure boot, management access, and encryption, presenting a clear divide between those upgrading with obsolete technology and those establishing a quantum-safe foundation. Mercadante cautioned that companies opting for outdated solutions could face another costly upgrade cycle soon, while those investing in future-proofing can significantly extend their infrastructure lifecycle.
Fortunately for enterprises and CFOs tasked with budget management, a complete shutdown for a network overhaul is not a necessity. Mercadante suggests a phased approach, urging organizations to start planning their refresh now, especially as many networks are already in need of updates due to compliance issues stemming from end-of-support policies. A collaborative effort between vendors and customers is essential for crafting a strategic refresh roadmap that identifies which components need immediate attention and which can remain functioning for the time being.
Modern network infrastructure is increasingly defined by the separation of hardware and software, allowing for feature expansions through software updates rather than hardware replacements. Cisco’s Silicon One architecture exemplifies this model, enabling organizations to roll out new functionalities after deployment, thus extending the lifecycle of their infrastructure significantly. This architecture allows for annual “network refreshes” that consist solely of software updates, contingent on robust hardware capable of supporting future requirements.
Cisco’s AI Readiness Index is a tool that measures an organization’s preparedness across various dimensions, including governance, vision, talent, data, and infrastructure. Mercadante stresses that infrastructure assessment should follow a free-thinking exploration of AI strategies, allowing organizations to envision their goals before evaluating whether current infrastructures can support that vision. For network managers, readiness for AI should be the top priority in refresh planning, focusing on how to enhance security, future-proof infrastructure, and prepare for quantum threats while maintaining operational continuity.
However, the infrastructure alone is not enough to tackle the AI data challenge. Companies require clean, reliable, and up-to-date data for their AI models to be effective. Cisco’s acquisition of Splunk plays a crucial role here, enabling the virtualization of data across enterprises. This capability allows organizations to leverage runtime data from various edges rather than consolidating all data into a single location, thereby creating a unified data repository for AI models.
As the conversation with Mercadante concluded, the urgency of completing network refreshes became clear. The timeline depends on the aggressiveness of an organization’s AI strategy and the competitiveness of its market. Nevertheless, Mercadante noted that every customer Cisco engages with has begun conceptualizing their refresh plans. The critical question is not whether to modernize, but how swiftly organizations can implement their strategies to remain competitive in an ever-evolving technological landscape.
See also
Continuous Learning Boosts AI Career Growth Amid Rapid Industry Changes
Tesseract Launches Site Manager and PRISM Vision Badge for Job Site Clarity
Affordable Android Smartwatches That Offer Great Value and Features
Russia”s AIDOL Robot Stumbles During Debut in Moscow
AI Technology Revolutionizes Meat Processing at Cargill Slaughterhouse
