Master Your Cloud Networking: Panel Discussion with Doug Merritt, Dave McCann, Rajeev Khanna, and Vinit Anshuman
In a recent webinar, Aviatrix CEO Doug Merritt led a panel discussion on how to "Master Your Cloud Networking" with Dave McCann, Strategic Advisor; Rajeev Khanna, former CIO and CTO at Aon , and Vinit Anshuman, leader at Amazon Web Services (AWS) Marketplace for Security and AI. The four discussed the industry’s greatest challenges when it comes to security, complexity, cost optimization, and integrating AI and how tech leaders can overcome them.?
Complex Network Requirements for Your Cloud Journey
In reviewing the complex challenges of the networking industry, Doug Merritt described some of the core use cases addressed by Aviatrix:?
Doug clarified that none of these use cases are mutually exclusive for Aviatrix customers: “The technology and platform that we bring to any of these six or other use cases is the exact same platform,” he said. “Same UI, same core capabilities. So if you master one of these insertion points, it's easy to spread to the others.”
Industry Challenges with Complex Networks
Dave McCann gave a brief overview of the complexity of the cloud networking industry. From the launch of AWS in 2006, Google Cloud Platform in 2008, Microsoft Azure in 2010, and later Alibaba in Asia, cloud networking has progressed rapidly in two decades. Many enterprises have moved apps to the cloud, and many have an average application portfolio of 500-2000, meaning that data is “quadrupling” every two years and is located everywhere. The industry’s paranoia about security means a very complex reality for networking teams.?
Dave reminded Rajeev of a past work conversation: "This complexity is only increasing exponentially,” Dave said, “and I always remember you telling me your people were not exponential.”?
Rajeev commented with some recommendations for dealing with this complexity:?
“You need people, you need the expertise, and it is hard to find the expertise. And having the budgets to be able to go out and hire enough people to be able to do this in the traditional ways — we do need some new modern practices to be able to manage our infrastructure and manage our network in a more simplified fashion,” Rajeev said.?
Rajeev and Doug also acknowledged the “invisible” aspect of networking — organizations expect networks to perform without effort or fanfare. “Without networking, without connectivity, nothing works,” Rajeev said. “You can't conduct business without connectivity.”
AI & Networking: Driving Innovation
Doug asked Vinit about the GenAI wave at AWS. Vinit explained that Amazon’s work in AI and ML has been going on for 35 years. He named innovative services like Rufus, GenAI, Power, Expert, Swapping, and Assistant that AWS has developed to help customers across the globe.?
“What we're seeing today is a democratization of GenAI for broad adoption across all industries and segments and areas,” he said.
Vinit highlighted the importance of data in this landscape: customers need a secured, reachable, and accurate form of data to succeed and build these applications at scale. He explained how AWS is continuing to help customers realize the benefits of GenAI by giving them the capabilities, tools, and resources. For example, Amazon Bedrock is “a fully managed service that offers you the choice of most performing foundational models,” giving organizations “flexibility to play around with the models which are most relevant to their use cases and then . . .? hit it in the market in the right time.”
Increasingly Complex Application, Data, and AI Landscape
Dave spoke about the complexity of AI layering in various industries. “Every industry is grappling with this modernization transformation and ‘now how do we use AI,’” he said.
He gave a real-world example of the complexity of AI layering. The state of California, the world’s 5th largest economy, has 40 million people who have a health care policy set by the federal government and the state of California Health and Human Services. A 1921 state law governs data sharing and privacy in healthcare. 4,000 organizations in California — including hospitals, clinics, charities, government organizations, Medicaid, and Medicare — have signed a state data-sharing agreement concerning personal health data. From a networking perspective, this data sharing is a massive project. AWS, GCP, and Azure all offer APIs, as do databases and applications, but then networking teams have to figure out what API to use, over which networks to exchange data, the ingress, the egress, the security, and the level of encryption.?
In this complex situation and across other industries, AI layering opens new opportunities. “Every network is going to be optimized by AI,” said Dave. But then new questions arise: “Have I got the right capacity? How do I take costs down? What API am I accessing? And where do I apply my GPUs?”
The Journey Toward Cloud Simplicity?
The four agreed on the importance of simplicity in every aspect of networking, from modernizing workloads with Kubernetes to leveraging exciting solutions like Amazon Bedrock.?
“In that very complex landscape that we started with — that's becoming more complex as we're adding in these mission-critical GenAI workloads — we all have got to be thoughtful on how you drive that set of capabilities,” said Doug. “Consistent security, consistent visibility, consistent cost understanding and management, consistent resiliency and reliability across this really, really diverse landscape.”?
Network & Cyber Transformation / Strategic GTM / Helping People & Companies Solve Big Problems / Category Creation / Advisor
3 个月Great opportunity to learn!!