Cloud computing is growing at an exponential pace with an increasing number of workloads being hosted in mega-scale public clouds such as Microsoft Azure. Designing and operating such large infrastructures requires not only a significant capital spend for provisioning datacenters, servers, networking and operating systems, but also R&D investments to capitalize on disruptive technology trends and emerging workloads such as AI/ML. This talk will cover the various infrastructure innovations being implemented in large scale public clouds and opportunities/challenges ahead to deliver the next generation of scale computing.
About the speaker
Kushagra Vaid is the general manager and distinguished engineer for Hardware Infrastructure in the Microsoft Azure division. He is accountable for the architecture and design of compute and storage platforms, which are the foundation for Microsoft’s global cloud-scale services. He and his team have successfully delivered four generations of hyperscale cloud hardware that have been deployed at massive scale across a global fleet of datacenters hosting a diverse set of workloads such as public cloud (Microsoft Azure), machine learning, internet search (Bing), big data analytics, cloud gaming (Xbox Live) and online productivity (Office 365). These hyperscale system designs have been fully open sourced and contributed to the Open Compute Project (OCP). Vaid is an active participant in OCP and has influenced several aspects of OCP policy and system specifications. He has published several papers for international research conferences and holds over 30 patents in the field of computer architecture and datacenter design.
Before joining Microsoft, he was a principal engineer at Intel, where his responsibilities included driving the technology direction for Intel’s Xeon microprocessors and platforms. He started his career as a CPU design engineer and architected enterprise-class CPUs and platforms for over a decade at Intel.