Optical Interconnect for Large Scale AI Clusters

  EXFO   LESSENGERS   LIGHTCOUNTING  

Date: Wednesday, September 4, 2024

Time: 07:00 AM PT

Duration: 1 Hour

Description:

Data and computation are fundamental to the success of AI, which is why hyperscale data center operators are racing to build ever larger accelerator-based compute clusters. The market jolt caused by LLMs and ChatGPT has further accelerated the race. Scaling clusters requires innovation across many domains: the accelerator chips, their linkages within a node, the fabrics linking the nodes, whether InfiniBand or Ethernet-based, and their topology. The input-output (I/O) growth of each new generation of accelerator and the growth of bandwidth that needs to escape a compute node means using optical I/O is a matter of time. Each hyperscale operator has its preference; some back co-packaged optics (CPO), while others favor pluggable optics. But the use of optics continues to grow in such clusters. For example, Google uses its optical circuit switches as part of its TPU v4 accelerator-based AI compute clusters to reduce cost and improve system reliability and flexibility. And Broadcom has already developed CPO-based Tomahawk Ethernet switches; its latest Jericho3-AI chip developed to scale AI compute clusters can now also be CPO-based. This webinar will discuss the challenges of scaling AI clusters as well as the emerging solutions for improving optical I/O density, bandwidth and power efficiency. 

During the webinar, you’ll learn:

  • AI workloads characteristics
  • Interconnect requirements of AI workloads
  • The various options of Optical Interconnect designs and topologies to support AI workloads
  • Optical Interconnect in large scale AI clusters - How big is the market opportunity?

Who should attend?

Switch System Vendors, Silicon Vendors, Optical Transceiver vendors, Cloud and Telecom Service Providers, IT buyers and Decision Makers, and Enterprise CIO and CTO

Register Today!


Speakers:

Cliff Grossner Chief Innovation Officer, Open Compute Project (OCP)Cliff Grossner, Ph.D.

Chief Innovation Officer, Open Compute Project Foundation (OCP)

Dr. Grossner has more than 25 years of telecommunications industry experience encompassing scientific research, market analysis, corporate and product strategy, product management  and marketing. At the OCP, Cliff will lead its market intelligence function and is responsible for driving new innovation programs including guiding inventors developing their early-stage company ideas, setting strategic direction and building awareness of OCP, establishing new alliances, and launching new activities in OCP’s Future Technologies Initiative. Previously, Dr. Grossner was head of the Cloud and Data Center Research Practice at Omdia, responsible for research quality across the practice and his research was focused on cloud services, data center compute and networking, and data center infrastructure. Prior to Omdia, he held senior positions in the information and communications technology industries, including heading strategic marketing for Alcatel-Lucent's enterprise network business, tenures at Bell Labs, Nortel and several early-stage companies. He earned his Doctor of Philosophy at McGill University, and his Master of Science in Computer Science at Concordia University, both in Montreal, Quebec, Canada, having been awarded national scholarships to support his graduate work. He holds more than 10 patents in computer networking, networking embedded security and telecommunications applications.


Vlad Kozlov, Ph.D. Founder & CEO at LightCountingVlad Kozlov, Ph.D.

Founder & CEO, LightCounting

Vladimir Kozlov is the founder and CEO of LightCounting, which he set up in 2004. By now, LightCounting is the leading market research company covering the global communications industry, including optical and wireless networks. Dr. Kozlov has more than 30 years of experience in optoelectronic technologies and devices, optical communications and market research. Dr. Kozlov held market analyst, product development and research staff positions at RHK Inc., Lucent Technologies and Princeton University. Dr. Kozlov holds several US patents and has numerous publications in the area of optoelectronics. He received M.Sc. at Moscow State University in Russia and Ph.D in Physics at Brown University in the United States.


Jones Huang, Director of Business Unit & Operations at EXFOJones Huang

Director of Business Unit & Operations, EXFO

Jones graduated from Feng Chia University (FCU) in Taiwan with an MBA and Electrical Engineering degree. Jones has held key positions in Taiwan's leading test companies and component manufacturers before founding InOpticals in 2018 (acquired by EXFO Inc. at the end of December 2020). Jones has specialized in communications test for more than 20 years and has deep insight into the development and innovation of high-speed test at all stages, from production to R&D.


Taeyong Kim, Chief Marketing Officer at LessengersTaeyong Kim

Chief Marketing Officer, Lessengers

Taeyong joined as Chief Marketing Officer in 2022 after leading his teams on a successful engagement and ramp-up of high-speed data center optics at Optomind Inc., where he ran Marketing, Quality Management and IP Management. He has more than 20 years of industry and academic experience in optical modules for data communications and his current role is focused on marketing and business development. He holds a PhD in Electrical Engineering from Ajou University and a Bachelor of Laws from KNOU.

Evolving DC Networking for Hybrid AI Deployments

  Dell'Oro Group   Juniper Networks      

Date: Thursday, September 19, 2024

Time: 11:00 AM ET

Duration: 1 Hour

Description:

The emergence of generative AI applications marks a new era for AI. The challenge, however, is that the characteristics of AI traffic are unique and quite distinct from the rest of the traffic carried by today’s data center network. The resulting data center build out is causing the need for a rethink of the data center network and strategies for deploying AI workloads in hybrid clouds. Once built predominantly in hyperscale operated clouds, enterprises are now investing in on-prem, private AI data centers for local AI processing.  This webinar will explore different network fabric options and network design approaches to meet these new on-premises and hybrid cloud AI driven requirements and discuss the size of the potential market opportunity for AI-tuned DC networks.

During the webinar, you’ll learn:

  • AI Networks – how big is the market opportunity?
  • AI applications and their potential deployment locations (Cloud, Edge)
  • AI workloads characteristics
  • AI-tuned networks for multivendor deployments
  • Economics of AI networks

Who should attend?

Switch System Vendors, Silicon Vendors, Optical Transceiver vendors, Colocation Facilities providers, Cloud and Telecom Service Providers, IT buyers and Decision Makers, and Enterprise CIO and CTO

Register Today!


Speakers:

Cliff Grossner Chief Innovation Officer, Open Compute Project (OCP)Cliff Grossner, Ph.D.

Chief Innovation Officer, Open Compute Project Foundation (OCP)

Dr. Grossner has more than 25 years of telecommunications industry experience encompassing scientific research, market analysis, corporate and product strategy, product management  and marketing. At the OCP, Cliff will lead its market intelligence function and is responsible for driving new innovation programs including guiding inventors developing their early-stage company ideas, setting strategic direction and building awareness of OCP, establishing new alliances, and launching new activities in OCP’s Future Technologies Initiative. Previously, Dr. Grossner was head of the Cloud and Data Center Research Practice at Omdia, responsible for research quality across the practice and his research was focused on cloud services, data center compute and networking, and data center infrastructure. Prior to Omdia, he held senior positions in the information and communications technology industries, including heading strategic marketing for Alcatel-Lucent's enterprise network business, tenures at Bell Labs, Nortel and several early-stage companies. He earned his Doctor of Philosophy at McGill University, and his Master of Science in Computer Science at Concordia University, both in Montreal, Quebec, Canada, having been awarded national scholarships to support his graduate work. He holds more than 10 patents in computer networking, networking embedded security and telecommunications applications.


Sameh BoujelbeneSameh Boujelbene

Vice President, Dell'Oro Group

Sameh Boujelbene joined Dell’Oro Group in 2011, where she currently oversees research in the areas of Ethernet Campus Switch, Ethernet Data Center Switch, and AI Networks for AI Workloads. During her tenure at the firm, Ms. Boujelbene expanded her research programs to address data center interconnect, AI/ML workloads, and digital transformation. She has published articles and been cited in various industries and trade publications, and she is a frequent speaker at industry conferences and events.


Praful LalchandaniPraful Lalchandani

VP Data Center Product Management, Juniper Networks

Praful Lalchandani is the VP of Product Management for Juniper's Data Center Platforms & AI solutions. Praful has over 20 years of technical, product management, and leadership experience delivering products and solutions for core routing, cybersecurity, edge computing, and SaaS cloud services. In his current role, Praful is responsible for the vision and strategy of Juniper's data center switching solutions with a an emphasis on defining and delivering advanced AI/ML networking solutions that streamline LLM training and optimize AI inference delivery.