For an organization to make effective use of an AI cluster, it is important to consider the entire process of designing, building, deploying and managing the resource. At each step, a cluster for AI presents new and different challenges that even experienced IT team members may not have encountered before. In this presentation, Penguin Solutions CTO Philip Pokorny will explore AI clusters from design to daily management and will speak to:
- Key considerations when designing an AI cluster
- Important areas that can compromise AI cluster performance
- Ways that software solutions like Penguin's unique Scyld ClusterWare can address complexities
- How to ensure maximum value from your AI cluster investment
Data Center
Systems
Hardware
Infrastructure
Phil Pokorny
Chief Technology Officer
Penguin Solutions
Phil Pokorny is the Chief Technology Officer (CTO) for SGH / Penguin Solutions. He brings a wealth of engineering experience and customer insight to the design, development, support, and vision for our technology solutions.
Phil joined Penguin in February of 2001 as an engineer, and steadily progressed through the organization, taking on more responsibility and influencing the direction of key technology and design decisions. Prior to joining Penguin, he spent 14 years in various engineering and system administration roles with Cummins, Inc. and Cummins Electronics. At Cummins, Phil participated in the development of internal network standards, deployed and managed a multisite network of multiprotocol routers, and supported a diverse mix of office and engineering workers with a variety of server and desktop operating systems.
He has contributed code to Open Source projects, including the Linux kernel, lm_sensors, and LCDproc.
Phil graduated from Rose-Hulman Institute of Technology with Bachelor of Science degrees in math and electrical engineering, with a second major in computer science.
Penguin Solutions
Website: https://www.penguinsolutions.com/
Penguin Solutions designs, builds, deploys, and manages AI and accelerated computing infrastructures at scale. With 25+ years of HPC experience – and more than 75,000 GPUs deployed and managed to date – Penguin is a trusted strategic partner for AI and HPC solutions and services for leading organizations around the world.
Designing, deploying, and operating “AI factories” is an incredibly complex endeavor and Penguin has successfully been delivering AI factories at scale since 2017. The company’s OriginAI infrastructure, which is backed by Penguin's specialized intelligent cluster management software and expert services, streamlines AI implementation and management, and enables predictable AI cluster performance that supports customers’ business needs and return on investment goals for clusters small or large, ranging in size from hundreds to thousands of GPUs.
The OriginAI solution builds on Penguin’s extensive AI infrastructure expertise to reduce complexity and accelerate return on investment, providing CEOs and CIOs alike the essential and reliable infrastructure they need to deploy and manage demanding AI workloads at scale in the data center and at the edge. To learn more visit their website at: https://www.penguinsolutions.com. Follow Penguin Solutions on LinkedIn, Twitter, YouTube, and Facebook.
Generative AI
Infrastructure
Sanchit Juneja
Director, Big Data & ML/AI
Booking.Com
Sanchit Juneja has 18+ years of Tech Leadership Experience in tech and product roles across The US, South-east Asia, Africa, South-east Asia, and Europe with organizations such as Booking.com, AppsFlyer, GoJek, Rocket Internet, and National Instruments. Currently Director- Product (Big Data & ML/AI) with Booking.com
Systems
Hardware
Infrastructure
Moderator
Steven Woo
Fellow and Distinguished Inventor
Rambus
I was drawn to Rambus to focus on cutting edge computing technologies. Throughout my 15+ year career, I’ve helped invent, create and develop means of driving and extending performance in both hardware and software solutions. At Rambus, we are solving challenges that are completely new to the industry and occur as a response to deployments that are highly sophisticated and advanced.
As an inventor, I find myself approaching a challenge like a room filled with 100,000 pieces of a puzzle where it is my job to figure out how they all go together – without knowing what it is supposed to look like in the end. For me, the job of finishing the puzzle is as enjoyable as the actual process of coming up with a new, innovative solution.
For example, RDRAM®, our first mainstream memory architecture, implemented in hundreds of millions of consumer, computing and networking products from leading electronics companies including Cisco, Dell, Hitachi, HP, Intel, etc. We did a lot of novel things that required inventiveness – we pushed the envelope and created state of the art performance without making actual changes to the infrastructure.
I’m excited about the new opportunities as computing is becoming more and more pervasive in our everyday lives. With a world full of data, my job and my fellow inventors’ job will be to stay curious, maintain an inquisitive approach and create solutions that are technologically superior and that seamlessly intertwine with our daily lives.
After an inspiring work day at Rambus, I enjoy spending time with my family, being outdoors, swimming, and reading.
Education
- Ph.D., Electrical Engineering, Stanford University
- M.S. Electrical Engineering, Stanford University
- Master of Engineering, Harvey Mudd College
- B.S. Engineering, Harvey Mudd College
Manoj Wadekar
AI Systems Technologist
Meta
Taeksang Song
CVP
Samsung Electronics
Taeksang is a Corporate VP at Samsung Electronics where he is leading a team dedicated to pioneering cutting-edge technologies including CXL memory expander, fabric attached memory solution and processing near memory to meet the evolving demands of next-generation data-centric AI architecture. He has almost 20 years' professional experience in memory and sub-system architecture, interconnect protocols, system-on-chip design and collaborating with CSPs to enable hetegeneous computing infrastructure. Prior to joining Samsung Electronics, he worked at Rambus Inc., SK hynix and Micron Technology in lead architect roles for the emerging memory controllers and systems.
Taeksang receives his Ph.D. degree from KAIST, South Korea, in 2006. Dr. Song has authored and co-authored over 20 technical papers and holds over 50 U.S. patents.
Markus Flierl
CVP Intel Cloud Services
Intel, Corp
Markus joined Intel in early 2022 to lead Intel Cloud Services which includes Intel Tiber Developer Cloud (ITDC/ cloud.intel.com), Intel Tiber App-Level Optimization (formerly known as Granulate). Intel Tiber Developer Cloud provides a range of cloud services based on Intel latest pre-production and production hardware and software with focus on AI workloads. ITDC hosts large production workloads for companies such as seekr or Prediction Guard. Before joining Intel Markus built out NVIDIA’s GPU cloud infrastructure services leveraging cutting edge NVIDIA and open source technologies. Today it is the foundation for NVIDIA’s GeForce Now cloud gaming service which has become the leader in cloud gaming with over 25 million registered users globally as well as NVIDIA’s DGX cloud and edge computing workloads like NVIDIA Omniverse™. Prior to that Markus led product strategy and product development of private and public cloud infrastructure and storage software at Oracle Corporation and Sun Microsystems.
Rambus
Website: https://www.rambus.com/
Rambus is a provider of industry-leading chips and silicon IP making data faster and safer. With over 30 years of advanced semiconductor experience, we are a pioneer in high-performance memory subsystems that solve the bottleneck between memory and processing for data-intensive systems. Whether in the cloud, at the edge or in your hand, real-time and immersive applications depend on data throughput and integrity. Rambus products and innovations deliver the increased bandwidth, capacity and security required to meet the world’s data needs and drive ever-greater end-user experiences. For more information, visit rambus.com.
Hardware
Infrastructure
Moderator
RK Anand
Co-Founder and CPO
RECOGNI
RK Anand is the Co-founder and Chief Product Officer (CPO) of Recogni, an artificial intelligence startup based in San Jose specializing in building multimodal GenAI inference systems for data centers.
At Recogni, RK spearheads the company’s product development and Go-To-Market strategies within the data center industry.
With an unwavering commitment to customer needs and value creation, RK and the Recogni team are striving to deliver the highest performing and most cost and energy efficient multi-modal GenAI systems to the market.
RK brings over 35 years of leadership experience in data center compute systems, networking, and silicon development. His distinguished career includes engineering roles at Sun Microsystems and serving as Executive Vice President and General Manager at Juniper Networks. As one of the earliest employees at Juniper, RK played a pivotal role in the company’s growth from a startup to generating billions of dollars in revenue.
Gaia Bellone
Chief Data Scientist
Prudential Financial
Gaia is a dynamic and accomplished leader in the field of Data Science and Artificial Intelligence. In her current role at Prudential Financial, she leads Global Data and AI Governance and serves as Chief Data Officer (CDO) for Emerging Markets.
Her contributions to Prudential Financial have been significant and impactful. As the former Chief Data Scientist at Prudential, she led the Data Science team in creating innovative solutions for Digital, Marketing, Sales, and Distribution, the AI/ML Platform team, and the GenAI Enterprise Program. Her leadership and strategic vision have been instrumental in driving business growth and enhancing operational efficiency.
Prior to her tenure at Prudential, she held prominent positions at Key Bank and JPMorgan Chase. At Key Bank, she served as the Head of Data Science for the Community Bank. Her leadership and expertise in data science were crucial in optimizing the bank's operations and improving customer experience. At JPMorgan Chase, she led the data science teams for Home Lending and Auto Finance. Her strategic insights and data-driven solutions significantly improved the business performance in these sectors, contributing to the overall success of the enterprise.
Throughout her career, she has consistently demonstrated her ability to leverage data and AI to drive business growth and improve operational efficiency. Her contributions to the businesses and the enterprise have been substantial and transformative.
Michael Stewart
Partner
M12
Alex Pham
GM, Chief Architect, Head of Data Infrastructure and AI Platforms
Toyota North America
Recogni
Website: https://www.recogni.com/
Recogni designs and builds multimodal GenAI inference systems for data centers. Recogni’s systems are powered by Pareto, the logarithmic math number system that supports AI inferencing at data center scale. Pareto radically simplifies AI compute by turning multiplications into additions making our chips smaller, faster, and less energy-hungry without compromising accuracy.
With a global footprint in Europe and North America, Recogni is home to industry-leading talent across chip design, AI/ML, systems engineering, networking, software, and business. Our mission: build the most compute-dense and energy-efficient GenAI inference system to help data centers maximize the utilization of compute, space, and energy.
Memory
Systems
Hardware
Infrastructure
Rochan Sankar
Co-Founder & CEO
Enfabrica
Rochan is Founder, President and CEO of Enfabrica. Prior to founding Enfabrica, he was Senior Director and leader of the Data Center Ethernet switch silicon business at Broadcom, where he defined and brought to market multiple generations of Tomahawk/Trident chips and helped build industry-wide ecosystems including 25G Ethernet and disaggregated whitebox networking.
Prior, he held roles in product management, chip architecture, and applications engineering across startup and public semiconductor companies. Rochan holds a B.A.Sc. in Electrical Engineering from the University of Toronto and an MBA from the Wharton School, and has 6 issued patents.
Enfabrica
Website: https://enfabrica.net/
Enfabrica is a cutting-edge silicon and software company building disruptive networking solutions for parallel, heterogeneous and accelerated computing infrastructure. Inventors of the Accelerated Compute Fabric SuperNIC (ACF-S), Enfabrica's groundbreaking chips, software stack design and partner-enabled systems are designed to scale GPU clusters faster and more efficiently than any other solution, while returning control of the AI network infrastructure stack to the customer. Enfabrica is elevating networking for the age of GenAI with the world's most advanced, performant and efficient solutions interconnecting compute, memory and network.
Generative AI
Infrastructure
Hardware
Software
Phil Pokorny
Chief Technology Officer
Penguin Solutions
Phil Pokorny is the Chief Technology Officer (CTO) for SGH / Penguin Solutions. He brings a wealth of engineering experience and customer insight to the design, development, support, and vision for our technology solutions.
Phil joined Penguin in February of 2001 as an engineer, and steadily progressed through the organization, taking on more responsibility and influencing the direction of key technology and design decisions. Prior to joining Penguin, he spent 14 years in various engineering and system administration roles with Cummins, Inc. and Cummins Electronics. At Cummins, Phil participated in the development of internal network standards, deployed and managed a multisite network of multiprotocol routers, and supported a diverse mix of office and engineering workers with a variety of server and desktop operating systems.
He has contributed code to Open Source projects, including the Linux kernel, lm_sensors, and LCDproc.
Phil graduated from Rose-Hulman Institute of Technology with Bachelor of Science degrees in math and electrical engineering, with a second major in computer science.
Penguin Solutions
Website: https://www.penguinsolutions.com/
Penguin Solutions designs, builds, deploys, and manages AI and accelerated computing infrastructures at scale. With 25+ years of HPC experience – and more than 75,000 GPUs deployed and managed to date – Penguin is a trusted strategic partner for AI and HPC solutions and services for leading organizations around the world.
Designing, deploying, and operating “AI factories” is an incredibly complex endeavor and Penguin has successfully been delivering AI factories at scale since 2017. The company’s OriginAI infrastructure, which is backed by Penguin's specialized intelligent cluster management software and expert services, streamlines AI implementation and management, and enables predictable AI cluster performance that supports customers’ business needs and return on investment goals for clusters small or large, ranging in size from hundreds to thousands of GPUs.
The OriginAI solution builds on Penguin’s extensive AI infrastructure expertise to reduce complexity and accelerate return on investment, providing CEOs and CIOs alike the essential and reliable infrastructure they need to deploy and manage demanding AI workloads at scale in the data center and at the edge. To learn more visit their website at: https://www.penguinsolutions.com. Follow Penguin Solutions on LinkedIn, Twitter, YouTube, and Facebook.
Systems
Hardware
Infrastructure
Matthew Burns
Technical Marketing Manager
Samtec
Matthew Burns develops go-to-market strategies for Samtec’s Silicon-to-Silicon solutions. Over the course of 20+ years, he has been a leader in design, applications engineering, technical sales and marketing in the telecommunications, medical and electronic components industries. Mr. Burns holds a B.S. in Electrical Engineering from Penn State University.
Samtec
Website: http://www.samtec.com/AI
Founded in 1976, Samtec is a privately held, $822 MM global manufacturer of a broad line of electronic interconnect solutions, including High-Speed Board-to-Board, High-Speed Cables, Mid-Board and Panel Optics, Precision RF, Flexible Stacking, and Micro/Rugged components and cables. With 40+ location severing approximately 125 countries, Samtec’s global presence enables its unmatched customer service.
Generative AI
Infrastructure
Steven Brightfield
Chief Marketing Officer
BrainChip
Steven Brightfield has over 20 years of success defining and bringing to market new semiconductor products with companies such as Qualcomm, SiMA.ai. LSI Logic, Plessey and Zoran for mobile, AR/VR, wearable, edge ML, cable/sat set-top and digital camera chips. He has 10 years of experience launching programmable semiconductor IP cores for CPU/GPU/DSP/NPUs at LSI Logic,ARC, MIPS, Silicon Arts, Improv, and BOPS and licensing into end products that are ubiquitous today. Steven’s technical foundation in digital signal processing led to using (DSPs in innovative products that digitized the world of speech, audio, multimedia, graphics, camera and video processing, most recently applying AI/ML in these same domains. Steven recently joined the BrainChip leadership team to further drive BrainChip’s brand recognition, go-to-market strategy and customer acquisition as BrainChip enters a growth phase for their flagship Akida products. Steven has a Bachelor of Science in Electrical Engineering from Purdue University.
BrainChip
Website: https://brainchip.com/
BrainChip is a leader in edge AI on-chip processing and learning. The company’s first-to-market convolutional, neuromorphic processor, AkidaTM, mimics the event-based processing method of the human brain in digital technology to classify sensor data at the point of acquisition, processing data with unparalleled energy-efficiency and independent of the CPU or MCU with high precision. On-device learning that is local to the chip without the need to access the cloud dramatically reduces latency while improving privacy and data security. In enabling effective edge computing to be universally deployable across real-world applications, such as connected cars, consumer electronics, and industrial IoT, BrainChip is proving that on-chip AI is the future for customers’ products, the planet and beyond.
Generative AI
Hardware
Infrastructure
Hyunsik Choi
Head of SW Platform
Furiosa AI
FuriosaAI
Website: https://furiosa.ai/
FuriosaAI, founded in 2017, specializes in high-performance data center AI chips targeting the most capable AI models and applications. Gen 1 product WARBOY (Samsung 14nm), targeting advanced computer vision applications, has successfully entered volume production and is now deployed in public clouds and on-prem data centers. Gen 2 product RNGD (TSMC 5nm; pronounced like Renegade) equipping HBM3 is set to launch this year to address the growing demand for more energy-efficient and powerful computing for LLM and Multimodal deployment. More information can be found on the official website.