Ayar Labs introduces Intel FPGA with 4 Tbps Optical Capability at Supercomputing 2023


November 9, 2023

Thumbnail

Summary: Ayar Labs will showcase its in-package optical I/O solution integrated with Intel's Agilex FPGA technology at Supercomputing 2023, promising 5x the bandwidth, 5x lower power, and 20x lower latency for data-intensive workloads like AI and ML.

  • 5 times the current industry bandwidth at 5 times lower power and 20 times lower latency
  • Potential to revolutionize the high-performance computing (HPC) landscape
  • 79% of HPC system and semiconductor vendors surveyed believe that disaggregation and composability of system resources will be a requirement for future systems


Ayar Labs, a leader in silicon photonics for chip-to-chip connectivity, will be showcasing its in-package optical I/O solution integrated with Intel’s Agilex Field-Programmable Gate Array (FPGA) technology. This integration promises to deliver 5 times the current industry bandwidth at 5 times lower power and 20 times lower latency. The optical FPGA, packaged in a common PCIe card form factor, has the potential to revolutionize the high-performance computing (HPC) landscape, particularly for data-intensive workloads like generative artificial intelligence (AI) and machine learning.

According to Charles Wuischpard, CEO of Ayar Labs, we are on the verge of a new era in high-performance computing where optical I/O is becoming an essential building block to meet the growing demands of emerging technologies. The integration of Ayar Labs’ silicon photonics and Intel’s FPGA technology at the upcoming Supercomputing 2023 conference is a concrete demonstration of the maturity and manufacturability of optical I/O to address these critical demands.

At the conference, Ayar Labs will showcase an optical FPGA consisting of two TeraPHY optical I/O chiplets, each capable of 4 Tbps bi-directional bandwidth. These chiplets are connected to a 10 nm FPGA fabric die, which is the core fabric used in Intel’s Agilex FPGAs. The optical communication is powered by two SuperNova light sources, supporting 64 optical channels of high-speed, error-free communication across eight fibers on each chiplet. This configuration enables 5 times the bandwidth at a fraction of the power.

Venkat Yadavalli, Intel Corporation’s VP and GM of Product Excellence Group, highlights that this capability goes well beyond 400G Ethernet and has the potential to unlock significant advancements in high-performance computing, AI, data centers, sensing, communications, edge computing, and more. An optical interface FPGA communicating at over 4 Terabits per second opens up a world of possibilities.

This integration is a prime example of the benefits of the emerging chiplet ecosystem. Ayar Labs’ optical I/O chiplets, developed on GlobalFoundries’ Fotonix monolithic silicon photonics platform, combined with Intel’s FPGA and leading packaging process, deliver a significant performance boost without altering the underlying compute silicon.

The optical FPGA is just one milestone in Ayar Labs’ long-term roadmap. The company is focused on achieving breakthrough performance, industry-standard UCIe and CXL connections, strategic partnerships, manufacturing options, and other technical and ecosystem advancements.

Optical I/O technology is crucial for addressing the future of HPC and AI application performance. It enhances system designs by enabling communication between compute, memory, and networking ASICs with increased bandwidth, lower latency, and reduced power consumption compared to existing electrical interfaces and optical interconnects. It also supports emerging disaggregated, pooled designs, heterogeneous compute systems, and unified memory architectures that are essential for driving future innovation.

A recent study by Hyperion Research confirms that optical I/O is expected to have the highest impact on addressing HPC system architecture challenges in the next two to six years. The study also reveals that 79% of HPC system and semiconductor vendors surveyed believe that disaggregation and composability of system resources will be a requirement for future systems, further emphasizing the need for optical I/O connectivity.

To experience Ayar Labs’ 4 Tbps optical FPGA with its in-package optical I/O solution, visit booth #228 at Supercomputing 2023. Additionally, Ayar Labs will be participating in panels discussing the chiplet ecosystem in high-performance computing, AI/ML, and data acceleration, as well as scalable and adaptable architectures for AI/HPC advancement.

Ayar Labs introduces Intel FPGA with 4 Tbps Optical Capability at Supercomputing 2023

(Source)

Background Information


About Intel: Intel Corporation, a global technology leader, is for its semiconductor innovations that power computing and communication devices worldwide. As a pioneer in microprocessor technology, Intel has left an indelible mark on the evolution of computing with its processors that drive everything from PCs to data centers and beyond. With a history of advancements, Intel's relentless pursuit of innovation continues to shape the digital landscape, offering solutions that empower businesses and individuals to achieve new levels of productivity and connectivity.

Intel Website: https://www.intel.com/
Intel LinkedIn: https://www.linkedin.com/company/intel-corporation/

Technology Explained


Agilex: Agilex is a technology that is used in the computer industry to help streamline the development process. It is based on the Agile methodology, which is a set of principles that focus on delivering high-quality software quickly and efficiently. Agilex helps to break down complex tasks into smaller, more manageable pieces, allowing developers to work on each piece independently. This helps to reduce the time it takes to develop a product, as well as improve the quality of the final product. Additionally, Agilex helps to reduce the risk of errors and bugs, as well as improve communication between developers and stakeholders. By using Agilex, companies can create better products faster and with fewer resources.


chiplet: Chiplets are a new type of technology that is revolutionizing the computer industry. They are small, modular components that can be used to build powerful computing systems. Chiplets are designed to be used in combination with other components, such as processors, memory, and storage, to create a complete system. This allows for more efficient and cost-effective production of computers, as well as more powerful and versatile systems. Chiplets can be used to create powerful gaming PCs, high-end workstations, and even supercomputers. They are also being used in the development of artificial intelligence and machine learning applications. Chiplets are an exciting new technology that is changing the way we build and use computers.


chiplets: Chiplets are a new type of technology that is revolutionizing the computer industry. They are small, modular components that can be used to build a variety of computer systems. Chiplets are designed to be highly efficient and cost-effective, allowing for the creation of powerful and complex systems without the need for large, expensive components. They are also highly customizable, allowing for the creation of systems tailored to specific needs. Chiplets are being used in a variety of applications, from high-end gaming PCs to embedded systems and even supercomputers. They are also being used to create powerful AI systems, allowing for the development of more advanced and intelligent machines. Chiplets are revolutionizing the computer industry, allowing for the creation of powerful and efficient systems at a fraction of the cost.


FPGA: Field Programmable Gate Arrays (FPGAs) are a type of technology used in the computer industry. They are integrated circuits that can be programmed to perform specific tasks. FPGAs are used in a variety of applications, including digital signal processing, networking, and embedded systems. They are also used in the development of artificial intelligence and machine learning algorithms. FPGAs are advantageous because they can be reprogrammed to perform different tasks, allowing for greater flexibility and faster development times. Additionally, FPGAs are more energy efficient than traditional processors, making them ideal for applications that require low power consumption.


latency: Technology latency is the time it takes for a computer system to respond to a request. It is an important factor in the performance of computer systems, as it affects the speed and efficiency of data processing. In the computer industry, latency is a major factor in the performance of computer networks, storage systems, and other computer systems. Low latency is essential for applications that require fast response times, such as online gaming, streaming media, and real-time data processing. High latency can cause delays in data processing, resulting in slow response times and poor performance. To reduce latency, computer systems use various techniques such as caching, load balancing, and parallel processing. By reducing latency, computer systems can provide faster response times and improved performance.


PCIe: PCIe (Peripheral Component Interconnect Express) is a high-speed serial computer expansion bus standard for connecting components such as graphics cards, sound cards, and network cards to a motherboard. It is the most widely used interface in the computer industry today, and is used in both desktop and laptop computers. PCIe is capable of providing up to 16 times the bandwidth of the older PCI standard, allowing for faster data transfer speeds and improved performance. It is also used in a variety of other applications, such as storage, networking, and communications. PCIe is an essential component of modern computing, and its applications are only expected to grow in the future.



Leave a Reply