Samsung Develops Industry’s First High Bandwidth Memory with AI Processing Power
17.2.2021 13:00:00 EET | Business Wire | Press release
Samsung Electronics Co., Ltd., the world leader in advanced memory technology, today announced that it has developed the industry's first High Bandwidth Memory (HBM) integrated with artificial intelligence (AI) processing power — the HBM-PIM. The new processing-in-memory (PIM) architecture brings powerful AI computing capabilities inside high-performance memory, to accelerate large-scale processing in data centers, high performance computing (HPC) systems and AI-enabled mobile applications.
This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20210217005384/en/
Samsung's HBM-PIM with AI processing power (Photo: Business Wire)
Kwangil Park, senior vice president of Memory Product Planning at Samsung Electronics stated, "Our groundbreaking HBM-PIM is the industry's first programmable PIM solution tailored for diverse AI-driven workloads such as HPC, training and inference. We plan to build upon this breakthrough by further collaborating with AI solution providers for even more advanced PIM-powered applications."
Rick Stevens, Argonne’s Associate Laboratory Director for Computing, Environment and Life Sciences commented, “I’m delighted to see that Samsung is addressing the memory bandwidth/power challenges for HPC and AI computing. HBM-PIM design has demonstrated impressive performance and power gains on important classes of AI applications, so we look forward to working together to evaluate its performance on additional problems of interest to Argonne National Laboratory.”
Most of today's computing systems are based on the von Neumann architecture, which uses separate processor and memory units to carry out millions of intricate data processing tasks. This sequential processing approach requires data to constantly move back and forth, resulting in a system-slowing bottleneck especially when handling ever-increasing volumes of data.
Instead, the HBM-PIM brings processing power directly to where the data is stored by placing a DRAM-optimized AI engine inside each memory bank — a storage sub-unit — enabling parallel processing and minimizing data movement. When applied to Samsung's existing HBM2 Aquabolt solution, the new architecture is able to deliver over twice the system performance while reducing energy consumption by more than 70%. The HBM-PIM also does not require any hardware or software changes, allowing faster integration into existing systems.
Samsung’s paper on the HBM-PIM has been selected for presentation at the renowned International Solid-State Circuits Virtual Conference (ISSCC) held through Feb. 22. Samsung’s HBM-PIM is now being tested inside AI accelerators by leading AI solution partners, with all validations expected to be completed within the first half of this year.
###
About Samsung Electronics Co., Ltd.
Samsung inspires the world and shapes the future with transformative ideas and technologies. The company is redefining the worlds of TVs, smartphones, wearable devices, tablets, digital appliances, network systems, and memory, system LSI, foundry and LED solutions. For the latest news, please visit the Samsung Newsroom at http://news.samsung.com.
To view this piece of content from cts.businesswire.com, please give your consent at the top of this page.
View source version on businesswire.com: https://www.businesswire.com/news/home/20210217005384/en/
Contact information
Ujeong Jahnke
Samsung Semiconductor Europe GmbH
Tel. +49(0)89-45578-1000
Email: sseg.comm@samsung.com
About Business Wire
For more than 50 years, Business Wire has been the global leader in press release distribution and regulatory disclosure.
Subscribe to releases from Business Wire
Subscribe to all the latest releases from Business Wire by registering your e-mail address below. You can unsubscribe at any time.
Latest releases from Business Wire
IQM and Zurich Instruments Launch Real-Time Quantum Error Correction Demonstrator with NVIDIA NVQLink16.3.2026 23:24:00 EET | Press release
Today, IQM Quantum Computers and Zurich Instruments announce a joint project to build and operate a real-time quantum error correction (QEC) demonstrator, enabled by the NVIDIA NVQLink platform. This project marks a significant milestone toward scalable and fault-tolerant quantum computing designed for enterprise and datacenter deployment. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20260316511715/en/ IQM and Zurich Instruments launch real-time quantum error correction demonstrator with NVIDIA NVQLink As enterprises and public institutions worldwide move from quantum exploration to long-term deployment, the challenge has evolved beyond simply accessing quantum hardware. The focus is now on reliably operating quantum computers, seamlessly integrating them into existing compute infrastructure, and scaling them toward fault tolerance. The announced project directly addresses these needs by focusing on full-system integration f
NetApp Accelerates Momentum in AI Leadership with NVIDIA16.3.2026 22:30:00 EET | Press release
NetApp® (NASDAQ: NTAP), the Intelligent Data Infrastructure company, today announced enhancements to its enterprise-grade data platform, enabling customers to remove roadblocks to AI innovation. In addition to supporting the latest innovations from NVIDIA announced at GTC, NetApp is launching NetApp AI Data Engine (AIDE)—a secure, unified AI data platform stack co-engineered with NVIDIA and integrated with the NVIDIA AI Data Platform reference design. A foundational challenge for AI is enabling enterprises to discover, understand, and govern the data they have across their global data estates. If data is AI’s fuel, finding and using the best data is essential to making truly transformative AI. NetApp AIDE helps enterprises solve this need through an automatically created—and continuously updated—global metadata catalog with powerful search capabilities. Critically, the NetApp AIDE metadata catalog goes beyond standard file system metadata and actively analyzes file content to semantica
Lenovo Brings Production-Scale AI to Global Sports: Enhancing Fan Experience, Driving Revenue Growth, Boosting Performance, and Improving Operational Efficiency with NVIDIA16.3.2026 22:30:00 EET | Press release
At NVIDIA GTC today, Lenovo (HKSE: 992) (ADR: LNVGY) announced an expanded multiyear collaboration with NVIDIA to help the global sports industry deploy production-scale AI across mission-critical environments, transforming live data into revenue growth, operational resilience, and real-time decision advantage. The global sports technology market is projected to grow from $23 billion in 2025 to more than $60 billion by 2030. Global sports events represent some of the most complex and demanding operating environments in any industry, combining unprecedented scale, technical sophistication, and public visibility. These events engage billions of viewers worldwide, generate and process petabytes of data in real time, and require highly coordinated, distributed operations across multiple countries, all within a context where reliability, resilience, and uninterrupted performance are non-negotiable. Scaling AI across this ecosystem requires validated infrastructure, domain-trained intelligen
Lattice Joins NVIDIA Halos Ecosystem to Advance Safety for Physical AI with Holoscan Sensor Bridge16.3.2026 22:30:00 EET | Press release
Lattice Semiconductor (NASDAQ: LSCC), the low power programmable leader, today announced it has joined the NVIDIA Halos AI Systems Inspection Lab ecosystem, the first ANSI National Accreditation Board (ANAB) accredited inspection lab for AI-driven physical systems. Announced at the NVIDIA GTC 2026, Lattice will engage with NVIDIA and other Halos ecosystem members to build Halos-certified Holoscan Sensor Bridge-based designs for physical AI and to help shape best practices as the industry evolves. “Physical AI is rapidly moving from controlled environments into the real world, where safety, reliability, and trust are paramount,” said Raemin Wang, Vice President, Segment Marketing, Lattice Semiconductor. “Through this collaboration, Lattice looks forward to contributing our expertise in low power FPGAs and award-winning solution stacks to enable scalable, trusted physical AI systems across robotics, industrial automation, and autonomous applications.” NVIDIA Halos is a comprehensive full
Kinaxis Advances Large-Scale Supply Chain Optimization with NVIDIA AI16.3.2026 22:30:00 EET | Press release
Kinaxis® Inc. (TSX: KXS), a global leader in supply chain orchestration, today announced a new milestone in advancing large-scale supply chain optimization within the Kinaxis Maestro™ platform. Maestro already delivers high-performance optimization across complex global supply chains, and Kinaxis is now extending that leadership by leveraging GPU acceleration powered by NVIDIA cuOpt™ and NVIDIA AI infrastructure. As supply chains grow in scale and complexity, planning models must reconcile tens of millions of variables across extended time horizons and multiple planning levels. As model size expands, the number of potential decisions can scale into billions, dramatically increasing computational needs. Organizations are no longer constrained by insight alone. They are constrained by how quickly they can iterate. In testing on a large-scale semiconductor planning model with nearly 50 million decision variables, Kinaxis achieved up to a 12X reduction in total end-to-end calculation time.
In our pressroom you can read all our latest releases, find our press contacts, images, documents and other relevant information about us.
Visit our pressroom