SK hynix Develops PIM, Next-Generation AI Accelerator

SK hynix | February 17, 2022

SK hynix or "the Company" announced on February 16 that it has developed PIM*, a next-generation memory chip with computing capabilities.

It has been generally accepted that memory chips store data and CPU or GPU, like human brain, process data. SK hynix, following its challenge to such notion and efforts to pursue innovation in the next-generation smart memory, has found a breakthrough solution with the development of the latest technology.

SK hynix plans to showcase its PIM development at the world's most prestigious semiconductor conference, 2022 ISSCC*, in San Francisco at the end of this month. The  company expects continued efforts for innovation of this technology to bring the memory-centric computing, in which semiconductor memory plays a central role, a step closer to the reality in devices such as smartphones.

For the first product that adopts the PIM technology, SK hynix has developed a sample of GDDR6-AiM (Accelerator* in memory). The GDDR6-AiM adds computational functions to GDDR6* memory chips, which process data at 16Gbps. A combination of GDDR6-AiM with CPU or GPU instead of a typical DRAM makes certain computation speed 16 times faster. GDDR6-AiM is widely expected to be adopted for machine learning, high-performance computing, and big data computation and storage.

GDDR6-AiM runs on 1.25V, lower than the existing product's operating voltage of 1.35V. In addition, the PIM reduces data movement to the CPU and GPU, reducing power consumption by 80%. This, accordingly, helps SK hynix meet its commitment to ESG management by reducing carbon emissions of the devices that adopt this product.

SK hynix also plans to introduce a technology that combines GDDR6-AiM with AI chips in collaboration with SAPEON Inc., an AI chip company that recently spun off from SK Telecom.

The use of artificial neural network data has increased rapidly recently, requiring computing technology optimized for computational characteristics. We aim to maximize efficiency in data calculation, costs, and energy use by combining technologies from the two companies."

Ryu Soo-jung, CEO of SAPEON Inc.

Ahn Hyun, Head of Solution Development who spearheaded the development of the latest technology and product, said that "SK hynix will build a new memory solution ecosystem using GDDR6-AiM, which has its own computing function." He added that "the company will continue to evolve its business model and the direction for technology development."

About SK hynix Inc.
SK hynix Inc., headquartered in Korea, is the world's top tier semiconductor supplier offering Dynamic Random Access Memory chips ("DRAM"), flash memory chips ("NAND flash") and CMOS Image Sensors ("CIS") for a wide range of distinguished customers globally. The Company's shares are traded on the Korea Exchange, and the Global Depository shares are listed on the Luxemburg Stock Exchange


Most people probably still envision AI in the form of a sentient robot that can talk, move around, and experience feelings – something like WALL-E or C-3PO from the movies. Although that still may be the dream, it turns out that the current iteration of AI is actually quite different. With modern AI, all the “thinking” gets done in the cloud, and the algorithms aren’t tied to the identity of a physical machine like we would have expected from the big screen.


Most people probably still envision AI in the form of a sentient robot that can talk, move around, and experience feelings – something like WALL-E or C-3PO from the movies. Although that still may be the dream, it turns out that the current iteration of AI is actually quite different. With modern AI, all the “thinking” gets done in the cloud, and the algorithms aren’t tied to the identity of a physical machine like we would have expected from the big screen.

Related News


Accenture Uses Artificial Intelligence-Driven Database to Help Reduce Electricity Rates for Colonial Pipeline’s Interstate System

Accenture | July 16, 2022

Accenture is using a proprietary database powered by artificial intelligence (AI) to help Colonial Pipeline, the largest refined products pipeline in the United States, reduce regulated and deregulated electric utility rates for its interstate pipeline system. The energy-management project leverages Accenture’s Utility Tracking System (UTS), a proprietary database of approximately 30 million anonymized utility bills that the company has been aggregating for more than 20 years. Built to identify power tariff options around the world, UTS uses AI-powered insights and automation as part of Accenture’s SynOps platform to continuously improve the efficiency and reliability of electricity rate-savings recommendations. Accenture is using insights generated by UTS to evaluate power bills for operations at approximately 80 Colonial Pipeline pump stations along its 5,500-mile pipeline system, which delivers approximately 100 million gallons of refined petroleum products daily to markets in the Southern and Eastern United States. Armed with information about tariff options, Accenture presents lower-cost options — e.g., reduced cost per kilowatt-hour — to Colonial Pipeline, and then works with utilities to help implement new rates. “This initiative, which is part of our ongoing effort to optimize utility rates, has shown encouraging results early on and should continue to help us in our comprehensive review of all utility accounts on our system. “We look forward to our continued partnership with Accenture as we identify opportunities to transition to renewable energy to power our pump stations.” Tony Leo, manager of energy management and power optimization for Colonial Pipeline Yusuf Tayob, group chief executive of Accenture Operations, added, “Our work for Colonial Pipeline is a great example of how Accenture’s long-term investments in technologies are helping clients deepen business resilience amid evolving market conditions. We look forward to making further progress on this project and to helping Colonial Pipeline evaluate how it can buy green energy, which increasingly is powering infrastructure as consumers and stakeholders demand cleaner air and less environmental impact.” About Accenture Accenture is a global professional services company with leading capabilities in digital, cloud and security. Combining unmatched experience and specialized skills across more than 40 industries, we offer Strategy and Consulting, Technology and Operations services and Accenture Song — all powered by the world’s largest network of Advanced Technology and Intelligent Operations centers. Our 710,000 people deliver on the promise of technology and human ingenuity every day, serving clients in more than 120 countries. We embrace the power of change to create value and shared success for our clients, people, shareholders, partners and communities. About Colonial Pipeline Colonial Pipeline Company, founded in 1962, connects refineries ­– primarily located in the Gulf Coast ­– with customers and markets throughout the Southern and Eastern United States through a pipeline system that spans more than 5,500 miles. The company delivers refined petroleum products such as gasoline, diesel, jet fuel, home heating oil, and fuel for the U.S. military. Colonial is committed to safety and environmental stewardship across its operations.

Read More


Unqork's Drag-and-Drop Software Development Platform for Advanced Apps Patented

Unqork | May 27, 2022

Unqork, the leading enterprise no-code platform based on Codeless Architecture, has received US Patent No. 11,204,690 from the US Patent and Trademark Office, which covers the company's enterprise-grade Codeless Architecture platform. The patent, titled "Systems and methods for software development and deployment platforms with advanced workflow and event processing capabilities and graphical version controls," solidifies Unqork's position as the first-ever no-code development platform designed to support the complexity scale and security that enterprise mission-critical applications demand. The Unqork platform's capacity to develop and deploy complex applications is reflected in US Patent No. 11,204,690, which provides users with drag-and-drop functionality to easily define content, workflow, and background activities for mission-critical applications. Unqork's solution comprises pre-built bundles that may be integrated into applications to define input fields, design layouts, application functionalities, preconfigured templates, and background operations for developers, as specified in the issued patent. This is Unqork's first patent grant, with several more patent applications pending. "Since day one, Unqork has had a unique approach – putting visual, drag-and-drop no-code technology in the hands of experienced users to help them build some of the most complex enterprise apps. This patent grant is further validation that what we're doing is truly one-of-a-kind." Gary Hoberman, CEO and founder, Unqork Since Unqork's inception in 2017, enterprise industry leaders have utilized Unqork's no-code platform to deliver apps to market faster, for less money, and with fewer vulnerabilities than traditional coding methods– all without writing a single line of code. Unqork's solution enables the removal of 188 million lines of code in 2021, saving customers over $678 million in tech debt and allowing them to focus on innovation and building better software to support and enhance their businesses. Customers utilize Unqork to implement Codeless Architecture, an open-standard approach to software development that allows businesses to create and maintain best-of-breed solutions without writing any code. This cutting-edge technique enables businesses to develop new software and maximize the value of existing applications without bothering about coding.

Read More


Algolia launches new AI models in spring 2022 to boost businesses' user engagement

Algolia | June 07, 2022

In its Recommend Spring Release 2022, Algolia, the leading API-First Search & Discovery platform, introduced new AI (artificial intelligence) models and features. Algolia Recommend and Algolia Search, which process over 30 billion search requests every week, allow any modern business with an online presence to enhance their end-discovery user's journey by quickly linking them with the most relevant, actionable recommendations. Algolia Recommend introduces powerful AI models that are powered by behavioral insights and are privacy-aware with this latest release. Customers can quickly put their most relevant and up-to-date material in motion for end-users thanks to Algolia's lightning-fast indexing capabilities. With its no-code/low-code tools, Algolia Recommend can be created and deployed in as little as four days. If developers want to fine-tune their implementation even further, they can dig behind the hood. "As shoppers we have an expectation that there will be recommendations on category pages, PDPs (product detail pages), and checkout pages. It speeds up our discovery of the most relevant products, and for the brands themselves, it increases average order value, conversion rates, and repeat visitors. We're focused on having the best quality recommendations and making it easy for our customers to implement across their web properties. Algolia Recommend Spring Release 2022 helps organizations anticipate their users' wants and needs, and recommends the right content to reach them within milliseconds." Bernadette Nixon, chief executive officer, Algolia According to a recent Forrester Consulting Total Economic Impact research, Algolia's ranking algorithm boosted the relevance of search results by 40 percent to 50 percent, particularly among mobile customers. "IDC expects the intelligent knowledge discovery software market to grow to $11.3 billion in revenue in 2025 with a CAGR of 26.4% over the next four years," said Hayley Sutherland, senior research analyst, Conversational AI and Intelligent Knowledge Discovery at IDC. "In this space, Algolia's Search and Discovery Platform, and particularly its Algolia Recommend Spring Release 2022, has the potential to play a significant role in unlocking previously 'hidden' data and making content available for organizational decision making."

Read More