Intel’s Latest Memory Enhancements Aim to Elevate Data Center Performance

Intel’s Bold Move in Memory Technology

When you think of Intel, your mind probably jumps straight to processors—the brains behind our computers. But did you know that system memory (yep, that’s DRAM) plays a crucial role in performance too? This is especially true in the world of servers, where the demand for processing power has skyrocketed, often outpacing the available memory bandwidth. In simpler terms, the amount of memory bandwidth per processing core has been on the decline. This mismatch can create serious bottlenecks in heavy-duty tasks like weather modeling, computational fluid dynamics, and even certain AI applications. But guess what? Intel has been hard at work to crack this code.

After years of collaboration with industry partners, Intel engineers have finally opened the floodgates by developing what they claim is the fastest system memory to date. Meet the Intel Xeon 6 data center processors, which are the first to harness this memory technology—dubbed MRDIMMs. And the best part? It’s designed to be as plug-and-play as you can get. Bhanu Jaiswal, Intel’s Xeon product manager in the Data Center and AI (DCAI) group, emphasizes that a significant portion of high-performance computing workloads are “memory-bandwidth bound,” making them prime candidates for MRDIMMs. Sounds intriguing, right? Let’s dive into the details of the DDR5 Multiplexed Rank Dual Inline Memory Module—MRDIMM for short.

Unlocking Parallelism in System Memory

So, what makes MRDIMMs so special? Well, the most common memory modules used in data centers—RDIMMs—have parallel resources built right in, similar to modern processors. However, these resources haven’t been fully utilized. George Vergis, a senior principal engineer in memory pathfinding at Intel, puts it this way: “Most DIMMs have two ranks for performance and capacity. It’s the sweet spot.” Think of ranks like banks: one set of memory chips on a module belongs to one rank, while the rest belong to another. With RDIMMs, data can be stored and accessed across multiple ranks, but not at the same time.

Vergis and his team had a lightbulb moment: “Wait a minute. We’ve got parallel resources that are just sitting there unused. Why not put them to work together?” The solution? A small interface chip—a multiplexer or “mux”—that allows data to flow across both ranks of memory simultaneously. This clever little addition consolidates the MRDIMM’s electrical load, enabling it to operate at much higher speeds compared to traditional RDIMMs. The result? A staggering increase in bandwidth—almost double, in fact. We’re talking about a leap from 6,400 megatransfers per second (MT/s) to 8,800 MT/s. That’s some serious speed!

Intel’s Role and the Future of Memory

Now, you might be wondering, is Intel diving back into the memory business? The answer is no. Although Intel has a rich history in memory technology, having invented EPROM and DRAM, the company has pivoted away from its various memory product lines over time. However, Intel remains committed to fostering interoperability and performance across different computing components. Vergis is even part of the JEDEC board, which sets open standards for the microelectronics industry. He’s been hard at work on the DDR6 standard and has previously earned accolades for his contributions to DDR5.

The journey to MRDIMM began back in 2018, and by 2021, the team had prototypes ready to go. Intel collaborated with the memory ecosystem to create the first components and generously donated the specifications to JEDEC as a new open standard in late 2022. What’s particularly impressive about MRDIMMs is their user-friendliness. They use the same connector and form factor as traditional RDIMMs, meaning no motherboard changes are necessary. Plus, they come equipped with all the reliability and error-correcting features that RDIMMs offer, ensuring data integrity remains intact even when requests are multiplexed across the data buffer.

This means data center operators can easily choose MRDIMMs when ordering new servers or swap out RDIMMs for MRDIMMs later on without any fuss. No code changes required to unlock that newfound performance!

Real-World Performance Gains

What do you need to make the most of MRDIMMs? A compatible CPU, of course! The first processor to support this new memory is the Intel Xeon 6, code-named Granite Rapids, which hit the market this year. Recent independent tests pitted two identical Xeon 6 systems against each other—one with MRDIMMs and the other with RDIMMs. The results were impressive: the MRDIMM-equipped system completed tasks up to 33% faster.

Jaiswal points out that the bandwidth improvements from MRDIMM are particularly beneficial for small language models and traditional deep learning tasks, making them a perfect match for Xeon processors. Leading memory vendors are already on board with MRDIMMs, and more are expected to join the fray. High-performance computing labs, including the National Institute for Quantum Science and Technology and the National Institute for Fusion Science, are already making the switch to Xeon 6 with MRDIMMs, supported by OEMs like NEC.

As Jaiswal puts it, “Intel definitely has a lead, backed by a strong ecosystem of OEMs and memory vendors.” With MRDIMMs setting a new standard for memory performance, the future of high-performance computing looks brighter than ever.

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.

  

Latest Articles about Intel

---

About JEDEC:

JEDEC is the global leader in the development of standards for the microelectronics industry. It was founded in 1958 as the Joint Electron Device Engineering Council to develop standards for the microelectronics industry. The organization's headquarters is located in Arlington, Virginia, United States. JEDEC sets standards for a wide range of technologies, including semiconductors, memory devices, integrated circuits, and more. Its standards are widely adopted and utilized throughout the industry to ensure compatibility and interoperability among various electronic devices and components.

  

Latest Articles about JEDEC

Technology Explained

---

CPU: The Central Processing Unit (CPU) is the brain of a computer, responsible for executing instructions and performing calculations. It is the most important component of a computer system, as it is responsible for controlling all other components. CPUs are used in a wide range of applications, from desktop computers to mobile devices, gaming consoles, and even supercomputers. CPUs are used to process data, execute instructions, and control the flow of information within a computer system. They are also used to control the input and output of data, as well as to store and retrieve data from memory. CPUs are essential for the functioning of any computer system, and their applications in the computer industry are vast.

Latest Articles about CPU

---

DDR5: DDR5 (Double Data Rate 5) is the next generation of memory technology for the computer industry. It is a modern day improvement on earlier DDR technologies, with faster speeds, greater bandwidth and higher capacities. DDR5 enables higher resolution, seamless gaming experiences and faster data transfer rates, making it an ideal choice for high-performance computing and 4K gaming. With its greater RAM compatibility, DDR5 provides faster buffering times and raised clock speeds, giving users an improved overall work system. DDR5 is also optimized for multi-tasking, allowing users to multitask without experiencing a significant drop in performance, increasing the productivity of digital tasks. As an ever-evolving technology, DDR5 is paving the way for the computer industry into a new and powerful era.

Latest Articles about DDR5

---

Xeon: The Intel Xeon processor is a powerful and reliable processor used in many computer systems. It is a multi-core processor that is designed to handle multiple tasks simultaneously. It is used in servers, workstations, and high-end desktop computers. It is also used in many embedded systems, such as routers and switches. The Xeon processor is known for its high performance and scalability, making it a popular choice for many computer applications. It is also used in many cloud computing applications, as it is capable of handling large amounts of data and providing high levels of performance. The Xeon processor is also used in many scientific and engineering applications, as it is capable of handling complex calculations and simulations.

Latest Articles about Xeon