Following its breakthrough achievement in May 2022 with the development of the first CXL 1.1-based CXL DRAM, Samsung has now taken a giant leap forward with the introduction of the 128GB CXL DRAM that operates on CXL 2.0. With this cutting-edge technology, Samsung is poised to accelerate the commercialization of next-gen memory solutions. The new CXL DRAM boasts a PCle 5.0 interface with x8 lanes and a whopping bandwidth of up to 35GB per second.
“As a member of the CXL Consortium Board of Directors, Samsung Electronics remains at the forefront of CXL technology,” Jangseok Choi, Samsung Electronics’ Vice President of New Business Planning Team, made a statement. “This breakthrough development underlines our commitment to expanding the CXL ecosystem even further through partnerships with data center, server and chipset companies across the industry.”
According to Jim Pappas, Director of Technology Initiatives at Intel Corporation, Intel is thrilled to collaborate with Samsung in advancing a thriving CXL ecosystem. Intel plans to continue partnering with Samsung to promote the growth and adoption of novel CXL products across the industry.
CXL 2.0 has introduced a groundbreaking feature – memory pooling – which is a memory management method that links multiple CXL memory blocks on a server platform to create a pool. This feature enables hosts to dynamically allocate memory from the pool as required, thereby enhancing efficiency and reducing operating costs. This innovation will empower customers to optimize their server memory and allocate resources towards strengthening it.
Samsung aims to commence mass production of CXL 2.0 DRAM later this year and plans to offer additional products in various capacities to cater to future computing applications’ needs.
CXL is a cutting-edge interface that increases the efficiency of accelerators, DRAM, and storage devices when used in conjunction with CPUs in high-performance server systems. The technology’s bandwidth and capacity can be expanded when combined with the main DRAM, and its progress is expected to revolutionize the next-gen computing market. The surging demand for high-speed data processing, driven by key technologies such as artificial intelligence (AI) and machine learning (ML), is propelling this innovation forward.