Is this your business? Claim it to manage your IP and profile
The NoC Bus Interconnect from OPENEDGES Technology is an integral component of the ORBIT Memory Subsystem, engineered to offer high performance and efficient area use in semiconductor designs. Its design is centered around facilitating exceptional connectivity and adaptability within system-on-chip (SoC) architectures, employing an automated end-to-end interconnect generation flow. This interconnect solution excels in delivering high-speed routing with low latency, thanks to its use of HyperPath technology which enables superior throughput and flexibility in physical design. The NoC Bus Interconnect's ability to dynamically control bandwidth and latency, coupled with efficient clocking methods, supports diverse applications ranging from AI/ML to automotive technology. Incorporating advanced ActiveQoS bandwidth management, this interconnect offers significant advantages in managing traffic, reducing congestion, and enhancing overall SoC performance. It features long-distance Asynchronous Bridge (LDA) technology, optimizing connections across distant domains on a chip without the added complexity of a typical register slice scheme.
The DDR Memory Controller from OPENEDGES Technology forms a crucial part of their ORBIT Memory Subsystem, aiming to deliver exceptional memory management performance. This controller, noted for low latency and high utilization, integrates seamlessly with various DDR PHYs, including both OPENEDGES’ own and third-party options. Engineered for next-generation semiconductor applications, it combines high-speed capability with advanced scheduling algorithms to optimize DRAM utilization. Equipped with JEDEC-compliant support for multiple DRAM types, such as LPDDR5, DDR5, and GDDR6, the controller ensures broad compatibility and scalability for various applications. Its out-of-order scheduling and dynamic DRAM tuning enable both significant area savings and power reductions, which are critical for conserving resources in high-demand scenarios. The memory controller's design includes advanced features like inline ECC for data integrity and dual-PHY control, which doubles DRAM channel bandwidth using a single controller instance. With a sophisticated pipeline architecture, this controller is designed to maximize efficiency in high-bandwidth applications, meeting the rigorous demands of AI/ML, HPC, and automotive uses.
The NPU, part of the ENLIGHT series by OPENEDGES Technology, is designed as a deep learning accelerator focusing on inferencing computations with superior efficiency and compute density. Developed for high-performance edge computing, this neural processing unit supports a range of operations pertinent to deep neural networks, including convolution and pooling, providing state-of-the-art capability in both power and performance. The NPU's architecture is based on mixed-precision computation using 4-/8-bit quantization which significantly reduces DRAM traffic, thereby optimizing bandwidth utilization and power consumption. Its design incorporates an advanced vector engine optimized for modern deep neural network architectures, enriching its ability to modernize and scale with evolving AI workloads. Accompanying the hardware capabilities, the NPU offers a comprehensive software toolkit featuring network conversion, quantization, and simulation tools. This suite is built for compatibility with mainstream AI frameworks and ensures seamless integration and efficiency in real-world applications ranging from automotive systems to surveillance.
The DDR PHY from OPENEDGES Technology is an essential component of the ORBIT Memory Subsystem, designed to provide low-latency, high-performance PHY IP solutions compatible with a wide array of DRAM standards, including LPDDR5, LPDDR4, DDR5, GDDR6, and HBM3. Utilizing a state-of-the-art mixed-signal architecture, the PHY addresses challenges in DRAM integration, focusing on high performance in low-power environments. It features built-in power management and advanced PLL design, allowing dynamic configuration and minimal power usage. Leveraging a programmable timing structure, the DDR PHY allows precise control and adjustments without impacting ongoing data operations. This flexibility makes it suitable for applications where exact timing is critical, offering low latency in read/write operations between memory controller and DRAM. Integral to its design is the ability to minimize the infrastructure at the system-level, which translates to fewer layers in both package substrates and PCB designs. Supporting frequencies up to 8533 Mbps, the DDR PHY is compliant with JEDEC standards, offering varied but efficient data management solutions, and enhancing overall system performance. Its adaptability makes it applicable in several cost-sensitive implementations, ensuring product competitiveness across a diverse array of applications.
The ORBIT Memory Subsystem by OPENEDGES Technology integrates a network-on-chip interconnect, memory controller, and PHY IPs into a cohesive solution, achieving remarkable system synergies. It is tailored for high-performance AI and computing applications, where its low latency, reduced power consumption, and extensive bandwidth support are paramount. This subsystem's structure is optimized for diverse next-gen semiconductor needs, incorporating ActiveQoS technology for advanced traffic control, managing data flow effectively to minimize latency and maximize performance. The combination of memory controller, PHY, and interconnect facilitates an adaptable and efficient ecosystem that can seamlessly manage a variety of DRAM protocols. By supporting multiple DRAM standards and optimizing for future technologies, the ORBIT Memory Subsystem extends the cycle life of semiconductor products, providing enhanced application coverage. Its design prioritizes energy efficiency and competitive functionality, reinforcing its role in state-of-the-art SoC development.
Join the world's most advanced semiconductor IP marketplace!
It's free, and you'll get all the tools you need to evaluate IP, download trial versions and datasheets, and manage your evaluation workflow!
To evaluate IP you need to be logged into a buyer profile. Select a profile below, or create a new buyer profile for your company.
