Is this your business? Claim it to manage your IP and profile
The TCP/UDP/IP Network Protocol Accelerator Platform (NPAP) is designed to expedite data transmission while ensuring low latency across Ethernet links. With high-bandwidth capabilities, this platform supports a range of Ethernet speeds from 1G to 100G. The solution benefits from custom hardware acceleration, offloading TCP/UDP/IP tasks to FPGAs, thus freeing up CPU resources for other computational tasks. Significant improvements in network throughput and latency reduction are achieved by integrating complete TCP/UDP/IP connectivity into FPGAs, essential for high-performance applications without using a CPU at all. The NPAP platform offers a highly modular TCP/UDP/IP stack, adaptable to various processing environments. Capable of operating at full line rates in both FPGA (70 Gbps) and ASIC (over 100 Gbps) domains, the platform features 128-bit wide bi-directional data paths and streaming interfaces. It supports scalable processing with multiple parallel TCP engines, allowing seamless operations in data centers and SmartNICs while providing deterministic performance thanks to embedded hardware processing. An additional feature of NPAP is its comprehensive integration within a remote evaluation system. Users can test the platform's capabilities remotely through a dedicated lab, aiding in rapid evaluation without the need for extensive on-site hardware setups. This makes it highly beneficial for applications such as networked storage, iSCSI systems, and automotive backbones, where high data throughput and minimal delay are critical requirements.
NVMe Streamer is a specialized solution intended for boosting storage performance by leveraging Non-Volatile Memory Express protocols integrated within FPGA platforms. It connects Solid-State Drives (SSDs) effectively, capitalizing on PCIe's high-speed data capabilities to outperform legacy protocols. The NVMe Streamer efficiently offloads NVMe tasks into programmable logic, thus maximizing data transfer speeds across various FPGA implementations. Built to satisfy the growing need for high-speed storage solutions, NVMe Streamer is designed as a fully integrated host subsystem. It is pre-validated and operates with various Xilinx and AMD Versal FPGAs, utilizing Xilinx Multi-Gigabit Transceivers to ensure robust PCIe connectivity. This enhancement allows streamlined data communication directly between SSDs and FPGA blocks without CPU intervention, providing what is referred to as "full acceleration". It supports seamless SSD connectivity with support for various PCIe generations, ensuring scalability and flexibility across different use-case scenarios. This IP core is crucial for applications requiring rapid data ingestion and storage. High-frequency, data-centric applications such as automotive data logging, aerospace data acquisition, and high-speed analog and digital data recording benefit greatly from NVMe Streamer. Its ability to provide lossless data streaming from and to SSDs highlights its strategic importance in automotive networks and integrated storage solutions.
The Low Latency Ethernet MAC core provides a vital solution for systems needing quick and effective network connectivity over 10G and 25G Ethernet protocols. Developed in collaboration with Fraunhofer HHI, this IP core supports high-speed networking with very low latency by efficiently managing data frames in FPGA-based systems. It supports the AXI4-Stream protocol, ensuring seamless data transmission and reception processes while optimizing resource usage. The core includes crucial mechanisms such as frame-padding for short frames, VLAN support, and a checksum feature for ensuring data integrity at full line rates. These features make it ideal for handling such network protocols within constrained latency budgets. Intended for a wide array of networking applications, this IP core offers considerable flexibility and customization options, including user-defined settings for maximum frame lengths. Applications span data centers, high-speed trading platforms, and real-time communication systems where low latency is a significant advantage. Its delivery as an integrated, licensable IP core, along with supporting design services, provides comprehensive support for system architects requiring robust Ethernet solutions.
Join the world's most advanced semiconductor IP marketplace!
It's free, and you'll get all the tools you need to discover IP, meet vendors and manage your IP 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.
