Is this your business? Claim it to manage your IP and profile
TimeServoPTP enhances the TimeServo's capabilities into a comprehensive IEEE 1588v2 PTP-compliant clock solution, engineered for FPGA environments. This advanced PTP ordinary clock slave is designed for seamless synchronization with network time grandmasters via Ethernet frames, allowing for both 1-step and 2-step time synchronization. The self-sufficiency in offering PTP services without host intervention highlights its robust adaptability. With an ability to provide up to 32 "now" time outputs, each equipped with a pulse per second (PPS) signal driven by independent clock domains, TimeServoPTP fosters a flexible control-plane integration through AXI interfaces. Outputs are selectable in binary, IEEE ordinary, or transparent formats, ensuring diverse application support. The Gardiner Type-2 Digital Phase Locked Loop and Clock Domain Crossing (CDC) logic significantly contribute to precision synchronization across network nodes. The capability to handle time-sensitive applications without dedicated processor engagement adds to notable power and resource efficiency, making it an excellent fit for mission-critical network implementations. FPGA resources are efficiently utilized, supporting devices like Intel Agilex and Xilinx UltraScalePlus, allowing effective low-energy processing within high-performance network frameworks.
ARDSoC delivers high-efficiency DPDK for ARM-based SoCs, including MPSoC architectures by leveraging Atomic Rules' extensive datacenter experience. By bypassing the Linux network stack, ARDSoC enhances ARM processor performance by drastically reducing memory pressure and power usage, while minimizing latency. This robust solution supports packet vector and container-aware applications in embedded environments, allowing for seamless protocol bridging for technologies such as CCIX, RDMA, RoCE, and NVMe-oF. By utilizing zero-copy DPDK memory structures, ARDSoC optimizes ARM cache performance, while maintaining full coherence and packet buffering capabilities. Users benefit from using pre-existing DPDK applications without modification, helping to transition smoothly to ARM-based platforms. The plugin services for Xilinx environments, provide notable flexibility and system efficiency, especially for cloud-edge devices. Key benefits include a reduced total cost of ownership (TCO) by utilizing ARM strengths over conventional x86 solutions, offering up to 64 Gbps throughput and perfect packet integrity. ARDSoC's features include advanced Poll Mode Driver (PMD) optimized for ARM A53 and A72 MPSoCs, offering microsecond-level latency interactions to ensure high-speed, lossless packet processing. Its design ensures ARM cores outperform traditional processors, forming a key element in efficient modern network infrastructures by enabling a plug-and-play experience with existing hardware setups.
TimeServo serves as a precise FPGA system timer or clock that supports line-rate independent packet timestamping while addressing high-resolution, modest accuracy timekeeping needs. Engineered with a PI-DPLL, it synchronizes local TCXO with an external 1 PPS signal, enabling high syntonicity and facilitating accurate timestamping with MAC-integrated environments. The component can be extended to TimeServoPTP for a fully compliant IEEE-1588v2/PTP configuration, creating a standalone slave device without the need for host intervention. Providing up to 32 runtime-tunable outputs, TimeServo operates independently across diverse clock domains and supports various output formats including binary, IEEE ordinary, and transparent. Management and observability through an AXI control plane enable dynamic time adjustments and phase-frequency monitoring, ensuring operational adaptability. The attribute of using a standard AXI4-Lite interface strengthens integration prospects with existing FPGA setups. TimeServo's robust phase lock capabilities are backed by a 120-bit resolution phase accumulator and phase-locked loop (PLL) systems capable of maintaining jitter accuracy and flexibility. The design advances are supplemented with application examples and software tools to ease setup and implementation, supporting a wide range of time-sensitive industrial applications.
The Arkville FPGA Gen5 PCIe DMA solution facilitates seamless data transport between FPGA logic and host memory, achieving remarkable bandwidth rates of up to 60 GBytes/s bidirectionally. This high-throughput, low-latency interconnect significantly enhances overall system efficiency by offloading CPU workload, eliminating unnecessary memory copies, and offering zero-copy user space memory buffers. The Arkville solution is equipped with industry-standard APIs for software engineers and RTL interfaces for hardware developers, thus optimizing operations across varied GPP/FPGA applications. Compatibility extends across both Intel and AMD/Xilinx FPGA platforms, ensuring vendor-agnostic deployment to support wide-ranging packet processing demands. Key features include concurrent full-duplex upstream and downstream data movement, AXI Streaming interfaces for packet handling, and support for up to 1 Tbps burst traffic. It houses a dedicated Application BAR (ABAR) AXI4-Master for enhanced application integration. Extensively tested within CI/CD pipelines using Jenkins, Arkville's adaptability and versatility are showcased through example designs like the four-port 10 GbE and single-port 100 GbE configurations, which help speed client development and market readiness. By future-proofing with DPDK and AXI standards, this IP core is highly adaptable for evolving network needs.
The Atomic Rules UDP Offload Engine is a leading-edge FPGA IP core enabling dynamic UDP/IP communication across network nodes at varying capacities, including 10, 25, 40, 50, 100, and 400 GbE. Designed to optimize throughput, it supports various operational requirements while future-proofing network applications. Essential UDP tasks, such as checksum verification, segmentation, and reassembly, are offloaded from software onto hardware, enhancing efficiency and reducing processing overhead. Capable of supporting datagrams of arbitrary size up to IPv4 limits, the UDP Offload Engine features a 16-entry ARP cache and has native support for VLAN configurations, enhancing layer 3 connectivity for non-UDP processes through direct application level interfacing. With robust support for super-jumbo frames up to 16K Bytes, it maintains connectivity integrity and speeds across diverse throughput environments without compromising on packet accuracy. Its low footprint implementation allows multiple core deployments within single FPGA instances. This component excels in various reference designs, such as maximizing data movement in cloud data centers, and enhancing multimedia streaming efficiency, showcasing its adaptability across different interface standards such as Avalon and AXI4. Its compatibility with a broad range of Ethernet MACs further underlines its suitability for integration in existing network systems.
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!
Join the world's most advanced AI-powered semiconductor IP marketplace!
It's free, and you'll get all the tools you need to advertise and discover semiconductor IP, keep up-to-date with the latest semiconductor news and more!
Plus we'll send you our free weekly report on the semiconductor industry and the latest IP launches!
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.
