Vendor: Sondrel Ltd Category: CPU

Adaptive Data Processing SoC

SFA 300 is designed to implement scalable processing applications supporting local networking.

Overview

SFA 300 is designed to implement scalable processing applications supporting local networking. Optional accelerators and/or custom logic may be integrated to increase performance and minimise for low power.

Active power management allows for mobile and static applications. Secure dedicated subsystem for communications and data management allows for commercial, industrial, and secure applications

Key features

  • High Performance scalable secure Processing engine with CNN/RNN Inference processing
    • Dual Socket Compute per channel Solution – Nominal CPU Cluster and NPU Cluster (CNN/RNN).
    • Data (x4) and System Ports(x2) - PCIe4 or Ethernet - 10Gbps max.
    • Quad External LPDDR4/4+ or LPDDR5 support with serial QSPI NVM interfaces.
    • Internal RAM scaled and distributed using framework for performance/cost optimisation.
    • Composable NOC System Fabric, Multibus, Multi-width (64b to 512b) data path nominal 800MHz.
    • Performance optimisation of data flow for application, QoS Arbitration, and scheduling support.
    • SMS Subsystem (Standard architecture):
      • A System Management Subsystem – manages the configuration, start-up and operation modes of the SoC.
      • Active power Policy implementation (including DVFS).
    • Security Subsystem – Arm® A53 or M33 based with integrated activity/intrusion monitors; Data ECC(2:1) logic:
      • Support for Intrusion detection, SW signing & Crypto, Authentication.
      • Multi-level resource access control – fragmented ecosystem support.
      • Additional Features: Watchman, Data logging.
    • High Performance Compute with CNN/RNN processing:
      • Quad Channel nominal SoC (option increase/reduction of channels)
      • x4 CPU (Arm® A53/A55) Subsystem per Channel A53 – 9.2/A55-16 DMIPS MHz at 1GHZ nominal (A73/75 option)
      • x2 NPU Subsystem/Channel
        • < 8 TOPS (8x8)
      • SMS (System) CPU(s) (A53)
        • 2.3 DMIPS/MHz
      • Quad LPDDR4/4+ Interface
        • +64GB/s, 32b IF
        • Optional LPDDR5
      • Interfaces –Quad Data and Dual System Port
        • PCIe4 x4 Lane/Ethernet 10G IF – 10Gbps/Port
      • PCIe4 Controller - Endpoint, Root Complex, Dual mode, Switch
      • Active Power Management
      • Optional Compute Selection – Dual Socket/channel
      • Optional – Accelerators and/or Customer

Block Diagram

Benefits

  • SFA 300 is targeted at signal and data processing, with a medium to large performance.
  • Multiple SoCs may be ganged to scale performance.
  • Other more general applications are possible with alternate IO and compute slices. For example, IO based around the general use of PCIe & network capabilities

Applications

  • 8K video, AI, facial recognition for surveillance, smart factories, blockchain servers and medical data analysis.

Specifications

Identity

Part Number
SFA 300
Vendor
Sondrel Ltd
Type
Silicon IP

Files

Note: some files may require an NDA depending on provider policy.

Provider

HQ: United Kingdom

Learn more about CPU IP core

Announcing Arm AGI CPU: The silicon foundation for the agentic AI cloud era

For the first time in our more than 35-year history, Arm is delivering its own silicon products – extending the Arm Neoverse platform beyond IP and Arm Compute Subsystems (CSS) to give customers greater choice in how they deploy Arm compute – from building custom silicon to integrating platform-level solutions or deploying Arm-designed processors.

Encarsia: Evaluating CPU Fuzzers via Automatic Bug Injection

Hardware fuzzing has recently gained momentum with many discovered bugs in open-source RISC-V CPU designs. Comparing the effectiveness of different hardware fuzzers, however, remains a challenge: each fuzzer optimizes for a different metric and is demonstrated on different CPU designs.

Frequently asked questions about CPU IP cores

What is Adaptive Data Processing SoC?

Adaptive Data Processing SoC is a CPU IP core from Sondrel Ltd listed on Semi IP Hub.

How should engineers evaluate this CPU?

Engineers should review the overview, key features, supported foundries and nodes, maturity, deliverables, and provider information before shortlisting this CPU IP.

Can this semiconductor IP be compared with similar products?

Yes. Buyers can compare this product with similar semiconductor IP cores or IP families based on category, provider, process options, and structured technical specifications.

×
Semiconductor IP