• Skip to main content
  • Skip to secondary menu
  • Skip to footer

Technologies.org

Technology Trends: Follow the Money

  • Technology Events 2026-2027
  • Sponsored Post
  • Technology Markets
  • About
    • GDPR
  • Contact

SiFive Announces New SiFive Performance P870-D Datacenter Processor

August 16, 2024 By admin Leave a Comment

SiFive, Inc., recognized as the gold standard in RISC-V computing, has unveiled its latest innovation, the SiFive Performance P870-D datacenter processor. This new processor is designed to meet the demanding needs of highly parallelizable infrastructure workloads such as video streaming, storage, and web appliances. By combining the P870-D with products from the SiFive Intelligence product family, datacenter architects can create a high-performance, energy-efficient compute subsystem tailored for AI-powered applications.

The P870-D builds on the success of its predecessor, the P870, and supports the open AMBA CHI protocol, offering customers increased flexibility to scale the number of clusters. This scalability allows for significant performance boosts while minimizing power consumption. By using a standard CHI bus, the P870-D enables customers to scale up to 256 cores, supporting industry-standard protocols like Compute Express Link (CXL) and CHI chip to chip (C2C). This facilitates the creation of coherent, high core count heterogeneous SoCs and chiplet configurations.

The processor’s design emphasizes improved compute density, offering superior performance per watt for workloads that require the execution of multiple compute tasks in parallel. This efficiency is crucial as the industry seeks to lower the cost of training AI models and aligns with the growing emphasis on sustainability.

SiFive’s comprehensive approach, including software, IOMMU, interrupt controller, and other uncore blocks, combined with their intelligence processors, allows customers to achieve optimal performance/watt/dollar metrics for AI and datacenter workloads. As stated by John Ronco, SiFive’s SVP of Product, “The new P870-D enhances our proven Performance architecture to bring new levels of performance, flexibility, and scalability.”

Additionally, SiFive has incorporated Reliability Availability Serviceability (RAS) features into the P870-D, which are designed to detect errors and protect data integrity, preventing downtime and ensuring system reliability. The processor also includes a distributed and scalable IOMMU for accelerating virtualized device IO, critical for addressing the latest functional safety and security requirements.

Key features of the P870-D include scalability to 256 cores, compatibility with the RVA23 profile, support for the RISC-V Sv57 extension, and an Advanced Interrupt Architecture compliant interrupt controller. Customers can leverage both open source and first-party software, benefiting from the collaborative efforts within the RISC-V community, such as the RISE Project.

The P870-D processor is currently sampling to lead customers, with a final production release expected by the end of 2024.

Filed Under: News

Reader Interactions

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Footer

Recent Posts

  • Dify Raises $30 Million to Power the Next Wave of Production AI Applications
  • Nscale’s $2 Billion Bet on the Physical Backbone of the AI Economy
  • Why USB-C Charging on the MacBook Neo Raises Questions About Port Durability
  • MagSafe Wireless Charging: The Magnetic Reinvention of Power
  • Apple Unveils MacBook Neo: A $599 Entry Into the Mac Ecosystem
  • Apple Unveils M5 Pro and M5 Max: A New Era for MacBook Pro, MacBook Air, and Studio Display
  • Apple iPhone 17e: Performance, Practicality, and a Smarter Entry Point into the iPhone 17 Family
  • Apple iPad Air M4 Arrives With 12GB Memory, Wi-Fi 7, and a Serious AI Push
  • Ericsson and Intel Are Redefining What 6G Is Actually For
  • Hollow-Core Fibre, Light Running Through Air Instead of Glass

Media Partners

  • Market Analysis
  • Cybersecurity Market
Beyond Shipyards: Stephen Carmel’s Maritime Warning and the Hard Reality of Rebuilding an Oceanic System
Memory Crunch: Why Prices Are Surging and Why Making More Memory Isn’t Easy
The End of Accounting as We Knew It
The Era of Superhuman Logistics Has Arrived: Building the First Autonomous Freight Network
Why Nvidia Shares Jumped on Meta, and Why the Market Cared
Accrual Launches With $75M to Push AI-Native Automation Into Core Accounting Workflows
Europe’s Digital Sovereignty Moment, or How Regulation Became a Competitive Handicap
Palantir Q4 2025: From Earnings Beat to Model Re-Rating
Baseten Raises $300M to Dominate the Inference Layer of AI, Valued at $5B
Nvidia’s China Problem Is Self-Inflicted, and Washington Should Stop Pretending Otherwise
Armadin Raises $189.9 Million to Build an AI Attacker That Defends the Enterprise
Day Zero Threat Research Summit, August 30 – September 1, 2026, Las Vegas
CrowdStrike Returns to Profit as Revenue Climbs to $1.31 Billion in Q4
Cloudflare 2026 Threat Report Signals the Automation of Cyberwar
Fal.Con Gov 2026, March 18, Washington, D.C.
Huper Corporation Raises $1.5M Pre-Seed to Build a Security-First AI Chief of Staff
CyberBay Summit 2026, March 11–13, Tampa, Florida
Zscaler’s Q2 Beat and the Market’s Reluctance to Celebrate
AI as the New Insider: Why Trust, Not Code, Is Now the Weakest Link
Cybersecurity Meets Corporate Travel: Darktrace Chooses AI-Driven Navan to Power Global Mobility

Media Partners

  • Market Research Media
  • Technology Conferences
The Rise of Faceless Creators: Picsart Launches Persona and Storyline for AI Character-Driven Content
Apple TV Arrives on The Roku Channel, Expanding the Streaming Platform Wars
Why Attraction-Grabbing Stations Win at Tech Events
Why Nvidia Let Go of Arm, and Why It Matters Now
When the Market Wants a Story, Not Numbers: Rethinking AMD’s Q4 Selloff
BBC and the Gaza War: How Disproportionate Attention Reshapes Reality
Parallel Museums: Why the Future of Art Might Be Copies, Not Originals
ClickHouse Series D, The $400M Bet That Data Infrastructure, Not Models, Will Decide the AI Era
AI Productivity Paradox: When Speed Eats Its Own Gain
Voice AI as Infrastructure: How Deepgram Signals a New Media Market Segment
COMPUTEX 2026, June 2–5, Taipei
360° Mobility Mega Shows 2026, April 14–17, Taipei
Forrester CX Summit Series 2026: Amsterdam, New York, San Francisco
IAMPHENOM 2026, March 10–12, Pennsylvania Convention Center, Philadelphia
Billington State and Local CyberSecurity Summit, March 9–11, 2026, Washington, D.C.
Mobile World Congress (MWC) 2026 – 2–5 March, Barcelona, Spain
The AI Summit London, 10–11 June 2026, Tobacco Dock, London
aim10x Digital 2026, March 18, Virtual
Harvard Business Review Strategy Summit, February 26, 2026, Virtual
International Compact Modeling Conference, July 30–31, 2026, Long Beach, California

Copyright © 2022 Technologies.org

Media Partners: Market Analysis & Market Research and Exclusive Domains, Photography