DOCSLIB.ORG

Intel® Quickassist Technology Components

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Intel® Quickassist Technology Components Intel® QuickAssist Technology Components Neal Oliver, PhD Principal Engineer Intel Corporation QATS002 Agenda • Intel® QuickAssist Architecture Overview • Selected Intel® QuickAssist Architecture Components • Intel® Embedded Processor for 2008 (Tolapai) • Hardware Architecture • Software Architecture • Use Cases • Intel® QuickAssist FSB-FPGA Accelerator Platform (FAP) • System Architecture • Accelerator Hardware Module (AHM) • Design Flow • Intel® QuickAssist Technology Accelerator Abstraction Layer (AAL) • AAL services and features • Software Architecture Intel’sSingle responseChip Solutions to customer - VPN/Firewall need for acceleration on2 Intel platforms Intel® QuickAssist Technology – Comprehensive Approach to Acceleration • Multiple accelerator and attach options with software and ecosystem support • Performance and scalability based on customer needs and priorities Intel’s response to customer need to deploy accelerators on Intel® architecture platforms. 3 Intel® Embedded Processor for 2008 (Tolapai) Single Die integrates - IA 32 based Core @ 600, 1066 and 1200MHz - DDR2 memory controller (MCH) - PCI Express* Technology - Standard IA PC peripherals (ICH) - 3x Gigabit Ethernet MACs - 3x TDM high-speed serial interfaces for 12 T1/E1 or SLIC/CODEC connections - Intel® QuickAssist Integrated Accelerator Vital Statistics - 148 Million transistors - 1,088-ball FCBGA w/1.092 mm pitch - 37.5 mm x 37.5 mm package SingleSystem-on-Chip Chip Solutions (SoC) - VPN/Firewall enabling performant, effective4 system solutions Intel® Embedded Processor for 2008 (Tolapai) System Architecture Current Intel® Embedded Processor Equiv Solution for 2008 (Tolapai) Intel® Pentium® M Processor MCH ATA Tolapai* Sec ATA ICH Co-Proc Enet Phy Wan/Lan Enet Wan/Lan Additional •IA SoC optimized for Power/Performance Area, Cost, Additional Core Area, Cost •In-line network/security acceleration Utilization, •Integrated I/O devices Architecture - •IA SW compatibility Lookaside Only •Highest Compute Cycles Available/$ SingleSoC Chip reduces Solutions area, - VPN/Firewallcost, power Tolapai = Intel® embedded processor for 2008 (Tolapai) 5 Acceleration HW Architecture ‡ Services Unit Local MDIO (x1) Expansion TDM GigE GigE GigE CAN (x2) ‡ yIA CPU Core w/ 256KB L2 cache Security Interface MAC MAC MAC ‡ Bus SSP (x1) Services Unit (16b @ (12 E1/T1) #2 #1 #0 - Intel® Pentium® M processor derivative (3DES, AES, (A)RC4, 80 MHz) IEEE-1588 MD5, SHA-x, PKE, yIntegrated Memory Controller TRNG) 1 channel 64-bit DDR2 - 256 KB - 4 channel DMA engine ASU SRAM - PCI Express* (1x8, 2x4, or 2x1) Acceleration and I/O Complex ‡ Enabling software required. yIntel® QuickAssist Acceleration IA Complex - Multi-core, Multi-threaded Engines IMCH Transparent EDMA - 256KB Internal SRAM PCI-to-PCI Bridge - Security Hardware Acceleration for Bulk: AES, 3DES, (A)RC4 Hash: MD5, SHA-x Public Key – RSA, DSA, DH Cache FSB Memory Controller Hub 256K L2 Internal True Random Number Generator (TRNG) IA32 Core yIntegrated I/O Interfaces - 3x TDM (12 T1/E1) IICH - 3x GbE MAC (RGMII or RMII) APIC, DMA, Timers, Watch Dog - 1x Local Expansion Bus (16b) Timer, RTC, HPET( x3) PCI - 2x Controller Area Network (CAN) Express* Interface Memory Controller - 1x Sync Serial Port (SSP) (x1) - 2x UART, 37x GPIO, (Gen1, DDR2 2x SMBus/I2C, LPC 1x8, 2x4 or (400/533/667/800, - UART (x2) 2x1 root 64b with ECC) - 2x USB, 2x SATA SATA2 .0 USB2 .0 GPIO (x37) complex) (x2) (x2) SMBus(x2) - WDT, RTC LPC1.1 SingleSoC integrates Chip Solutions processor, - VPN/Firewall chipset, accelerators6 Acceleration Packet Processing ‡ Services Unit Local MDIO (x1) Expansion TDM GigE GigE GigE CAN (x2) ‡ Security Interface MAC MAC MAC ‡ Bus SSP (x1) Flows Services Unit (16b @ (12 E1/T1) #2 #1 #0 (3DES, AES, (A)RC4, 80 MHz) IEEE-1588 Classic IA (blue) MD5, SHA-x, PKE, y TRNG) - GigE Rx DMA packets to DRAM (includes IA snoop) 256 KB - IA interrupt ASU SRAM - IA CPU runs protocol Acceleration and I/O Complex ‡ Enabling software required. - IA CPU controls GigE TX IA Complex IMCH Transparent y Fastpath (red) PCI-to-PCI Bridge EDMA - GigE Rx DMA packets to DRAM - Interrupt routed to accelerator Cache FSB Memory Controller Hub 256K L2 - Accelerator operates on IA32 Core packet - Forwarding/filtering and security functions can be IICH handled w/o IA CPU APIC, DMA, Timers, Watch Dog intervention Timer, RTC, HPET( x3) PCI - Accelerator controls GigE Tx Express* Interface Memory Controller y Exception Packets (green) (x1) (Gen1, DDR2 - Move packet to coherent 1x8, 2x4 or (400/533/667/800, DRAM (includes IA snoop) UART (x2) 2x1 root 64b with ECC) SATA2 .0 USB2 .0 GPIO (x37) complex) - Accelerator signals IA CPU (x2) (x2) SMBus(x2) LPC1.1 SingleProcessing Chip Solutions flows preserve - VPN/Firewall legacy, provide next-gen7 performance Intel® Embedded Processor for 2008 (Tolapai) Architecture Overview High Level SW Model Customer App IA Core OS/Stack Driver/Shim …API IO/Acceleration/Security Access Library Acceleration/Security Services (or Unit) SingleSoftware Chip Solutionsframework - VPN/Firewall goal: enable scalable software8 solutions Intel® Embedded Processor for 2008 (Tolapai) Software Architecture Application y Drivers for Linux* (Red Hat) and FreeBSD y Intel® QuickAssist FreeS/WAN Open SSL OCF Technology Shim Layer Shim Layer Shim Layer Security API - Low level crypto API – PKCS #11 Intel® QuickAssist Technology Security API compliant - High level IA Acceleration Drivers protocol support Integrated with Protocol y Accelerators IPSec SSL/TLS IKE Acceleration middleware frameworks Bulk processing Public Key – Combined Encrypt and - OpenSSL RSA, DH, DSA Operation Hash - OCF Bulk Packet Low Level Rand Mod Exp Authentication - FreeS/WAN, etc. Crypto Classify Acceleration SingleConvenient, Chip Solutions efficient - VPN/Firewall access to security9 accelerators Intel® QuickAssist Acceleration Technology Security Applications – Single Chip VPN/FW User Data App TCP IP Auth. State Bulk Keys MAC&PL L5-7 L4 L3 L2&1 Encrypt & Hash Authenticate Traditional Solutions Tolapai with Acceleration High Performance Tolapai RISC or CISC IA 32 Core 1 GbE PCIX NIC Intel®QuickAssist Acceleration Security IMCH Technology Co-Processor IICH layer 72 Bit layer DDR2 GbE Single Chip Solutions - VPN/Firewall Tolapai = Intel® Embedded Processor for 2008 (Tolapai) 10 Intel® Embedded Processor for 2008 (Tolapai) - Summary y The Tolapai System-on-a-Chip enables single chip, small form factor developments, bringing x86 performance and cost effectiveness to new applications! y The Intel® QuickAssist Integrated Accelerator within Tolapai draws its identity from the QuickAssist Software Services Modules, enabling customers to develop complete communications, or security solutions in a single chip design. y The Tolapai Integrated Accelerators make possible single chip solutions such as SMB IP PBX, and VPN/Firewall. Tolapai = Intel® Embedded Processor for 2008 (Tolapai) SingleSoC bringsChip Solutions X86 performance - VPN/Firewall and Cost-effectiveness11 to new apps FSB-FPGA Accelerator Platform (FAP) Field Programmable Gate Arrays (FPGA) allow the development of domain-specific hardware-based, parallel algorithms that execute significantly faster than equivalent algorithm in software Front Side Bus (FSB) provides a high performance, low latency interconnect between AHM and CPU SingleFSB provides Chip Solutions high-bandwidth - VPN/Firewall access to third party vendor12 (TPV) accelerators A A A A M M M M B B B B A A AM A M M B M B B B A A A A M M M M B B B B A A A A M M M M B B B B )/ )/ ) ) Modules FSB – FPGA – FSB Bensley Bensley Entry Two Entry Two 5000 Series that plug into Workstation Workstation ® ® Stoakley Stoakley ® ® ( ( Two Socket Two Socket 2007 ( 2007 ( Intel Intel Platform for 2008 Platform for 2008 Intel Intel Socket Platform for Socket Socket Platform for Socket (AHMs) nce, lowest latency interconnect 13 Server Platforms – A A A A M M M M DP & MP B B B B A A A A ® M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B Xeon ® Modules FSB – FPGA – FSB ) ) processor sockets ® Intel 7000 Series Single Chip Solutions - VPN/Firewall Caneland Caneland Xeon Expandable Expandable Expandable Expandable ® ® ® Attach directly to the Front Side Bus (FSB) FSB provides the highest performa 2007 ( 2007 ( Intel Intel Server Platform for Server Platform for FPGA Accelerator Hardware Modules Intel - What are FSB-FPGA Accelerators? Serial Links Multi-Gigabit A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M B B B B A A A A M M M M connected in a ring topology B B B B A A A A M M M M B B B B 14 Modules FSB – FPGA – FSB ) ) Modules FSB – FPGA – FSB Expandable Expandable Expandable Expandable for 2007 for 2007 ® ® Caneland Caneland ( ( Modules FSB – FPGA – FSB Server Platform Server Platform Intel Intel Flexible hardware configuration Single Chip Solutions - VPN/Firewall Partition complex algorithms across multiple accelerator modules Higher degrees of parallelization for even higher performance Connect to external I/O sources Multiple FPGA modules can be - Multiple AHMs Shared Memory: SMP y In a typical SMP system, all XEON XEON CPUs can see system memory, but they can’t see Inter-Processor Interrupts each other because they (IPI) don’t occupy address space Memory Controller Hub y CPUs usually communicate through shared memory and signal each other using Inter-Processor Interrupts (IPIs) System Memory SHARED MEMORY SingleHow Chip SMP Solutions architectures - VPN/Firewall work 15 Shared Memory: CPU+AHM y A similar technique is used to communicate with FSB- attached AHMs y The AHM device driver allocates (and pins) system FSB FSB memory for AHM device registers, command/response queues and shared workspaces y The physical address of the device register memory area is sent to the AHM y CPU communicates with AHM through this memory-based interface AHMSingle Core Chip participates Solutions - VPN/Firewallin FSB protocol 16 FAP System Architecture y Accelerator Hardware Modules (AHM) contains FPGAs, SRAM, flash memory and control logic.
Load more

Recommended publications
  • QATS001 Agenda
    Overview of Intel® QuickAssist Technology Dr. Nash Palaniswamy Intel Corporation and Guest Speaker Richard Kaufmann HP QATS001 Agenda • Intel® QuickAssist Technology Strategy • Overview of Solutions – Intel® Embedded Processor for 2008 (Tolapai), FSB-FPGA, AAL, etc. • A viewpoint from HP* on Accelerators • Summary 2 Intel® QuickAssist Technology – Comprehensive Approach to Acceleration • Multiple accelerator and attach options with software and ecosystem support • Performance and scalability based on customer needs and priorities Comprehensive Initiative to optimize the use and deployment of accelerators on Intel® architecture platforms. 3 Design Alternatives as Diverse as Application Requirements Accelerator Accelerator Software Design Attach Models Customized Solutions Power Software Maximum Efficiency Scalability Performance 4 Intel® Embedded Processor for 2008 (Tolapai) Single Die integrates - IA 32 based Core @ 600, 1066 and 1200MHz - DDR2 memory controller (MCH) - PCI Express* Technology - Standard IA PC peripherals (ICH) - 3x Gigabit Ethernet MACs - 3x TDM high-speed serial interfaces for 12 T1/E1 or SLIC/CODEC connections - Intel® QuickAssist Integrated Accelerator Vital Statistics - 148 million transistors - 1,088-ball FCBGA w/1.092 mm pitch - 37.5 mm x 37.5 mm package Intel's first integrated x86 processor, chipset and memory controller since 1994's 80386EX 5 Intel® QuickAssist® FPGA Acceleration Platform Source : Pat Gelsinger Keynote Fall IDF 2006 1. Open Ubiquitous Standards Based Approach PCI Express* Gen1, PCI Express* Gen2, Gen 3 in the future 2. Enable third party FSB-FPGA Modules – targeted for Financial Services Industry, Oil and Gas, Life Sciences, Digital Health, etc. FSB-FPGA Modules Targeted 1H08 3. Intel® QuickAssist Technology Accelerator Geneseo – PCI Express* Abstraction Layer (AAL) that seamlessly allows Source : Intel Internal the SW to access acceleration across various technologies.
    [Show full text]
  • System Design for Telecommunication Gateways
    P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come SYSTEM DESIGN FOR TELECOMMUNICATION GATEWAYS Alexander Bachmutsky Nokia Siemens Networks, USA A John Wiley and Sons, Ltd., Publication P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come SYSTEM DESIGN FOR TELECOMMUNICATION GATEWAYS P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come SYSTEM DESIGN FOR TELECOMMUNICATION GATEWAYS Alexander Bachmutsky Nokia Siemens Networks, USA A John Wiley and Sons, Ltd., Publication P1: OTE/OTE/SPH P2: OTE FM BLBK307-Bachmutsky August 30, 2010 15:13 Printer Name: Yet to Come This edition first published 2011 C 2011 John Wiley & Sons, Ltd Registered office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com. The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
    [Show full text]
  • Intel's Core 2 Family
    Intel’s Core 2 family - TOCK lines References Dezső Sima Vers. 1.0 Januar 2019 Contents (1) • 1. Introduction • 2. The Core 2 line • 3. The Nehalem line • 4. The Sandy Bridge line • 5. The Haswell line • 6. The Skylake line • 7. The Kaby Lake line • 8. The Kaby Lake Refresh line • 9. The Coffee Lake line • 10. The Coffee Lake line Refresh Contents (2) • 11. The Cannon Lake line (outlook) • 12. Sunny Cove • 13. References 13. References 12. References (1) [1]: Singhal R., “Next Generation Intel Microarchitecture (Nehalem) Family: Architecture Insight and Power Management, IDF Taipeh, Oct. 2008, http://intel.wingateweb.com/taiwan08/ published/sessions/TPTS001/FA08%20IDFTaipei_TPTS001_100.pdf [2]: Bryant D., “Intel Hitting on All Cylinders,” UBS Conf., Nov. 2007, http://files.shareholder.com/downloads/INTC/0x0x191011/e2b3bcc5-0a37-4d06- aa5a-0c46e8a1a76d/UBSConfNov2007Bryant.pdf [3]: Fisher S., “Technical Overview of the 45 nm Next Generation Intel Core Microarchitecture (Penryn),” IDF 2007, ITPS001, http://isdlibrary.intel-dispatch.com/isd/89/45nm.pdf [4]:Pabst T., The New Athlon Processor: AMD Is Finally Overtaking Intel, Tom's Hardware, August 9, 1999, http://www.tomshardware.com/reviews/athlon-processor,121-2.html [5]: Carmean D., “Inside the Pentium 4 Processor Micro-architecture,” Aug. 2000, http://people.virginia.edu/~zl4j/CS854/pda_s01_cd.pdf [6]: Shimpi A. L. & Clark J., “AMD Opteron 248 vs. Intel Xeon 2.8: 2-way Web Servers go Head to Head,” AnandTech, Dec. 17 2003, http://www.anandtech.com/showdoc.aspx?i=1935&p=1 [7]: Völkel F., “Duel of the Titans: Opteron vs. Xeon : Hammer Time: AMD On The Attack,” Tom’s Hardware, Apr.
    [Show full text]
  • Michael Eisenstadt David Pollock How the United States Benefits from Its
    A WASHINGTON INSTITUTE STRATEGIC REPORT How the United States Benefits from Its Alliance with Israel Michael Eisenstadt David Pollock STRATEGIC REPORT 7 Michael Eisenstadt David Pollock ASSET TEST How the United States Benefits from Its Alliance with Israel STRATEGIC REPORT 7 SEPTEMBER 2012 All rights reserved. Printed in the United States of America. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. © 2012 by The Washington Institute for Near East Policy The Washington Institute for Near East Policy, 1828 L Street NW, Suite 1050, Washington, DC 20036. COVER: Detail from die photo of Intel ‘s Tolapai System on Chip (SoC ). Computer processors developed by Intel’s Israel R&D center account for 40 percent of the company’s revenues worldwide. Contents About the Authors v Acknowledgments vii Executive Summary ix 1 | Introduction 1 2 | The Enduring Strategic Logic 3 3 | Cooperation on Hard Security Issues 10 4 | Cooperation on Soft Security Challenges 30 5 | Future Challenges 51 6 | Conclusion 57 ILLUSTRATIONS Fig. 1a | Arab-U.S. Ties Trend 5 Fig. 1b | Israel-U.S. Ties Trend 5 Fig. 2a | U.S. Exports to Israel (2011) 31 Fig. 2b | U.S. Imports from Israel (2011) 31 Fig. 2c | U.S. Exports to Saudi Arabia (2011) 31 Fig. 2d | U.S. Imports from Saudi Arabia (2011) 31 TABLES Table 1 | Select Israel-Origin Systems in Recent Use by the U.S. Military 18 Table 2 | Select U.S.
    [Show full text]
  • Intel in 1968, Became President and CEO in 1975, Chairman in 1979, Chairman Emeritus in 1997
    [Beginning of recorded material] [Music] [Video] Pat Gelsinger: There is no person on the planet that I would be more pleased to introduce than our next speaker. And as you come to IDF, technology, technologists, great innovation, profound insights, and who better to speak on such topics than Gordon Moore? Gordon co-founded Intel in 1968, became president and CEO in 1975, chairman in 1979, chairman emeritus in 1997. You know, he was the guy who sort of codified what we today call Moore's Law, , this insight that's become not just an insight, but the very nature of our industry's heartbeat and direction. I remember some of my personal interactions with Gordon very early on. And one time we were struggling on the 46th compaction, so this is back when, , we were struggling to make a 1.2 million transistor chip. Myself was leading the design team for that. [Youssef A. El-Mansy] was leading the process technology for it. And we were struggling with this low-yield problem. And were running all sorts of experiments and struggling with trying to figure this out. And Gordon summons me to his office along with Youssef, , on this problem. I felt like I was going and seeing God, right, you know, walk in, the Gordon Moore, right? And Gordon started shelling us with questions. Have you run this experiment? Have you tried that? And so on. And by this point Gordon was not a practicing technologist for quite a number of years, so we felt like we were sort of pacifying the old man, Intel Developers' Forum, 8.18.07, Afternoon Session Page 2 , by answering his questions.
    [Show full text]
  • SMI GPU Roadmap Update
    SMI GPU Roadmap Update October, 2009 Agenda . Company Overview . GPU Roadmap . GPU Technology and Success Story . GPU Selection Guide . GPU Software Driver and Design Tool . Why Choose SMI? About Silicon Motion . Founded in Nov. 1995 . Market Position: A global leader in microcontrollers for NAND flash storage and low-power multimedia devices . Mission: Silicon Motion is a leading fabless semiconductor company that provides universally compatible, high- performance, low-power semiconductor solutions for the multimedia electronics markets. 3 Pacific Rim Operations Support Global Focus Beijing, China Seoul, Korea Yokohama, Japan Shanghai, China Milpitas, CA Jhubei, Taiwan Shenzhen, China Headquarters: Jhubei, Taiwan Other Operations: Seoul (KOR), Milpitas (US), Shenzhen, Shanghai, Beijing (CN), Shin Yokohama (JP), and Taipei (TW) Employees: 600+ (400+ Engineers) Distributors: N. America, Europe, China, Singapore, Taiwan, Korea, Singapore Our Products >>Mobile Storage >>Multimedia SoC >>Mobile Communications • Flash Card Controllers • Portable Multimedia SoC • Mobile TV Tuner IC • Mobile TV SoC • Solid State Drive Controllers • PC Camera Controllers (tuner + demodulator) • USB Flash Drive Controllers • Embedded Graphics • CDMA RF ICs • Embedded Flash Controllers • Electronic Toll Collection System RF ICs • Card Reader Controllers Embedded GPU Roadmap SM107 SM502 SM502 4 MB 8 MB 0 MB •2D Engine •2D Engine Main RISC •Host/PCI interface •Host/PCI interface •4MB Int. SDRAM •8MB Int. SDRAM Interface •1280x1024 •1280x1024 •CRT& DVO •CRT& DVO •I/O Peripherals (limited) •I/O Peripherals •15x15mm BGA •19x19mm BGA •MP •MP SM718 SM718 16 MB 0 MB SM712 SM722 SM750 SM750 •PCI & 16/32b Host I/F 4 MB 8 MB 16 MB 0 MB •2D Engine •16MB Int. DDR •1920x1440 •2D Engine •2D/3D Engine •2D Engine •CRT& Dual-D •PCI interface •PCI/AGP interface •1 lane PCI-E Main X86 •17x17mm BGA •4MB Int.
    [Show full text]
  • Valpont.Com, a Technology Content Platform Valpont.Com, a Technology Content Platform
    Valpont.com, a Technology Content Platform Valpont.com, a Technology Content Platform Software Development for Embedded Multi-core Systems Valpont.com, a Technology Content Platform This page intentionally left blank Valpont.com, a Technology Content Platform Software Development for Embedded Multi-core Systems A Practical Guide Using Embedded Intel ® Architecture Max Domeika AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Newnes is an imprint of Elsevier Valpont.com, a Technology Content Platform Cover image by iStockphoto Newnes is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA Linacre House, Jordan Hill, Oxford OX2 8DP, UK Copyright © 2008, Elsevier Inc. All rights reserved. Intel® and Pentium® are registered trademarks of Intel Corporation. *Other names and brands may be the property of others. The author is not speaking for Intel Corporation. This book represents the opinions of author. Performance tests and ratings are measured using specifi c computer systems and/or components and refl ect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or confi guration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher.
    [Show full text]
  • The Intel X86 Microarchitectures Map Version 3.2
    The Intel x86 Microarchitectures Map Version 3.2 8086 (1978, 3 µm) 80386 (1985, 1.5 to 1 µm) P6 (1995, 0.50 to 0.35 μm) Skylake (2015, 14 nm) NetBurst (2000 , 180 to 130 nm) Series: Alternative Names: iAPX 386, 386, i386 Alternative Names: i686 Alternative Names: SKL (Desktop and Mobile), SKX (Server) (Note : all U and Y processors are MCPs) P5 (1993, 0.80 to 0.35 μm) Alternative Names: Pentium 4, Pentium IV, P4 • 16-bit data bus: Series: Series: Pentium Pro (used in desktops and servers) Series: Alternative Names: Pentium, Series: 8086 (iAPX 86) • Desktop/Server: i386DX Variant: Klamath (1997, 0.35 μm) • Desktop: Desktop 6th Generation Core i5 (i5-6xxx, S-H) 80586, 586, i586 • Desktop: Pentium 4 1.x (except those with a suffix, Willamette, 180 nm), Pentium 4 2.0 (Willamette, 180 • 8-bit data bus: • Desktop lower-performance: i386SX Alternative Names: Pentium II, PII • Desktop higher-performance: Desktop 6th Generation Core i7 (i7-6xxx, S-H), Desktop 7th Generation Core i7 X (i7-7xxxX, X), Desktop 7th Series: nm) 8088 (iAPX 88) • Mobile: i386SL, 80376, i386EX, Series: Pentium II 233/266/300 steppings C0 and C1 (Klamath, used in desktops) Generation Core i9 X (i7-7xxxXE, i7-7xxxX, X), Desktop 9th Generation Core i7 X (i7-9xxxX, X, 14 nm++?), Desktop 9th Generation Core i9 X • Desktop/Server: P5, P54C • Desktop higher-performance: Pentium 4 1.xA (Northwood, 130 nm), Pentium 4 2.x (except 2.0, 2.40A, i386CXSA, i386SXSA, i386CXSB New instructions: Deschutes (1998, 0.25 to 0.18 μm) (i7-9xxxXE, i7-9xxxX, X, 14 nm++?), Xeon W-3175X (W, 14 nm++?)
    [Show full text]
  • Power TECH FOCUS O
    www.elettronica-plus.it COVER STORY Conrad punta sul servizio e potenzia la spedizione in 24 ore Power TECH FOCUS O E Il risveglio delle reti neurali VERTICAL MAGAZINE VERTICAL GIUGNO/LUGLIO 2016 - N° 455 - e 4,50 Mensile di elettronica professionale, componenti, strumentazione e tecnologie Velocità x Spedizione 24h Tekkie = Elettronica In caso di mancato recapito inviare al CMP/CPO di Roserio-Milano per la restituzione al mittente previo pagaamento resi al CMP/CPO di Roserio-Milano per la restituzione al mittente previo In caso di mancato recapito inviare SUM Copertina di MA Conrad Business Supplies N° 455 realizzata GIUGNO/LUGLIO 2016 RY da Franco Tedeschi Accanto ai big, vi sono varie mettente ambito dell’intel- piccole società che operano ligenza artificiale che trova nel settore delle reti neurali potenziali e innumerevoli e di ciò che viene definito applicazioni in un ampio “machine learning”, un pro- TECH-FOCUS spettro di realtà odierne 8 ADVERTISERS 10 WE SPEAK ABOUT 15 EDITORIAL 18 Conrad punta sul servizio e potenzia COVER STORY la spedizione in 24 ore - Massimo Valtorta 22 Memorie prossime venture - Peter Lieberwirth TECH INSIGHT 24 L’evoluzione dei System-on-a-Chip - Andrea Cattania 26 Un tool di progettazione veramente completo - Alessandro Nobile ANALOG/MIXED SIGNAL 28 Come realizzare un amplificatore audio in classe D - Steve Bowling TECH-FOCUS 32 Il risveglio delle reti neurali - Paolo De Vittor 38 La sicurezza delle reti si trasferisce dal dominio software a quello hardware - Ryan Kenny DIGITAL 42 Utilizzo di Sram veloci
    [Show full text]
  • The Intel X86 Microarchitectures Map Version 4.1
    The Intel x86 Microarchitectures Map Version 4.1 8086 (1978, 3 µm) P5 (1993, 0.80 to 0.35 μm) Willamette (2000, 180 nm) Skylake (2015, 14 nm to 14 nm++) 80386 (1985, 1.5 to 1 µm) P6 (1995, 0.50 to 0.35 μm) Alternative Names: Alternative Names: Pentium, Alternative Names: NetBurst, Pentium 4, Pentium IV, P4, , Pentium 4 180 nm Alternative Names: SKL (Desktop and Mobile), SKX (Server) (The Desktop 9th Generation Core processors have the same security fixes as CSL)(All Alternative Names: iAPX 386, 386, i386 Alternative Names: i686 MCS-86 80586, 586, i586 Series: U and Y processors have on-package PCH) Series: Series: Pentium Pro (used in desktops and servers) Series: Series: • Desktop: Pentium 4 1.x (except those with a suffix), Pentium 4 2.0 Series: • Desktop/Server: i386DX Variant: Klamath (1997, 0.35 μm) • 16-bit data bus: • Desktop/Server: P5, P54C • Desktop lower-performance: Celeron 1.x (where x is a single digit between 5-9, except Celeron 1.6 SL7EZ, • Desktop: Desktop 6th Generation Core i5 (i5-6xxx, S-H, 14 nm) • Desktop lower-performance: i386SX Alternative Names: Pentium II, PII 8086 (iAPX 86) • Desktop/Server higher- Willamette-128), Celeron 2.0 SL68F • Desktop higher-performance: Desktop 6th Generation Core i7 (i7-6xxx, S-H , 14 nm), Desktop 7th Generation Core i7 X (i7-7xxxX, X, 14 nm+), • Mobile: i386SL, 80376, i386EX, Series: Pentium II 233/266/300 steppings C0 and C1 (Klamath, used in desktops) • 8-bit data bus: performance: P54CQS, P54CS • Server: Foster (Xeon), Foster-MP (Xeon) Desktop 7th Generation Core i9 X (i9-7xxxXE,
    [Show full text]
  • Pegasus: Coordinated Scheduling for Virtualized Accelerator-Based Systems
    Pegasus: Coordinated Scheduling for Virtualized Accelerator-based Systems Vishakha Gupta Karsten Schwan Niraj Tolia Georgia Institute of Technology Georgia Institute of Technology Maginatics Vanish Talwar Parthasarathy Ranganathan HP Labs HP Labs Abstract higher level programming APIs [19, 28]. Unfortunately, Heterogeneous multi-cores—platforms comprised of technically, this implies that drivers rather than operating both general purpose and accelerator cores—are becom- systems or hypervisors determine how accelerators are ing increasingly common. While applications wish to shared, which restricts scheduling policies and thus, the freely utilize all cores present on such platforms, operat- optimization criteria applied when using such heteroge- ing systems continue to view accelerators as specialized neous systems. devices. The Pegasus system described in this paper uses A driver-based execution model can not only poten- an alternative approach that offers a uniform resource tially hurt utilization, but also make it difficult for appli- usage model for all cores on heterogeneous chip mul- cations and systems to obtain desired benefits from the tiprocessors. Operating at the hypervisor level, its novel combined use of heterogeneous processing units. Con- scheduling methods fairly and efficiently share acceler- sider, for instance, an advanced image processing ser- ators across multiple virtual machines, thereby making vice akin to HP’s Snapfish [32] or Microsoft’s Photo- accelerators into first class schedulable entities of choice Synth [25] applications, but offering additional compu- for many-core applications. Using NVIDIA GPGPUs tational services like complex image enhancement and coupled with x86-based general purpose host cores, a watermarking, hosted in a data center. For such appli- Xen-based implementation of Pegasus demonstrates im- cations, the low latency responses desired by end users proved performance for applications by better managing require the combined processing power of both general combined platform resources.
    [Show full text]
  • Parallelism for the Masses “Opportunities and Challenges” ©Intel Corporation 2 Parallelismparallelism Isis Here…Here… and and Growing!Growing!
    ParallelismParallelism forfor thethe Masses:Masses: OpportunitiesOpportunities andand ChallengesChallenges Andrew A. Chien Vice President of Research Intel Corporation Carnegie Mellon University Parallel Thinking Seminar OctobeOctoberr 2299, 2008 OutlineOutline • Is Parallelism a crisis? • Opportunities in Parallelism • Expectations and Challenges • Moving Parallel Programming Forward • What’s Going on at Intel • Questions Parallelism for the Masses “Opportunities and Challenges” ©Intel Corporation 2 ParallelismParallelism isis here…here… And And Growing!Growing! Future: 100+ Larrabee: 12-32 Nehalem: 8+ Dunnington (6) Core2 Quad (4) Number of Cores Core 2 Duo (2) 2006 20072008 2009 2010 … 2015 Parallelism for the Masses “Opportunities and Challenges” ©Intel Corporation 3 Q.Q. IsIs parallelismparallelism aa crisis?crisis? A.A. Parallelism Parallelism isis anan opportunity.opportunity. Parallelism for the Masses “Opportunities and Challenges” ©Intel Corporation 4 Parallelism is a key driver for Energy and Performance Dual-Core 1.73x Performance 1.73x Power 1.13x 1.00x 1.02x 0.87x 0.51x Over-clocked Design Dual-core (+20%) Frequency Underclocked (-20%) Parallelism for the Masses “Opportunities and Challenges” ©Intel Corporation 5 Unified IA and Parallelism Vision – a Foundation Across Platforms PERFORMANCE Integration Options Example: Graphics Notebook, Desktop and Server Canmore, Tolapai, Atom Intel® Core™ 2, i7 Parallelism for the Masses “Opportunities and Challenges” POWER Visual Computing, Gaming ©Intel Corporation 6 Opportunity in Low-power
    [Show full text]