2013 Silvermont Intel presentation
-
Upload
artem-russakovskii -
Category
Documents
-
view
15.140 -
download
1
description
Transcript of 2013 Silvermont Intel presentation
Introducing Next Generation Low Power Microarchitecture:
Silvermont
Dadi PerlmutterExecutive Vice President General Manager, Intel Architecture Group Chief Product Officer
Risk FactorsTodays presentations contain forward-looking statements. All statements made that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially. Please refer to our most recent earnings release, Form 10-Q and 10-K filing available for more information on the risk factors that could cause actual results to differ. If we use any non-GAAP financial measures during the presentations, you will find on our website, intc.com, the required reconciliation to the most directly comparable GAAP financial measure.
Rev. 4/16/13
Legal DisclaimersSoftware and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance . Intel, Intel Atom and the Intel logo are trademarks of Intel Corporation in the United States and other countries. Based on the geometric mean of a variety of power and performance measurements across various benchmarks. Benchmarks included in this geomean are measurements on browsing benchmarks and workloads including SunSpider* and page load tests on Internet Explorer*, FireFox*, & Chrome*; Dhrystone*; EEMBC* workloads including CoreMark*; Android* workloads including CaffineMark*, AnTutu*, Linpack* and Quadrant* as well as measured estimates on SPECint* rate_base2000 & SPECfp* rate_base2000; on Silvermont preproduction systems compared to Atom processor Z2580. Individual results will vary. SPEC* CPU2000* is a retired benchmark. * Other names and brands may be claimed as the property of others.1
Based on a geometric mean of the projected power and performance of SPECint* rate_base2000 on Silvermont compared to expected configurations of main ARM*-based mobile competitors using descriptions of the architectures; assumes similar configurations. Numbers may be subject to change once verified with the actual parts. Individual results will vary. SPEC* CPU2000* is a retired benchmark; results are estimates. * Other names and brands may be claimed as the property of others2
Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance.3
Introducing SilvermontNext Gen Microarchitecture22nm SoC Process
~3X Higher Performance1
~5X Lower Power1
Based on the geometric mean of a variety of power and performance measurements across various benchmarks. Benchmarks included in this geomean are measurements on browsing benchmarks and workloads including SunSpider* and page load tests on Internet Explorer*, FireFox*, & Chrome*; Dhrystone*; EEMBC* workloads including CoreMark*; Android* workloads including CaffineMark*, AnTutu*, Linpack* and Quadrant* as well as measured estimates on SPECint* rate_base2000 & SPECfp* rate_base2000; on Silvermont preproduction systems compared to Atom processor Z2580. Individual results will vary. SPEC* CPU2000* is a retired benchmark. * Other names and brands may be claimed as the property of others.1
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .
SilvermontBest In Class Performance Out of Order Execution engine New multi-core and system fabric architecture Capable of delivering 8 cores
Higher single thread performance New IA instructions and core technologies
Best In Class Power & Power Efficiency Wider dynamic power operating range
Enhanced power management Fast standby entry/exit
Great Scalability Across Multiple Market Segments
Architecture & Microarchitecture DefinitionsExampleInstruction Set Definition
AnalogyEngine Type
Architecture
Intel 64
IA-32
HW Implementation for ISA
Microarchitecture
Engine SchematicIvy Bridge Haswell Saltwell Silvermont
SoC (System on a Chip)Single die with integrated IPs
Crypto
Audio CPU Video Display
Type of Vehicle
Haswell
Storage Graphics
Cloverview
The Constantly Evolving
Intel ArchitectureMedia
Performance
Power Efficiency
Manageability
Security
SSE Extensions
AVX InstructionsAES-NI Smart Caches Turbo Boost Technology Intel Hyper-Threading Intel Matrix Storage ...
Thermal Monitoring Technologies Idle States Intel Smart Idle Intel SpeedStep Technology Intel Demand Based Switching ...
Multimedia Instructions Intel Quick Sync Video Intel HD Graphics Intel HD Audio Intel Clear Video HD Intel Flexible Display Technolgy Intel Insider ...
Intel Virtualization Technology Intel vPro Technology Intel CIRA Technology Intel AMT Intel Remote PC Assist Intel Remote Wake ...
Intel Anti-Theft Technology Intel Trusted Execution Technology Intel Secure Key Intel Run Secure Execute Disable Bit ...
Intel Solutions For All Segments
Intel Atom
Datacenter and HPC
Ultrabook
Tablets
Phones
Intelligent Systems
From TERAFLOPS to
MILLIWATTS
Intel Solutions For All Segments
Intel Atom
Datacenter and HPC
Ultrabook
Tablets
Phones
Intelligent Systems
From TERAFLOPS to
MILLIWATTS
Enabling a Full Spectrum of Mobile Products
Ultrabooks
Convertible Mobility Spectrum
Tablet
Smartphone
Best Performance and Outstanding Battery Life Best Visuals Best Battery Life Thinnest and Lightest Form Factors
Intel Atom
As Well as Datacenter and Intelligent Systems SolutionsServer Network Storage Embedded IVI
Highest Performance Best Performance/Watt Highest Density Designs Lowest PowerIntel Atom
in Next Generation Intel ProductsFrom DATACENTER to
Silvermont Microarchitecture
DEVICES
Avoton
Rangeley
Baytrail
Merrifield
TBA
Port of Choice
Enabling Exceptional Experiences* Other names and brands may be claimed as the property of others.
Committed Leadership Roadmap45nm 32nm 22nm 14nm 14nm
Silvermont; The First in a Family of Yearly Refreshes
Intels IDM AdvantagePROCESS TECHNOLOGY INTEL ARCHITECTURE PRODUCT DESIGN
Common Tools
Common Goals
SOFTWARE PACKAGING
MANUFACTURING
Silvermont Technical OverviewBelli KuttannaIntel Fellow Intel Architecture Group
Key MessagesNext Generation Low Power Microarchitecture Manufactured in custom Intel 22nm SOC process technology Present in a wide range of low power products from Phones to Servers ~3x peak performance1 improvement or the same performance at ~5x lower power1 over the current generation Atom core Leading performance and performance per watt efficiency2 First in a family of cores that will be refreshed every year
Based on the geometric mean of a variety of power and performance measurements across various benchmarks. Benchmarks included in this geomean are measurements on browsing benchmarks and workloads including SunSpider* and page load tests on Internet Explorer*, FireFox*, & Chrome*; Dhrystone*; EEMBC* workloads including CoreMark*; Android* workloads including CaffineMark*, AnTutu*, Linpack* and Quadrant* as well as measured estimates on SPECint* rate_base2000 & SPECfp* rate_base2000; on Silvermont preproduction systems compared to Atom processor Z2580. Individual results will vary. SPEC* CPU2000* is a retired benchmark. * Other names and brands may be claimed as the property of others.1
Based on a geometric mean of the projected power and performance of SPECint* rate_base2000 on Silvermont compared to expected configurations of main ARM*-based mobile competitors using descriptions of the architectures; assumes similar configurations. Numbers may be subject to change once verified with the actual parts. Individual results will vary. SPEC* CPU2000* is a retired benchmark; results are estimates. * Other names and brands may be claimed as the property of others.2
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .
Silvermont HighlightsBetter Performance Better Power Efficiency Full Advantage of Intel 22nm SoC Process Technology
Out of Order Execution engine New multi-core and system fabric architecture New IA instructions extensions (Core Westmere level) New security and virtualization technologies
Wider dynamic operating range Enhanced active and idle power management
3-D Tri-gate transistors tuned for SoC products Architecture and design cooptimized with the process
~3X the Performance Or ~5X Lower Power1Based on the geometric mean of a variety of power and performance measurements across various benchmarks. Benchmarks included in this geomean are measurements on browsing benchmarks and workloads including SunSpider* and page load tests on Internet Explorer*, FireFox*, & Chrome*; Dhrystone*; EEMBC* workloads including CoreMark*; Android* workloads including CaffineMark*, AnTutu*, Linpack* and Quadrant* as well as measured estimates on SPECint* rate_base2000 & SPECfp* rate_base2000; on Silvermont preproduction systems compared to Atom processor Z2580. Individual results will vary. SPEC* CPU2000* is a retired benchmark. * Other names and brands may be claimed as the property of others.1
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .
Agenda
Agenda
Silvermont Micro-ArchitectureBENEFITSHigh Performance Without Sacrificing Power Efficiency
FEATURESOut-of-Order Execution Pipeline Macro operation execution pipeline Improved instruction latencies and throughput Smart pipeline resource managementEfficient Branch Processing Accurate branch predictors Fast recovery pipeline
Power and Performance Improvements
Faster and More Efficient Access to Memory
Low Latency, high bandwidth caches Out of order memory transactions Multiple advanced hardware prefetchers Balanced core and memory subsystems
Building upon Intels Strengths and Expertise In Defining Microarchitectures
Multi-Core Support Module building block-based expansion from 1 to 8 cores Module contains: Two coresModule A Module B
Core CoreL2 Cache
Core CoreL2 Cache
Memory Controller DRAM
IDISystem Agent
IDI
Tightly coupled second-level cache (up to 1MB): very low latency, high bandwidth Dedicated point-to-point interface (IDI) to SOC Fabric Independent read, write channels
Higher bandwidth, Lower Latency, OOO transaction support
Per-core frequency and power management supportBalanced Core and Memory Subsystem Design
New Instructions and TechnologiesPerformance VirtualizationVM 1VMM
Evolving Architecture:
Security
VM 2
New Instructions
Intel Core2 64b ISA + Core Westmere SSE4.1, SSE4.2, POPCNT
Intel Core Westmere AES-NI, Intel Secure Key
New Technologies
Real Time Instruction Tracing TSC Deadline Timer
Intel VT-x2: Extended Page Tables Virtual Processor IDs Unrestricted Guest
Intel OS Guard Support for McAfee DeepSAFE
Fully Compatible with the Breadth of IA Software Installed Base
Agenda
Process/Design/Architecture Co-optimization22nm SoC process
+ + =
3D Transistors that offer unprecedented improvement in performance and power Metal stack co-optimized with key SOC IPs for density and performance Multiple 22nm process versions optimized for specific products
Architecture for Low PowerCompletely redefined for efficiency and scalability
Design for Low PowerCustom arrays and libraries optimized for power and performance Power-Aware design and automation
Wide Dynamic Range of Operation + High Efficiency
Unique Intel Leadership in Process, Design and Architecture
Burst Mode Improvements Prior Atom cores Opportunistically exposed additional P-states based on available thermal headroomPower Sharing Between Cores
Silvermont enhancements Burst frequency managed in hardware based on Thermal, Electrical and Power Delivery constraints Power sharing between CPU cores and SOC IPs (e.g. Graphics) is supported Burst operating points can be dynamically adjusted
Core0
Core1
GFX
Maximize Performance within Platform Capabilities
Burst Mode Improvements Prior Atom cores Opportunistically exposed additional P-states based on available thermal headroomPower Sharing Between Cores
Silvermont enhancements Burst frequency managed in hardware based on Thermal, Electrical and Power Delivery constraints Power sharing between CPU cores and SOC IPs (e.g. Graphics) is supported Burst operating points can be dynamically adjusted
Core0
Core1
GFX
Maximize Performance within Platform Capabilities
Burst Mode Improvements Prior Atom cores Opportunistically exposed additional P-states based on available thermal headroomPower Sharing Between Cores Power Sharing Between Cores and GFX
Silvermont enhancements Burst frequency managed in hardware based on Thermal, Electrical and Power Delivery constraints Power sharing between CPU cores and SOC IPs (e.g. Graphics) is supported Burst operating points can be dynamically adjusted
Core0
Core1
GFX
Core0
Core1
GFX
Maximize Performance within Platform Capabilities
Burst Mode Improvements Prior Atom cores Opportunistically exposed additional P-states based on available thermal headroomPower Sharing Between Cores Power Sharing Between Cores and GFX Dynamic Burst
Silvermont enhancements Burst frequency managed in hardware based on Thermal, Electrical and Power Delivery constraints Power sharing between CPU cores and SOC IPs (e.g. Graphics) is supported Burst operating points can be dynamically adjusted
Core0
Core1
GFX
Core0
Core1
GFX
Core0
Core1
GFX
Maximize Performance within Platform Capabilities
Power C-StatesC0 Core State Core Clock L1 Caches C1/C2 C6
OFF
OFFoff
Core VoltageWakeup Time
active
Power C-StatesC0 Core State Core Clock L1 Caches C1/C2 C6
OFF
OFFoff
Improved C6 Modes with Cache State Retention Fast Standby Mode Entry/Exit Fine Grained Power Gating
Core VoltageWakeup Time L2 Cache
activepartial flush retention off
Module State
PLLWakeup Time Power
OFF
OFF
active
active
3D Transistors Enable Low Leakage and Low Voltage Operation
Silvermonts Wide Dynamic Range of OperationComp - largeSilvermont
Core PowerComp - small Performance3Wide Dynamic Range of Silvermont Makes It More Efficient Than Asymmetric CoresResults have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .3
Silvermonts Wide Dynamic Range of OperationHigher power Lower efficiency
Comp - largeSilvermontHigh Burst frequencies at lower power Higher efficiency longer residencies higher performance
Core Power
Long switching times Lost performance & Lower efficiency
Comp - smallComplex switching algorithms Low frequencies + Lower voltage Higher efficiency
Fast, Seamless Transitions
Performance3
Wide Dynamic Range of Silvermont Makes It More Efficient Than Asymmetric CoresResults have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .3
Agenda
Tablet: Silvermont vs. SaltwellCore Power and Performance14.0 3.5 3.0 2.5 2.0 1.5
Peak-to-Peak
Perf
Iso Power
Perf
Lower Poweriso Perf
Peak-to-Peak
Perf
Iso Power
Perf
Lower Poweriso Perf
1.00.5 0.0 STW SLM 1C1T 1C1T peak to peak STW SLM 1C1T 1C1T iso-power Single Threaded1
STW SLM 1C1T 1C1T iso-perf
STW SLM 2C4T 4C4T peak to peak
STW SLM 2C4T 4C4T iso-power Multi-Threaded
STW SLM 2C4T 4C4T iso-perf
Based on the geometric mean of a variety of power and performance measurements across various benchmarks. Benchmarks included in this geomean are measurements on browsing benchmarks and workloads including SunSpider* and page load tests on Internet Explorer*, FireFox*, & Chrome*; Dhrystone*; EEMBC* workloads including CoreMark*; Android* workloads including CaffineMark*, AnTutu*, Linpack* and Quadrant* as well as measured estimates on SPECint* rate_base2000 & SPECfp* rate_base2000; on Silvermont preproduction systems compared to Atom processor Z2580. Individual results will vary. SPEC* CPU2000* is a retired benchmark. * Other names and brands may be claimed as the property of others. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .
Not All Cores Are Created EqualDC Competition QC Competition DC Silvermont
Core Power
Phone Power Target
Performance3
Dual-Core Silvermont is more efficient than Dual-Core Competition2 Efficient dual-cores outperform2 inefficient quad-cores under power constraintsBased on a geometric mean of the projected power and performance of SPECint* rate_base2000 on Silvermont compared to expected configurations of main ARM*-based mobile competitors using descriptions of the architectures; assumes similar configurations. Numbers may be subject to change once verified with the actual parts. Individual results will vary. SPEC* CPU2000* is a retired benchmark; results are estimates. * Other names and brands may be claimed as the property of others. 3 Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .2
Not All Cores Are Created EqualDual Core Silvermont vs. Quad Core Competition2Silvermont Performance Speedup at 1W Core Power
Dual Core Silvermont vs. Quad Core Competition2X Less Core Power at 1W Comp Performance
2.1x
3.1x
2.8x
1.5x
1.4x
1.6x
1.6x The Performance2
Comp1
Comp2
Comp3
Comp1
2.4x Lower Power
Comp2
Comp3
Based on a geometric mean of the projected power and performance of SPECint* rate_base2000 on Silvermont compared to expected configurations of main ARM*-based mobile competitors using descriptions of the architectures; assumes similar configurations. Numbers may be subject to change once verified with the actual parts. Individual results will vary. SPEC* CPU2000* is a retired benchmark; results are estimates. * Other names and brands may be claimed as the property of others. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .
Iso core-count comparisons
Competitive View TabletsSilvermont vs Competition2Silvermont Performance Speedup at 1.5W Core Power 2.3X 2.3X 5.8X 1.7X 5.2X
Silvermont vs Competition2X Less Core Power at Peak Comp Performance
1.6X
3.6X3.0X
Comp1
Comp2
Comp3
Comp4
Comp1
Comp2
Comp3
Comp4
2.0x The Performance2
4.3x Lower Power
Based on a geometric mean of the projected power and performance of SPECint* rate_base2000 on Silvermont compared to expected configurations of main ARM*-based mobile competitors using descriptions of the architectures; assumes similar configurations. Numbers may be subject to change once verified with the actual parts. Individual results will vary. SPEC* CPU2000* is a retired benchmark; results are estimates. * Other names and brands may be claimed as the property of others. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to: http://www.intel.com/performance .