SlideShare ist ein Scribd-Unternehmen logo
1 von 26
Advance Computer Architecture
Danish Shehzad
MSCS
 Multi-Core Computers

Multithreading


GPUs
Introduction To Multi-Core System

 Integration of multiple processor cores on a single chip.
 Multi-core processor is a special kind of multiprocessors:
    All   processors are on the same chip also called Chip
     Multiprocessor.
    Different cores execute different codes (threads) operating on
     different data.
    A shared memory multiprocessor: all cores share the same
     memory via some cache organization
 Provides a cheap parallel computer solution.
 Increases the computation power to PC platform.
Why Multi-Core?
 Limitations of single core architectures:
     High power consumption due to high clock rates (2-3% power
       increase per 1% performance increase).
      Heat generation (cooling is expensive).
      Limited parallelism (Instruction Level Parallelism only).
      Design time and complexity increased due to complex methods to
       increase ILP.
   Many new applications are multithreaded, suitable for multi-core.
      Ex. Multimedia applications.
   General trend in computer architecture (shift towards more
    parallelism).
   Much faster cache coherency circuits, in a single chip.
   Smaller in physical size than SMP.
Single Core VS Multi Core
A Dual-Core Intel Processor
Intel Polaris
 80 cores with Teraflop (10^12)
  performance on a single chip (1st
  chip to do so).
 Mesh network-on-a-chip.
 Frequency target at 5GHz.
 Workload-aware power
  management:
    Instructions to make any core
     sleep or wake as apps
     demand.
    Chip voltage & frequency
     control.
    Peak of 1.01 Teraflops at 62
     watts.
    Peak power efficiency of 19.4
    Short design time.
Innovations on Intel’s Polaris
 Rapid design – The tiled-design approach allows designers to use
  smaller cores that can easily be repeated across the chip.

 Network-on-a-chip – The cores are connected in a 2D mesh
  network that implement message-passing. This scheme is much
  more scalable.

 Fine-grain power management - The individual compute engines
  and data routers in each core can be activated or put to sleep based
  on the performance required by the applications.
What applications benefit From Multicore
Processors


 • Database servers
 • Web servers (Web commerce)
 • Compilers
 • Multimedia applications
 • Scientific applications,
 • In general, applications with
  Thread-level parallelism(as opposed to instruction
  level parallelism)
Multi-Core Computers



Multithreading

GPUs
2. Multi-Threading
Thread-Level Parallelism (TLP)
 This is parallelism on a more coarser scale than instruction-level
  parallelism (ILP).
 Instruction stream divided into smaller streams (threads) to be
  executed in parallel.
 Thread has its own instructions and data.
    May be part of a parallel program or independent programs.
    Each thread has all state (instructions, data, PC, etc.) needed to
     execute.
 Single-core superscalar processors cannot fully exploit TLP.
 Multi-core architectures can exploit TLP efficiently.
    Use multiple instruction streams to improve the throughput of
     computers that run several programs .
 TLP are more cost-effective to exploit than ILP.
Threads vs. Processes
 Process: An instance of a program running on a computer.
    Resource ownership.
        Virtual address space to hold process image including
         program, data, stack, and attributes.
    Execution of a process follows a path though the program.
    Process switch - An expensive operation due to the need to
     save the control data and register contents.
 Thread: A dispatchable unit of work within a process.
    Interruptible: processor can turn to another thread.
    All threads within a process share code and data segments.
    Thread switch is usually much less costly than process switch.
Multithreading Approaches
 Interleaved (fine-grained)
    Processor deals with several thread contexts at a time.
    Switching thread at each clock cycle (hardware need).
    If a thread is blocked, it is skipped.
    Hide latency of both short and long pipeline stalls.
 Blocked (coarse-grained)
    Thread executed until an event causes delay (e.g., cache miss).
    Relieves the need to have very fast thread switching.
    No slow down for ready-to-go threads.
 Simultaneous (SMT)
    Instructions simultaneously issued from multiple threads to
     execution units of superscalar processor.
 Chip multiprocessing
    Each processor handles separate threads.
Multithreading Paradigms
Programming for Multi-Core (In Context Of
Multithreading)
 There must be many threads or processes:
    Multiple applications running on the same machine.
        Multi-tasking is becoming very common.
        OS software tends to run many threads as a part of its normal
         operation.

 An application may also have multiple threads.
        In most cases, it must be specifically written.

 OS scheduler should map the threads to different cores, in
  order to balance the work load or to avoid hot spots due to
  heat generation.
Multi-Core Computers


Multithreading



GPUs
Introduction To GPU
What is GPU?
• It is a processor optimized for 2D/3D
  graphics, video, visual computing, and display.
• It is highly parallel, highly multithreaded multiprocessor
  optimized for visual computing.
• It provide real-time visual interaction with computed
  objects via graphics images, and video.
• It serves as both a programmable graphics processor
  and a scalable parallel computing platform.
• Heterogeneous Systems: combine a GPU with a CPU
Graphics Processing Unit, Why?
Graphics applications are:
    Rendering of 2D or 3D images with complex optical effects;
    Highly computation intensive;
    Massively parallel;
    Data stream based.
 General purpose processors (CPU) are:
    Designed to handle huge data volumes;
    Serial, one operation at a time;
    Control flow based;
    High flexible, but not very well adapted to graphics.
 GPUs are the solutions!
GPU Design
 Process pixels in parallel
    2.3M pixels per frame
    lots of work
 All pixels are independent
    no synchronization
 Lots of spatial locality
    regular memory access
 Great speedups:
    Limited only by the amount of hardware
GPU Design
2) Focus on throughput, not latency
Each pixel can take a long time……as long as we process
  many at the same time.
 Great scalability
 Lots of simple parallel processors
 Low clock speed
CPU vs. GPU Architecture
 GPUs are throughput-optimized
    Each thread may take a long time, but thousands of threads
 CPUs are latency-optimized
    Each thread runs as fast as possible, but only a few threads

 GPUs have hundreds of simple cores
 CPUs have a few massive cores

 GPUs excel at regular math-intensive work
    Lots of ALUs for math, little hardware for control
 CPUs excel at irregular control-intensive work
    Lots of hardware for control, few ALUs
GPU Architecture Features
 Massively Parallel, 1000s of processors (today).
 Power Efficient:
    Fixed Function Hardware = area & power efficient.
    Lack of speculation. More processing, less leaky cache.
 Memory Bandwidth:
    Memory Bandwidth is limited in CPU.
    GPU is not dependent on large caches for performance.
 Computing power = Frequency * Transistors
    GPUs: 1.7X (pixels) to 2.3X (vertices) annual growth.
    CPUs: 1.4X annual growth.
Computational Power
CUDA
 CUDA = Computer Unified Device Architecture.
 A scalable parallel programming model and a software
  environment for parallel computing.
 Threads:
   GPU threads are extremely light-weight, with little creation
    overhead.
   GPU needs 1000s of threads for full efficiency.
   Multi-core CPU needed only a few.
 GPU/CUDA address all three levels of parallelism:
    Thread parallelism;
    Data parallelism;
    Task parallelism.
Summary
 All computers are now parallel computers!
 Multi-core processors represent an important new trend in
  computer architecture.
    Decreased power consumption and heat generation.
    Minimized wire lengths and interconnect latencies.
 They enable true thread-level parallelism with great energy
  efficiency and scalability.
 Graphics requires a lot of computation and a huge amount
  of bandwidth - GPUs.
 GPUs are coming to general purpose computing, since they
  deliver huge performance with small power.
THANK YOU

Weitere ähnliche Inhalte

Was ist angesagt?

Parallel computing with Gpu
Parallel computing with GpuParallel computing with Gpu
Parallel computing with GpuRohit Khatana
 
Gpu and The Brick Wall
Gpu and The Brick WallGpu and The Brick Wall
Gpu and The Brick Wallugur candan
 
Indian Contribution towards Parallel Processing
Indian Contribution towards Parallel ProcessingIndian Contribution towards Parallel Processing
Indian Contribution towards Parallel ProcessingAjil Jose
 
GPU Computing: A brief overview
GPU Computing: A brief overviewGPU Computing: A brief overview
GPU Computing: A brief overviewRajiv Kumar
 
Nvidia (History, GPU Architecture and New Pascal Architecture)
Nvidia (History, GPU Architecture and New Pascal Architecture)Nvidia (History, GPU Architecture and New Pascal Architecture)
Nvidia (History, GPU Architecture and New Pascal Architecture)Saksham Tanwar
 
Multi_Core_Processor_2015_(Download it!)
Multi_Core_Processor_2015_(Download it!)Multi_Core_Processor_2015_(Download it!)
Multi_Core_Processor_2015_(Download it!)Sudip Roy
 
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...Bharath Sudharsan
 
Quad Core Processors - Technology Presentation
Quad Core Processors - Technology PresentationQuad Core Processors - Technology Presentation
Quad Core Processors - Technology Presentationvinaya.hs
 
FIne Grain Multithreading
FIne Grain MultithreadingFIne Grain Multithreading
FIne Grain MultithreadingDharmesh Tank
 
Graphics processing unit ppt
Graphics processing unit pptGraphics processing unit ppt
Graphics processing unit pptSandeep Singh
 
Gpu with cuda architecture
Gpu with cuda architectureGpu with cuda architecture
Gpu with cuda architectureDhaval Kaneria
 
Multiprocessor Architecture for Image Processing
Multiprocessor Architecture for Image ProcessingMultiprocessor Architecture for Image Processing
Multiprocessor Architecture for Image Processingmayank.grd
 
Ximea - the pc camera, 90 gflps smart camera
Ximea  - the pc camera, 90 gflps smart cameraXimea  - the pc camera, 90 gflps smart camera
Ximea - the pc camera, 90 gflps smart cameraXIMEA
 

Was ist angesagt? (19)

Parallel computing with Gpu
Parallel computing with GpuParallel computing with Gpu
Parallel computing with Gpu
 
Gpu and The Brick Wall
Gpu and The Brick WallGpu and The Brick Wall
Gpu and The Brick Wall
 
Indian Contribution towards Parallel Processing
Indian Contribution towards Parallel ProcessingIndian Contribution towards Parallel Processing
Indian Contribution towards Parallel Processing
 
GPU Computing: A brief overview
GPU Computing: A brief overviewGPU Computing: A brief overview
GPU Computing: A brief overview
 
Nvidia (History, GPU Architecture and New Pascal Architecture)
Nvidia (History, GPU Architecture and New Pascal Architecture)Nvidia (History, GPU Architecture and New Pascal Architecture)
Nvidia (History, GPU Architecture and New Pascal Architecture)
 
Tensor Processing Unit (TPU)
Tensor Processing Unit (TPU)Tensor Processing Unit (TPU)
Tensor Processing Unit (TPU)
 
Multi_Core_Processor_2015_(Download it!)
Multi_Core_Processor_2015_(Download it!)Multi_Core_Processor_2015_(Download it!)
Multi_Core_Processor_2015_(Download it!)
 
Gpu databases
Gpu databasesGpu databases
Gpu databases
 
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...
ECML PKDD 2021 ML meets IoT Tutorial Part II: Creating ML based Self learning...
 
Quad Core Processors - Technology Presentation
Quad Core Processors - Technology PresentationQuad Core Processors - Technology Presentation
Quad Core Processors - Technology Presentation
 
Parallel Computing
Parallel Computing Parallel Computing
Parallel Computing
 
FIne Grain Multithreading
FIne Grain MultithreadingFIne Grain Multithreading
FIne Grain Multithreading
 
GPU Programming
GPU ProgrammingGPU Programming
GPU Programming
 
Graphics processing unit ppt
Graphics processing unit pptGraphics processing unit ppt
Graphics processing unit ppt
 
Gpu with cuda architecture
Gpu with cuda architectureGpu with cuda architecture
Gpu with cuda architecture
 
Multiprocessor Architecture for Image Processing
Multiprocessor Architecture for Image ProcessingMultiprocessor Architecture for Image Processing
Multiprocessor Architecture for Image Processing
 
Ximea - the pc camera, 90 gflps smart camera
Ximea  - the pc camera, 90 gflps smart cameraXimea  - the pc camera, 90 gflps smart camera
Ximea - the pc camera, 90 gflps smart camera
 
GPU Computing
GPU ComputingGPU Computing
GPU Computing
 
Google TPU
Google TPUGoogle TPU
Google TPU
 

Ähnlich wie Modern processor art

Stream Processing
Stream ProcessingStream Processing
Stream Processingarnamoy10
 
Multicore processor by Ankit Raj and Akash Prajapati
Multicore processor by Ankit Raj and Akash PrajapatiMulticore processor by Ankit Raj and Akash Prajapati
Multicore processor by Ankit Raj and Akash PrajapatiAnkit Raj
 
Intel new processors
Intel new processorsIntel new processors
Intel new processorszaid_b
 
Study of various factors affecting performance of multi core processors
Study of various factors affecting performance of multi core processorsStudy of various factors affecting performance of multi core processors
Study of various factors affecting performance of multi core processorsateeq ateeq
 
Maxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorialMaxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorialmadhuinturi
 
Computer architecture multi core processor
Computer architecture multi core processorComputer architecture multi core processor
Computer architecture multi core processorMazin Alwaaly
 
Blue gene- IBM's SuperComputer
Blue gene- IBM's SuperComputerBlue gene- IBM's SuperComputer
Blue gene- IBM's SuperComputerIsaaq Mohammed
 
The evolution of computer
The evolution of computerThe evolution of computer
The evolution of computerLolita De Leon
 
Implementation of RISC-Based Architecture for Low power applications
Implementation of RISC-Based Architecture for Low power applicationsImplementation of RISC-Based Architecture for Low power applications
Implementation of RISC-Based Architecture for Low power applicationsIOSR Journals
 
trends of microprocessor field
trends of microprocessor fieldtrends of microprocessor field
trends of microprocessor fieldRamya SK
 
Parallel and Distributed Computing chapter 3
Parallel and Distributed Computing chapter 3Parallel and Distributed Computing chapter 3
Parallel and Distributed Computing chapter 3AbdullahMunir32
 
Hyper threading technology
Hyper threading technologyHyper threading technology
Hyper threading technologyNikhil Venugopal
 
Area Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips ProcessorArea Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips ProcessorIOSR Journals
 

Ähnlich wie Modern processor art (20)

Stream Processing
Stream ProcessingStream Processing
Stream Processing
 
Multicore processor by Ankit Raj and Akash Prajapati
Multicore processor by Ankit Raj and Akash PrajapatiMulticore processor by Ankit Raj and Akash Prajapati
Multicore processor by Ankit Raj and Akash Prajapati
 
Intel new processors
Intel new processorsIntel new processors
Intel new processors
 
Distributed Computing
Distributed ComputingDistributed Computing
Distributed Computing
 
Module 1 unit 3
Module 1  unit 3Module 1  unit 3
Module 1 unit 3
 
Study of various factors affecting performance of multi core processors
Study of various factors affecting performance of multi core processorsStudy of various factors affecting performance of multi core processors
Study of various factors affecting performance of multi core processors
 
Maxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorialMaxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorial
 
Computer architecture multi core processor
Computer architecture multi core processorComputer architecture multi core processor
Computer architecture multi core processor
 
Blue gene- IBM's SuperComputer
Blue gene- IBM's SuperComputerBlue gene- IBM's SuperComputer
Blue gene- IBM's SuperComputer
 
Lecture1
Lecture1Lecture1
Lecture1
 
The evolution of computer
The evolution of computerThe evolution of computer
The evolution of computer
 
Implementation of RISC-Based Architecture for Low power applications
Implementation of RISC-Based Architecture for Low power applicationsImplementation of RISC-Based Architecture for Low power applications
Implementation of RISC-Based Architecture for Low power applications
 
trends of microprocessor field
trends of microprocessor fieldtrends of microprocessor field
trends of microprocessor field
 
Exascale Capabl
Exascale CapablExascale Capabl
Exascale Capabl
 
Multi-Core on Chip Architecture *doc - IK
Multi-Core on Chip Architecture *doc - IKMulti-Core on Chip Architecture *doc - IK
Multi-Core on Chip Architecture *doc - IK
 
CPU vs GPU Comparison
CPU  vs GPU ComparisonCPU  vs GPU Comparison
CPU vs GPU Comparison
 
High performance computing
High performance computingHigh performance computing
High performance computing
 
Parallel and Distributed Computing chapter 3
Parallel and Distributed Computing chapter 3Parallel and Distributed Computing chapter 3
Parallel and Distributed Computing chapter 3
 
Hyper threading technology
Hyper threading technologyHyper threading technology
Hyper threading technology
 
Area Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips ProcessorArea Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips Processor
 

Kürzlich hochgeladen

Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 

Kürzlich hochgeladen (20)

Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 

Modern processor art

  • 3. Introduction To Multi-Core System  Integration of multiple processor cores on a single chip.  Multi-core processor is a special kind of multiprocessors:  All processors are on the same chip also called Chip Multiprocessor.  Different cores execute different codes (threads) operating on different data.  A shared memory multiprocessor: all cores share the same memory via some cache organization  Provides a cheap parallel computer solution.  Increases the computation power to PC platform.
  • 4. Why Multi-Core?  Limitations of single core architectures:  High power consumption due to high clock rates (2-3% power increase per 1% performance increase).  Heat generation (cooling is expensive).  Limited parallelism (Instruction Level Parallelism only).  Design time and complexity increased due to complex methods to increase ILP.  Many new applications are multithreaded, suitable for multi-core.  Ex. Multimedia applications.  General trend in computer architecture (shift towards more parallelism).  Much faster cache coherency circuits, in a single chip.  Smaller in physical size than SMP.
  • 5. Single Core VS Multi Core
  • 6. A Dual-Core Intel Processor
  • 7. Intel Polaris  80 cores with Teraflop (10^12) performance on a single chip (1st chip to do so).  Mesh network-on-a-chip.  Frequency target at 5GHz.  Workload-aware power management:  Instructions to make any core sleep or wake as apps demand.  Chip voltage & frequency control.  Peak of 1.01 Teraflops at 62 watts.  Peak power efficiency of 19.4  Short design time.
  • 8. Innovations on Intel’s Polaris  Rapid design – The tiled-design approach allows designers to use smaller cores that can easily be repeated across the chip.  Network-on-a-chip – The cores are connected in a 2D mesh network that implement message-passing. This scheme is much more scalable.  Fine-grain power management - The individual compute engines and data routers in each core can be activated or put to sleep based on the performance required by the applications.
  • 9. What applications benefit From Multicore Processors  • Database servers  • Web servers (Web commerce)  • Compilers  • Multimedia applications  • Scientific applications,  • In general, applications with Thread-level parallelism(as opposed to instruction level parallelism)
  • 11. 2. Multi-Threading Thread-Level Parallelism (TLP)  This is parallelism on a more coarser scale than instruction-level parallelism (ILP).  Instruction stream divided into smaller streams (threads) to be executed in parallel.  Thread has its own instructions and data.  May be part of a parallel program or independent programs.  Each thread has all state (instructions, data, PC, etc.) needed to execute.  Single-core superscalar processors cannot fully exploit TLP.  Multi-core architectures can exploit TLP efficiently.  Use multiple instruction streams to improve the throughput of computers that run several programs .  TLP are more cost-effective to exploit than ILP.
  • 12. Threads vs. Processes  Process: An instance of a program running on a computer.  Resource ownership.  Virtual address space to hold process image including program, data, stack, and attributes.  Execution of a process follows a path though the program.  Process switch - An expensive operation due to the need to save the control data and register contents.  Thread: A dispatchable unit of work within a process.  Interruptible: processor can turn to another thread.  All threads within a process share code and data segments.  Thread switch is usually much less costly than process switch.
  • 13. Multithreading Approaches  Interleaved (fine-grained)  Processor deals with several thread contexts at a time.  Switching thread at each clock cycle (hardware need).  If a thread is blocked, it is skipped.  Hide latency of both short and long pipeline stalls.  Blocked (coarse-grained)  Thread executed until an event causes delay (e.g., cache miss).  Relieves the need to have very fast thread switching.  No slow down for ready-to-go threads.  Simultaneous (SMT)  Instructions simultaneously issued from multiple threads to execution units of superscalar processor.  Chip multiprocessing  Each processor handles separate threads.
  • 15. Programming for Multi-Core (In Context Of Multithreading)  There must be many threads or processes:  Multiple applications running on the same machine.  Multi-tasking is becoming very common.  OS software tends to run many threads as a part of its normal operation.  An application may also have multiple threads.  In most cases, it must be specifically written.  OS scheduler should map the threads to different cores, in order to balance the work load or to avoid hot spots due to heat generation.
  • 17. Introduction To GPU What is GPU? • It is a processor optimized for 2D/3D graphics, video, visual computing, and display. • It is highly parallel, highly multithreaded multiprocessor optimized for visual computing. • It provide real-time visual interaction with computed objects via graphics images, and video. • It serves as both a programmable graphics processor and a scalable parallel computing platform. • Heterogeneous Systems: combine a GPU with a CPU
  • 18. Graphics Processing Unit, Why? Graphics applications are:  Rendering of 2D or 3D images with complex optical effects;  Highly computation intensive;  Massively parallel;  Data stream based.  General purpose processors (CPU) are:  Designed to handle huge data volumes;  Serial, one operation at a time;  Control flow based;  High flexible, but not very well adapted to graphics.  GPUs are the solutions!
  • 19. GPU Design  Process pixels in parallel  2.3M pixels per frame  lots of work  All pixels are independent  no synchronization  Lots of spatial locality  regular memory access  Great speedups:  Limited only by the amount of hardware
  • 20. GPU Design 2) Focus on throughput, not latency Each pixel can take a long time……as long as we process many at the same time.  Great scalability  Lots of simple parallel processors  Low clock speed
  • 21. CPU vs. GPU Architecture  GPUs are throughput-optimized  Each thread may take a long time, but thousands of threads  CPUs are latency-optimized  Each thread runs as fast as possible, but only a few threads  GPUs have hundreds of simple cores  CPUs have a few massive cores  GPUs excel at regular math-intensive work  Lots of ALUs for math, little hardware for control  CPUs excel at irregular control-intensive work  Lots of hardware for control, few ALUs
  • 22. GPU Architecture Features  Massively Parallel, 1000s of processors (today).  Power Efficient:  Fixed Function Hardware = area & power efficient.  Lack of speculation. More processing, less leaky cache.  Memory Bandwidth:  Memory Bandwidth is limited in CPU.  GPU is not dependent on large caches for performance.  Computing power = Frequency * Transistors  GPUs: 1.7X (pixels) to 2.3X (vertices) annual growth.  CPUs: 1.4X annual growth.
  • 24. CUDA  CUDA = Computer Unified Device Architecture.  A scalable parallel programming model and a software environment for parallel computing.  Threads:  GPU threads are extremely light-weight, with little creation overhead.  GPU needs 1000s of threads for full efficiency.  Multi-core CPU needed only a few.  GPU/CUDA address all three levels of parallelism:  Thread parallelism;  Data parallelism;  Task parallelism.
  • 25. Summary  All computers are now parallel computers!  Multi-core processors represent an important new trend in computer architecture.  Decreased power consumption and heat generation.  Minimized wire lengths and interconnect latencies.  They enable true thread-level parallelism with great energy efficiency and scalability.  Graphics requires a lot of computation and a huge amount of bandwidth - GPUs.  GPUs are coming to general purpose computing, since they deliver huge performance with small power.

Hinweis der Redaktion

  1. smp
  2. L1, L2 cache