SlideShare a Scribd company logo
VEMANA INSTITUTE OF
TECHNOLOGY
(Affiliated to VTU, Approved by AICTE)
Koramangala, Bengaluru- 560034
“ARTIFICIAL INTELLIGENCE
HARDWARE”
Presented By
Name: Karthik
USN: 1VI17EC033
Under the Guidance of
Dr. Suneeta
Associate. Professor
Department of ECE, Vemana IT
Table of Contents
9 June 2021 Department of ECE, Vemana IT 2
• Abstract
• Introduction
• Literature survey
• Technical Aspects
• Applications
• Future Scope
• Conclusion
• References
ARTIFICIAL INTELLIGENCE HARDWARE
Abstract
9 June 2021 Department of ECE, Vemana IT 3
• Artificial Intelligence is one of the fastest growing field in tech
sector. Scope of AI is expanded to many fields including
healthcare, transport, security, entertainment, defense and
shopping.
• AI takes a large amount of data, and by processing that data it
recognizes patterns.
• Conventional CPUs are not enough to process large amount of
data for the AI applications. Hence we require a special
hardware called ‘AI accelerators’.
• In this seminar we are discussing about those AI accelerators.
ARTIFICIAL INTELLIGENCE HARDWARE
Introduction
9 June 2021 Department of ECE, Vemana IT 4
What is AI, Machine Learning and Deep Learning?
ARTIFICIAL INTELLIGENCE HARDWARE
Artificial Intelligence
Machine Learning
Deep Learning
fx
Fig:. AI
Where AI is used?
9 June 2021 Department of ECE, Vemana IT 5
• Astronomy
• Healthcare
• Gaming
• Finance
• Data Security
• Social Media
• Travel
• Automotive Industry
• E commerce
• Apps
ARTIFICIAL INTELLIGENCE HARDWARE
Literature survey
9 June 2021 Department of ECE, Vemana IT 6
PAPER NO.1: 2017
TITLE: “In-Datacenter Performance Analysis of a Tensor
Processing 𝐔𝐧𝐢𝐭𝐓𝐌”
AUTHORS: Norman P. Jouppi, Cliff Young, Nishant Patil and
others. Google, Inc., Mountain View, CA USA
DESCRIPTION:
• This paper evaluates an AI hardware called Tensor Processing
Unit (TPU).
• It compares performance of TPU to the Intel Haswell CPU and
Nvidia GPU.
ARTIFICIAL INTELLIGENCE HARDWARE
Literature survey
9 June 2021 Department of ECE, Vemana IT 7
PAPER NO.2: 2018
TITLE: “Hardware-Enabled Artificial Intelligence”
AUTHORS: William J. Dally, C. Thomas Gray, John Poulton,
Brucek Khailany and others. NVIDIA
DESCRIPTION:
• This paper discusses the circuit challenges in building deep-
learning hardware both for inference and for training.
• This paper also discusses about GPU system for deep NN.
ARTIFICIAL INTELLIGENCE HARDWARE
Literature survey
9 June 2021 Department of ECE, Vemana IT 8
PAPER NO.3: 2018
TITLE: “AI Hardware-Mapping the $100bn IT spending
opportunity”
AUTHORS: Toshiya Hari, Charles Long and others. Goldman S.
DESCRIPTION:
• This paper discusses the scope of AI in hardware industry.
ARTIFICIAL INTELLIGENCE HARDWARE
Neural Network
9 June 2021 Department of ECE, Vemana IT 9
ARTIFICIAL INTELLIGENCE HARDWARE
Fig:. Block diagram of neural network
Neural Network
9 June 2021 Department of ECE, Vemana IT 10
ARTIFICIAL INTELLIGENCE HARDWARE
0 1
0 1
2×2
1
f(𝒘𝟏𝟏𝒙𝟏 + 𝒘𝟏𝟐𝒙𝟐 + 𝒘𝟏𝟑𝒙𝟑 + ⋯ + ⋯ . ) =
𝒚
Neural Network
9 June 2021 Department of ECE, Vemana IT 11
ARTIFICIAL INTELLIGENCE HARDWARE
500×500
X56,675,768,…..
CPU
9 June 2021 Department of ECE, Vemana IT 12
ARTIFICIAL INTELLIGENCE HARDWARE
CALCULATED
RESULT
MEMORY
MULTIPLY AND ADD
• Von Neumann Architecture
Fig.: CPU (Central Processing Unit)
GPU
9 June 2021 Department of ECE, Vemana IT 13
ARTIFICIAL INTELLIGENCE HARDWARE
RESULT
Fig:. GPU( Graphics Processing Unit)
TPU
9 June 2021 Department of ECE, Vemana IT 14
ARTIFICIAL INTELLIGENCE HARDWARE
RESULT
Fig.: TPU (Tensor Processing Unit)
TPU
9 June 2021 Department of ECE, Vemana IT 15
ARTIFICIAL INTELLIGENCE HARDWARE
Fig.: TPU Printed Circuit Board
TPU
9 June 2021 Department of ECE, Vemana IT 16
ARTIFICIAL INTELLIGENCE HARDWARE
Fig.: TPU Block Diagram
Applications
9 June 2021 Department of ECE, Vemana IT 17
ARTIFICIAL INTELLIGENCE HARDWARE
• In Datacenters:
AWS
IBM
Google
Microsoft
Equinix
• In Consumer Electronics:
Fig.: Apple products Fig.: Snapdragon 888 5G
Applications
9 June 2021 Department of ECE, Vemana IT 18
ARTIFICIAL INTELLIGENCE HARDWARE
• In Edge AI:
Fig.: Nvidia Jetson Nano Fig.: Google Coral Fig:. Intel Movidius NC stick
Future Scope
9 June 2021 Department of ECE, Vemana IT 19
ARTIFICIAL INTELLIGENCE HARDWARE
• As per the goldman sachs Market research, TAM for AI
Hardware will reach $109bn by 2025.
Conclusion
9 June 2021 Department of ECE, Vemana IT 20
ARTIFICIAL INTELLIGENCE HARDWARE
Conclusion
9 June 2021 Department of ECE, Vemana IT 21
ARTIFICIAL INTELLIGENCE HARDWARE
TPU vs K80 GPU
• 25 times as many MACs than GPU
• 3.5 times more on chip memory than GPU
• Uses 50% less power for the same numbers of operations.
• 15 times as fast as the GPU
Disadvantages:
• Cannot be used in graphic processing.
• Requires optimized softwares.
References
9 June 2021 Department of ECE, Vemana IT 22
ARTIFICIAL INTELLIGENCE HARDWARE
1. Jouppi, N. P., Borchers, A., Boyle, R., Cantin, P., Chao, C., Clark, C., … Dean,
J. (2017). “In-Datacenter Performance Analysis of a Tensor Processing Unit.”
Proceedings of the 44th Annual International Symposium on Computer
Architecture - ISCA ’17. doi:10.1145/3079856.3080246 .
2. W. J. Dally, C. T. Gray, J. Poulton, B. Khailany, J. Wilson and L. Dennison,
"Hardware-Enabled Artificial Intelligence," 2018 IEEE Symposium on VLSI
Circuits, 2018, pp. 3-6, doi: 10.1109/VLSIC.2018.8502368.
3. Toshiya Hari, Charles Long, Daiki Takayama, Donald Lu, Ph. D., Mark
Delaney, CFA and Rod Hall. (2018). “AI Hardware-Mapping the $100bn IT
spending opportunity”. Goldman Sachs Global Investment Research.
4. www.cloud.google.com
5. www.nvidia.com
6. www.wikipedia.com
7. www.ark.intel.com
9 June 2021 Department of ECE, Vemana IT 23
ARTIFICIAL INTELLIGENCE HARDWARE
Thank you!

More Related Content

What's hot

HPC Top 5 Stories: March 29, 2017
HPC Top 5 Stories: March 29, 2017HPC Top 5 Stories: March 29, 2017
HPC Top 5 Stories: March 29, 2017
NVIDIA
 
Top 5 Deep Learning and AI Stories April 7th
Top 5 Deep Learning and AI Stories April 7th Top 5 Deep Learning and AI Stories April 7th
Top 5 Deep Learning and AI Stories April 7th
NVIDIA
 
Neural Networks Hardware Accelerators (An Introduction)
Neural Networks Hardware Accelerators (An Introduction)Neural Networks Hardware Accelerators (An Introduction)
Neural Networks Hardware Accelerators (An Introduction)
Hamidreza Bolhasani
 
Vertex perspectives ai optimized chipsets (part i)
Vertex perspectives   ai optimized chipsets (part i)Vertex perspectives   ai optimized chipsets (part i)
Vertex perspectives ai optimized chipsets (part i)
Yanai Oron
 
Vertex Perspectives | AI Optimized Chipsets | Part II
Vertex Perspectives | AI Optimized Chipsets | Part IIVertex Perspectives | AI Optimized Chipsets | Part II
Vertex Perspectives | AI Optimized Chipsets | Part II
Vertex Holdings
 
How Can AI and IoT Power the Chemical Industry?
How Can AI and IoT Power the Chemical Industry?How Can AI and IoT Power the Chemical Industry?
How Can AI and IoT Power the Chemical Industry?
Xiaonan Wang
 
AI in Healthcare 2017
AI in Healthcare 2017AI in Healthcare 2017
AI in Healthcare 2017
Peter Morgan
 
Presentation of the IMU / ICCS lab
Presentation of the IMU / ICCS labPresentation of the IMU / ICCS lab
Presentation of the IMU / ICCS lab
Gregoris Mentzas
 
9/23 Top 5 Deep Learning
9/23 Top 5 Deep Learning9/23 Top 5 Deep Learning
9/23 Top 5 Deep Learning
NVIDIA
 
Top 5 Deep Learning and AI Stories - November 30, 2018
Top 5 Deep Learning and AI Stories - November 30, 2018Top 5 Deep Learning and AI Stories - November 30, 2018
Top 5 Deep Learning and AI Stories - November 30, 2018
NVIDIA
 
Artificial intelligence in civil engineering seminar report
Artificial intelligence in civil engineering seminar reportArtificial intelligence in civil engineering seminar report
Artificial intelligence in civil engineering seminar report
DhanushS51
 
HPC Top 5 Stories: March 22, 2017
HPC Top 5 Stories: March 22, 2017HPC Top 5 Stories: March 22, 2017
HPC Top 5 Stories: March 22, 2017
NVIDIA
 
0831 presention
0831 presention0831 presention
0831 presention
AI.academy
 
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
Journal For Research
 
Exploring the Momentum: The Intersection of AI and HPC
Exploring the Momentum: The Intersection of AI and HPCExploring the Momentum: The Intersection of AI and HPC
Exploring the Momentum: The Intersection of AI and HPC
NVIDIA
 
GTC China 2017 Highlights
GTC China 2017 HighlightsGTC China 2017 Highlights
GTC China 2017 Highlights
NVIDIA
 
Transforming Healthcare at GTC Silicon Valley
Transforming Healthcare at GTC Silicon ValleyTransforming Healthcare at GTC Silicon Valley
Transforming Healthcare at GTC Silicon Valley
NVIDIA
 
Artificial intelligence in civil engineering technicial seminar ppt
Artificial intelligence in civil engineering technicial seminar pptArtificial intelligence in civil engineering technicial seminar ppt
Artificial intelligence in civil engineering technicial seminar ppt
DhanushS51
 
The Next Generation of AI and Deep Learning - GTC17
The Next Generation of AI and Deep Learning - GTC17The Next Generation of AI and Deep Learning - GTC17
The Next Generation of AI and Deep Learning - GTC17
NVIDIA
 
iThings-2012, Besançon, France, 20 November, 2012
iThings-2012, Besançon, France, 20 November, 2012iThings-2012, Besançon, France, 20 November, 2012
iThings-2012, Besançon, France, 20 November, 2012
Charith Perera
 

What's hot (20)

HPC Top 5 Stories: March 29, 2017
HPC Top 5 Stories: March 29, 2017HPC Top 5 Stories: March 29, 2017
HPC Top 5 Stories: March 29, 2017
 
Top 5 Deep Learning and AI Stories April 7th
Top 5 Deep Learning and AI Stories April 7th Top 5 Deep Learning and AI Stories April 7th
Top 5 Deep Learning and AI Stories April 7th
 
Neural Networks Hardware Accelerators (An Introduction)
Neural Networks Hardware Accelerators (An Introduction)Neural Networks Hardware Accelerators (An Introduction)
Neural Networks Hardware Accelerators (An Introduction)
 
Vertex perspectives ai optimized chipsets (part i)
Vertex perspectives   ai optimized chipsets (part i)Vertex perspectives   ai optimized chipsets (part i)
Vertex perspectives ai optimized chipsets (part i)
 
Vertex Perspectives | AI Optimized Chipsets | Part II
Vertex Perspectives | AI Optimized Chipsets | Part IIVertex Perspectives | AI Optimized Chipsets | Part II
Vertex Perspectives | AI Optimized Chipsets | Part II
 
How Can AI and IoT Power the Chemical Industry?
How Can AI and IoT Power the Chemical Industry?How Can AI and IoT Power the Chemical Industry?
How Can AI and IoT Power the Chemical Industry?
 
AI in Healthcare 2017
AI in Healthcare 2017AI in Healthcare 2017
AI in Healthcare 2017
 
Presentation of the IMU / ICCS lab
Presentation of the IMU / ICCS labPresentation of the IMU / ICCS lab
Presentation of the IMU / ICCS lab
 
9/23 Top 5 Deep Learning
9/23 Top 5 Deep Learning9/23 Top 5 Deep Learning
9/23 Top 5 Deep Learning
 
Top 5 Deep Learning and AI Stories - November 30, 2018
Top 5 Deep Learning and AI Stories - November 30, 2018Top 5 Deep Learning and AI Stories - November 30, 2018
Top 5 Deep Learning and AI Stories - November 30, 2018
 
Artificial intelligence in civil engineering seminar report
Artificial intelligence in civil engineering seminar reportArtificial intelligence in civil engineering seminar report
Artificial intelligence in civil engineering seminar report
 
HPC Top 5 Stories: March 22, 2017
HPC Top 5 Stories: March 22, 2017HPC Top 5 Stories: March 22, 2017
HPC Top 5 Stories: March 22, 2017
 
0831 presention
0831 presention0831 presention
0831 presention
 
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
ADVANCED CIVIL ENGINEERING OPTIMIZATION BY ARTIFICIAL INTELLIGENT SYSTEMS: RE...
 
Exploring the Momentum: The Intersection of AI and HPC
Exploring the Momentum: The Intersection of AI and HPCExploring the Momentum: The Intersection of AI and HPC
Exploring the Momentum: The Intersection of AI and HPC
 
GTC China 2017 Highlights
GTC China 2017 HighlightsGTC China 2017 Highlights
GTC China 2017 Highlights
 
Transforming Healthcare at GTC Silicon Valley
Transforming Healthcare at GTC Silicon ValleyTransforming Healthcare at GTC Silicon Valley
Transforming Healthcare at GTC Silicon Valley
 
Artificial intelligence in civil engineering technicial seminar ppt
Artificial intelligence in civil engineering technicial seminar pptArtificial intelligence in civil engineering technicial seminar ppt
Artificial intelligence in civil engineering technicial seminar ppt
 
The Next Generation of AI and Deep Learning - GTC17
The Next Generation of AI and Deep Learning - GTC17The Next Generation of AI and Deep Learning - GTC17
The Next Generation of AI and Deep Learning - GTC17
 
iThings-2012, Besançon, France, 20 November, 2012
iThings-2012, Besançon, France, 20 November, 2012iThings-2012, Besançon, France, 20 November, 2012
iThings-2012, Besançon, France, 20 November, 2012
 

Similar to 1VI17EC033_KARTHIK_TECHNICAL_SEMINAR_PPT.pptx

IRJET- Classification of Assembly (W-Section) using Artificial Intelligence
IRJET- Classification of Assembly (W-Section) using Artificial IntelligenceIRJET- Classification of Assembly (W-Section) using Artificial Intelligence
IRJET- Classification of Assembly (W-Section) using Artificial Intelligence
IRJET Journal
 
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
Sri Ambati
 
team12.project_ver_1_(1).pptx
team12.project_ver_1_(1).pptxteam12.project_ver_1_(1).pptx
team12.project_ver_1_(1).pptx
RitwikShrivastava1
 
Microcontrollers for Artificial Intelligence and Machine Learning
Microcontrollers for Artificial Intelligence and Machine LearningMicrocontrollers for Artificial Intelligence and Machine Learning
Microcontrollers for Artificial Intelligence and Machine Learning
IRJET Journal
 
Big Data Re-Told
Big Data Re-ToldBig Data Re-Told
Big Data Re-Told
Widy Widyawan
 
Software Realibility on the Big Data Era
Software Realibility on the Big Data EraSoftware Realibility on the Big Data Era
Software Realibility on the Big Data Era
Angel Conde Manjon
 
seminar ppt.pptx
seminar ppt.pptxseminar ppt.pptx
seminar ppt.pptx
SuprithC2
 
GTC 2017: The AI Revolution
GTC 2017: The AI RevolutionGTC 2017: The AI Revolution
GTC 2017: The AI Revolution
NVIDIA
 
Computer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
Computer Science Dissertation Topic Ideas For Phd Scholar - PhdassistanceComputer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
Computer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
PhD Assistance
 
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
Raybaen
 
Accelerating algorithmic and hardware advancements for power efficient on-dev...
Accelerating algorithmic and hardware advancements for power efficient on-dev...Accelerating algorithmic and hardware advancements for power efficient on-dev...
Accelerating algorithmic and hardware advancements for power efficient on-dev...
Qualcomm Research
 
Machine learning at the edge
Machine learning at the edgeMachine learning at the edge
Machine learning at the edge
Mehmet Ali Anıl
 
Pushing the boundaries of AI research
Pushing the boundaries of AI researchPushing the boundaries of AI research
Pushing the boundaries of AI research
Qualcomm Research
 
Chip design with AI inside—designed by AI
Chip design with AI inside—designed by AIChip design with AI inside—designed by AI
Chip design with AI inside—designed by AI
Abacus Technologies
 
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
IRJET Journal
 
Dl 0n mobile jeff shomaker_jan-2018_final
Dl 0n mobile jeff shomaker_jan-2018_finalDl 0n mobile jeff shomaker_jan-2018_final
Dl 0n mobile jeff shomaker_jan-2018_final
Jeffrey Shomaker
 
Internship report on AI , ML & IIOT and project responses
Internship report on AI , ML & IIOT and project responsesInternship report on AI , ML & IIOT and project responses
Internship report on AI , ML & IIOT and project responses
Rakesh Arigela
 
Dell AI and HPC University Roadshow
Dell AI and HPC University RoadshowDell AI and HPC University Roadshow
Dell AI and HPC University Roadshow
Bill Wong
 
Lecture_IIITD.pptx
Lecture_IIITD.pptxLecture_IIITD.pptx
Lecture_IIITD.pptx
achakracu
 
Artificial Intelligence in Small Embedded System
Artificial Intelligence in Small Embedded SystemArtificial Intelligence in Small Embedded System
Artificial Intelligence in Small Embedded System
GlobalLogic Ukraine
 

Similar to 1VI17EC033_KARTHIK_TECHNICAL_SEMINAR_PPT.pptx (20)

IRJET- Classification of Assembly (W-Section) using Artificial Intelligence
IRJET- Classification of Assembly (W-Section) using Artificial IntelligenceIRJET- Classification of Assembly (W-Section) using Artificial Intelligence
IRJET- Classification of Assembly (W-Section) using Artificial Intelligence
 
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
Meg Mude, Intel - Data Engineering Lifecycle Optimized on Intel - H2O World S...
 
team12.project_ver_1_(1).pptx
team12.project_ver_1_(1).pptxteam12.project_ver_1_(1).pptx
team12.project_ver_1_(1).pptx
 
Microcontrollers for Artificial Intelligence and Machine Learning
Microcontrollers for Artificial Intelligence and Machine LearningMicrocontrollers for Artificial Intelligence and Machine Learning
Microcontrollers for Artificial Intelligence and Machine Learning
 
Big Data Re-Told
Big Data Re-ToldBig Data Re-Told
Big Data Re-Told
 
Software Realibility on the Big Data Era
Software Realibility on the Big Data EraSoftware Realibility on the Big Data Era
Software Realibility on the Big Data Era
 
seminar ppt.pptx
seminar ppt.pptxseminar ppt.pptx
seminar ppt.pptx
 
GTC 2017: The AI Revolution
GTC 2017: The AI RevolutionGTC 2017: The AI Revolution
GTC 2017: The AI Revolution
 
Computer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
Computer Science Dissertation Topic Ideas For Phd Scholar - PhdassistanceComputer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
Computer Science Dissertation Topic Ideas For Phd Scholar - Phdassistance
 
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
Rajat Bandejiya(14uec076)Lusip (Smart Campus) Report
 
Accelerating algorithmic and hardware advancements for power efficient on-dev...
Accelerating algorithmic and hardware advancements for power efficient on-dev...Accelerating algorithmic and hardware advancements for power efficient on-dev...
Accelerating algorithmic and hardware advancements for power efficient on-dev...
 
Machine learning at the edge
Machine learning at the edgeMachine learning at the edge
Machine learning at the edge
 
Pushing the boundaries of AI research
Pushing the boundaries of AI researchPushing the boundaries of AI research
Pushing the boundaries of AI research
 
Chip design with AI inside—designed by AI
Chip design with AI inside—designed by AIChip design with AI inside—designed by AI
Chip design with AI inside—designed by AI
 
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
IRJET- Implementing Automatic Object Categorizing by Applying Deep Learning a...
 
Dl 0n mobile jeff shomaker_jan-2018_final
Dl 0n mobile jeff shomaker_jan-2018_finalDl 0n mobile jeff shomaker_jan-2018_final
Dl 0n mobile jeff shomaker_jan-2018_final
 
Internship report on AI , ML & IIOT and project responses
Internship report on AI , ML & IIOT and project responsesInternship report on AI , ML & IIOT and project responses
Internship report on AI , ML & IIOT and project responses
 
Dell AI and HPC University Roadshow
Dell AI and HPC University RoadshowDell AI and HPC University Roadshow
Dell AI and HPC University Roadshow
 
Lecture_IIITD.pptx
Lecture_IIITD.pptxLecture_IIITD.pptx
Lecture_IIITD.pptx
 
Artificial Intelligence in Small Embedded System
Artificial Intelligence in Small Embedded SystemArtificial Intelligence in Small Embedded System
Artificial Intelligence in Small Embedded System
 

Recently uploaded

Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
Kari Kakkonen
 
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdfObservability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
Paige Cruz
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
James Anderson
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
Safe Software
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
sonjaschweigert1
 
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
Neo4j
 
By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024
Pierluigi Pugliese
 
Removing Uninteresting Bytes in Software Fuzzing
Removing Uninteresting Bytes in Software FuzzingRemoving Uninteresting Bytes in Software Fuzzing
Removing Uninteresting Bytes in Software Fuzzing
Aftab Hussain
 
20240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 202420240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 2024
Matthew Sinclair
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
Mind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AIMind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AI
Kumud Singh
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
KAMESHS29
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
Uni Systems S.M.S.A.
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
Zilliz
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
Ana-Maria Mihalceanu
 
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
Neo4j
 
PCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase TeamPCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase Team
ControlCase
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
KatiaHIMEUR1
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Albert Hoitingh
 
Epistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI supportEpistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI support
Alan Dix
 

Recently uploaded (20)

Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
 
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdfObservability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdf
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
 
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
 
By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024By Design, not by Accident - Agile Venture Bolzano 2024
By Design, not by Accident - Agile Venture Bolzano 2024
 
Removing Uninteresting Bytes in Software Fuzzing
Removing Uninteresting Bytes in Software FuzzingRemoving Uninteresting Bytes in Software Fuzzing
Removing Uninteresting Bytes in Software Fuzzing
 
20240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 202420240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 2024
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
Mind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AIMind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AI
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
 
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
GraphSummit Singapore | Neo4j Product Vision & Roadmap - Q2 2024
 
PCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase TeamPCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase Team
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
 
Epistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI supportEpistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI support
 

1VI17EC033_KARTHIK_TECHNICAL_SEMINAR_PPT.pptx

  • 1. VEMANA INSTITUTE OF TECHNOLOGY (Affiliated to VTU, Approved by AICTE) Koramangala, Bengaluru- 560034 “ARTIFICIAL INTELLIGENCE HARDWARE” Presented By Name: Karthik USN: 1VI17EC033 Under the Guidance of Dr. Suneeta Associate. Professor Department of ECE, Vemana IT
  • 2. Table of Contents 9 June 2021 Department of ECE, Vemana IT 2 • Abstract • Introduction • Literature survey • Technical Aspects • Applications • Future Scope • Conclusion • References ARTIFICIAL INTELLIGENCE HARDWARE
  • 3. Abstract 9 June 2021 Department of ECE, Vemana IT 3 • Artificial Intelligence is one of the fastest growing field in tech sector. Scope of AI is expanded to many fields including healthcare, transport, security, entertainment, defense and shopping. • AI takes a large amount of data, and by processing that data it recognizes patterns. • Conventional CPUs are not enough to process large amount of data for the AI applications. Hence we require a special hardware called ‘AI accelerators’. • In this seminar we are discussing about those AI accelerators. ARTIFICIAL INTELLIGENCE HARDWARE
  • 4. Introduction 9 June 2021 Department of ECE, Vemana IT 4 What is AI, Machine Learning and Deep Learning? ARTIFICIAL INTELLIGENCE HARDWARE Artificial Intelligence Machine Learning Deep Learning fx Fig:. AI
  • 5. Where AI is used? 9 June 2021 Department of ECE, Vemana IT 5 • Astronomy • Healthcare • Gaming • Finance • Data Security • Social Media • Travel • Automotive Industry • E commerce • Apps ARTIFICIAL INTELLIGENCE HARDWARE
  • 6. Literature survey 9 June 2021 Department of ECE, Vemana IT 6 PAPER NO.1: 2017 TITLE: “In-Datacenter Performance Analysis of a Tensor Processing 𝐔𝐧𝐢𝐭𝐓𝐌” AUTHORS: Norman P. Jouppi, Cliff Young, Nishant Patil and others. Google, Inc., Mountain View, CA USA DESCRIPTION: • This paper evaluates an AI hardware called Tensor Processing Unit (TPU). • It compares performance of TPU to the Intel Haswell CPU and Nvidia GPU. ARTIFICIAL INTELLIGENCE HARDWARE
  • 7. Literature survey 9 June 2021 Department of ECE, Vemana IT 7 PAPER NO.2: 2018 TITLE: “Hardware-Enabled Artificial Intelligence” AUTHORS: William J. Dally, C. Thomas Gray, John Poulton, Brucek Khailany and others. NVIDIA DESCRIPTION: • This paper discusses the circuit challenges in building deep- learning hardware both for inference and for training. • This paper also discusses about GPU system for deep NN. ARTIFICIAL INTELLIGENCE HARDWARE
  • 8. Literature survey 9 June 2021 Department of ECE, Vemana IT 8 PAPER NO.3: 2018 TITLE: “AI Hardware-Mapping the $100bn IT spending opportunity” AUTHORS: Toshiya Hari, Charles Long and others. Goldman S. DESCRIPTION: • This paper discusses the scope of AI in hardware industry. ARTIFICIAL INTELLIGENCE HARDWARE
  • 9. Neural Network 9 June 2021 Department of ECE, Vemana IT 9 ARTIFICIAL INTELLIGENCE HARDWARE Fig:. Block diagram of neural network
  • 10. Neural Network 9 June 2021 Department of ECE, Vemana IT 10 ARTIFICIAL INTELLIGENCE HARDWARE 0 1 0 1 2×2 1 f(𝒘𝟏𝟏𝒙𝟏 + 𝒘𝟏𝟐𝒙𝟐 + 𝒘𝟏𝟑𝒙𝟑 + ⋯ + ⋯ . ) = 𝒚
  • 11. Neural Network 9 June 2021 Department of ECE, Vemana IT 11 ARTIFICIAL INTELLIGENCE HARDWARE 500×500 X56,675,768,…..
  • 12. CPU 9 June 2021 Department of ECE, Vemana IT 12 ARTIFICIAL INTELLIGENCE HARDWARE CALCULATED RESULT MEMORY MULTIPLY AND ADD • Von Neumann Architecture Fig.: CPU (Central Processing Unit)
  • 13. GPU 9 June 2021 Department of ECE, Vemana IT 13 ARTIFICIAL INTELLIGENCE HARDWARE RESULT Fig:. GPU( Graphics Processing Unit)
  • 14. TPU 9 June 2021 Department of ECE, Vemana IT 14 ARTIFICIAL INTELLIGENCE HARDWARE RESULT Fig.: TPU (Tensor Processing Unit)
  • 15. TPU 9 June 2021 Department of ECE, Vemana IT 15 ARTIFICIAL INTELLIGENCE HARDWARE Fig.: TPU Printed Circuit Board
  • 16. TPU 9 June 2021 Department of ECE, Vemana IT 16 ARTIFICIAL INTELLIGENCE HARDWARE Fig.: TPU Block Diagram
  • 17. Applications 9 June 2021 Department of ECE, Vemana IT 17 ARTIFICIAL INTELLIGENCE HARDWARE • In Datacenters: AWS IBM Google Microsoft Equinix • In Consumer Electronics: Fig.: Apple products Fig.: Snapdragon 888 5G
  • 18. Applications 9 June 2021 Department of ECE, Vemana IT 18 ARTIFICIAL INTELLIGENCE HARDWARE • In Edge AI: Fig.: Nvidia Jetson Nano Fig.: Google Coral Fig:. Intel Movidius NC stick
  • 19. Future Scope 9 June 2021 Department of ECE, Vemana IT 19 ARTIFICIAL INTELLIGENCE HARDWARE • As per the goldman sachs Market research, TAM for AI Hardware will reach $109bn by 2025.
  • 20. Conclusion 9 June 2021 Department of ECE, Vemana IT 20 ARTIFICIAL INTELLIGENCE HARDWARE
  • 21. Conclusion 9 June 2021 Department of ECE, Vemana IT 21 ARTIFICIAL INTELLIGENCE HARDWARE TPU vs K80 GPU • 25 times as many MACs than GPU • 3.5 times more on chip memory than GPU • Uses 50% less power for the same numbers of operations. • 15 times as fast as the GPU Disadvantages: • Cannot be used in graphic processing. • Requires optimized softwares.
  • 22. References 9 June 2021 Department of ECE, Vemana IT 22 ARTIFICIAL INTELLIGENCE HARDWARE 1. Jouppi, N. P., Borchers, A., Boyle, R., Cantin, P., Chao, C., Clark, C., … Dean, J. (2017). “In-Datacenter Performance Analysis of a Tensor Processing Unit.” Proceedings of the 44th Annual International Symposium on Computer Architecture - ISCA ’17. doi:10.1145/3079856.3080246 . 2. W. J. Dally, C. T. Gray, J. Poulton, B. Khailany, J. Wilson and L. Dennison, "Hardware-Enabled Artificial Intelligence," 2018 IEEE Symposium on VLSI Circuits, 2018, pp. 3-6, doi: 10.1109/VLSIC.2018.8502368. 3. Toshiya Hari, Charles Long, Daiki Takayama, Donald Lu, Ph. D., Mark Delaney, CFA and Rod Hall. (2018). “AI Hardware-Mapping the $100bn IT spending opportunity”. Goldman Sachs Global Investment Research. 4. www.cloud.google.com 5. www.nvidia.com 6. www.wikipedia.com 7. www.ark.intel.com
  • 23. 9 June 2021 Department of ECE, Vemana IT 23 ARTIFICIAL INTELLIGENCE HARDWARE Thank you!