SlideShare une entreprise Scribd logo
1  sur  28
Field Programmable Gate Arrays
in Finance
14/Oct/2014Antonio Roldao, Ph.D.
Presentation Overview
• What are FPGAs
• How do FPGAs work
• Why are FPGAs in-hype
• Where are FPGAs being used
• FPGAs in Finance
• Summary
• Future Vision
Antonio Roldao, Ph.D. 2
What are FPGAs?
• Invented in 1985 by Ross Freeman
• Re-programmable Digital Device
• Equivalent to employing millions of logic gates
• Can be configured in the outside in the “field”
• Able to implement
• very simple logic; or
• fully featured systems like multi-core-
microprocessors w/ RAM, video, and sound,
neural networks, trading systems, …
Antonio Roldao, Ph.D. 3
How do FPGAs work?
Antonio Roldao, Ph.D. 4
How do FPGAs work: Look-Up-Tables - 1/2
Antonio Roldao, Ph.D. 5
How do FPGAs work: Look-Up-Tables - 2/2
Antonio Roldao, Ph.D. 6
How do FPGAs work: Routing Fabric
Antonio Roldao, Ph.D. 7
How do FPGAs work: Embedded Cores
Antonio Roldao, Ph.D. 8
Why are FPGAs in-hype?
Antonio Roldao, Ph.D. 9
Why are FPGA in-hype: CPU’s Trends
Antonio Roldao, Ph.D. 10
Why are FPGA in-hype: FPGA’s Density
Antonio Roldao, Ph.D. 11
Why are FPGA in-hype: FPGA’s FP Capability
Antonio Roldao, Ph.D. 12
Why are FPGA in-hype: Better Tools
• Better Integrated Development Environments
that detect and flag most common mistakes
• Integrated simulation
including embeddable probes that can read live FPGA signals
• Faster synthesis and bitstream generation
that can infer standard embedded blocks (e.g. RAMs/Multipliers)
• Comprehensive libs of commonly used logic blocks
with highly optimized designs targeting the specific FPGA
• Hybrid software and hardware methodologies
using embedded microprocessors (e.g. ARM)
Antonio Roldao, Ph.D. 13
Why are FPGA in-hype: Gartner Hype Cycle
Antonio Roldao, Ph.D. 14
Where are FPGAs being used?
Antonio Roldao, Ph.D. 15
• Spatial Parallelizable - Monte Carlo, Map-Reduce, Neural Nets
• Temporal Parallelizable - Monte Carlo, Streaming, Encryption
• Deterministic - Precise Measurements and Actuators
• Low Power - Space Probes, Cameras, Hard-disks
• High Integration - Satellite Systems, F1 Steering Wheel
• High Throughput - Video Encoding/Decoding, Filtering
• Customizable - Data-paths, Number Systems, ISAs
• Wire-Speed / Low-Latency - Software Defined Networks, HFTs
• Architecture Independence
General FPGA Applications
Antonio Roldao, Ph.D. 16
FPGAs in Finance
Antonio Roldao, Ph.D. 17
Why is speed so important?
• Winner eventually takes it all due to:
Positive Network Effects
Antonio Roldao, Ph.D. 18
Simplified Client-side HFT Platform
Antonio Roldao, Ph.D. 19
Simplified Exchange-side Platform
Antonio Roldao, Ph.D. 20
1/5 - FPGA Applications in Finance
• Off-loading Network Computations
• Checksum calculations
• ARP Requests/Replies
• UDP packet crafting
• TCP packet crafting, sliding window, re-request, throttling
• Protocol Normalization / Transliteration / Tagging / Stamping
• Time stamping
• Tagging with Unique Identifier
• Transliteration from {b}FIX/ProtoBuffers/SBE/ASCII into binary and
vice-versa
• Façade for different protocols
Antonio Roldao, Ph.D. 21
• Wire Speed Logging
• Capture traffic in real-time
• Non-intrusive logging
• Reactive alerts
• Network traffic replay and compare
• Market Data Processing
• Manage connections to various data sources
• Transliterate and compact data
• A/B arbitration
• Aggregate multiple feeds into a unified stream
• Filter non-subscribed names
Antonio Roldao, Ph.D. 22
2/5 - FPGA Applications in Finance
• Rules and Limit Checks
• Verify an order is within market ranges
• Track and limit exposure in real-time
• Verify order complies with regulatory requirements
• Smart Order Routing
• Route by lowest latency
• Route by best price
• Route by increased liquidity
• Route by lowest transaction fees
• Route by highest historical success rate
• Route by a combination of above or any other factor
Antonio Roldao, Ph.D. 23
3/5 - FPGA Applications in Finance
• Risk Calculation
• Instrument pricing
• Scenario building and evaluation
• Monte-Carlo simulations
• Real-time portfolio / PnL calculations
• Risk Mitigation and Hedging Strategies
• Generate alerts for certain risky conditions
• React to certain risky conditions
• Generate alerts when opportunities for hedging arise
• Automatically rebalance portfolios to minimize exposure
Antonio Roldao, Ph.D. 24
4/5 - FPGA Applications in Finance
• Matching and Crossing Engines
• Sort orders in parallel
• Match many and rank trades in parallel
• Preemptively calculate crossing results
• Generate market data feeds
• Reply to both sides of the trade in parallel
• Money Making Engines
• Arbitrage strategies
• Momentum strategies
• Market making strategies
• Statistical and pattern-based strategies
Antonio Roldao, Ph.D. 25
5/5 - FPGA Applications in Finance
Summary
BASIC INTERMEDIATE COMPLEX
Packet Filtering Various Checks Real-time PnL
Protocol Transliteration Market Data Consolidation Trading Strategies
A/B Arbitration Order Routing Matching Engines
UDP/TCP Off-Loading Risk Calculation
• There are a wide range of applications
• Winner eventually takes it all due to:
Positive Network Effects
Antonio Roldao, Ph.D. 26
Future Vision
Trading-System-on-a-Chip
Antonio Roldao, Ph.D. 27
Thank You
http://anton.io
@roldao

Contenu connexe

Similaire à FPGA Applications in Finance

An introduction to RFID
An introduction to RFIDAn introduction to RFID
An introduction to RFID
Miguel Pardal
 
Towards the Internet of Things: An introduction to RFID
Towards the Internet of Things: An introduction to RFIDTowards the Internet of Things: An introduction to RFID
Towards the Internet of Things: An introduction to RFID
Miguel Pardal
 

Similaire à FPGA Applications in Finance (20)

Fpga Device Selection
Fpga Device SelectionFpga Device Selection
Fpga Device Selection
 
In-house OAuth/OIDC Infrastructure as a Competitive Advantage #eic2021
In-house OAuth/OIDC Infrastructure as a Competitive Advantage #eic2021In-house OAuth/OIDC Infrastructure as a Competitive Advantage #eic2021
In-house OAuth/OIDC Infrastructure as a Competitive Advantage #eic2021
 
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision SystemHai Tao at AI Frontiers: Deep Learning For Embedded Vision System
Hai Tao at AI Frontiers: Deep Learning For Embedded Vision System
 
Why FPGA
Why FPGAWhy FPGA
Why FPGA
 
Unified Threat Management
Unified Threat ManagementUnified Threat Management
Unified Threat Management
 
FPGA Selection Methodology for Real time projects
FPGA Selection Methodology for Real time projectsFPGA Selection Methodology for Real time projects
FPGA Selection Methodology for Real time projects
 
Avi solution2
Avi solution2Avi solution2
Avi solution2
 
An introduction to RFID
An introduction to RFIDAn introduction to RFID
An introduction to RFID
 
Myles firewalls
Myles firewallsMyles firewalls
Myles firewalls
 
Striving for ultimate Low Latency
Striving for ultimate Low LatencyStriving for ultimate Low Latency
Striving for ultimate Low Latency
 
MIPI DevCon 2016: Image Sensor and Display Connectivity Disruption
MIPI DevCon 2016: Image Sensor and Display Connectivity DisruptionMIPI DevCon 2016: Image Sensor and Display Connectivity Disruption
MIPI DevCon 2016: Image Sensor and Display Connectivity Disruption
 
ITRI ICL LTE SmallCell & Multi-RAT G/W 技術介紹 20140529
ITRI ICL LTE SmallCell & Multi-RAT G/W 技術介紹 20140529ITRI ICL LTE SmallCell & Multi-RAT G/W 技術介紹 20140529
ITRI ICL LTE SmallCell & Multi-RAT G/W 技術介紹 20140529
 
Tego's CEO Presentation at Woodside on August 15th, 2014
Tego's CEO Presentation at Woodside on August 15th, 2014Tego's CEO Presentation at Woodside on August 15th, 2014
Tego's CEO Presentation at Woodside on August 15th, 2014
 
ODSA PHY Layer
ODSA PHY LayerODSA PHY Layer
ODSA PHY Layer
 
ODSA - PHY Layer
ODSA - PHY LayerODSA - PHY Layer
ODSA - PHY Layer
 
RFID Technology
RFID TechnologyRFID Technology
RFID Technology
 
Toward a Methodology to turn Smalltak code into FPGA
Toward a Methodology to turn Smalltak code into FPGAToward a Methodology to turn Smalltak code into FPGA
Toward a Methodology to turn Smalltak code into FPGA
 
PLNOG14: Fortinet, Carrier and MSSP - Robert Dąbrowski
PLNOG14: Fortinet, Carrier and MSSP - Robert DąbrowskiPLNOG14: Fortinet, Carrier and MSSP - Robert Dąbrowski
PLNOG14: Fortinet, Carrier and MSSP - Robert Dąbrowski
 
Towards the Internet of Things: An introduction to RFID
Towards the Internet of Things: An introduction to RFIDTowards the Internet of Things: An introduction to RFID
Towards the Internet of Things: An introduction to RFID
 
RFID and NFC Providing the last yards for IoT
RFID and NFC Providing the last yards for IoTRFID and NFC Providing the last yards for IoT
RFID and NFC Providing the last yards for IoT
 

Dernier

Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
vu2urc
 

Dernier (20)

A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of Brazil
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
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
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
[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
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
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...
 
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
 

FPGA Applications in Finance

  • 1. Field Programmable Gate Arrays in Finance 14/Oct/2014Antonio Roldao, Ph.D.
  • 2. Presentation Overview • What are FPGAs • How do FPGAs work • Why are FPGAs in-hype • Where are FPGAs being used • FPGAs in Finance • Summary • Future Vision Antonio Roldao, Ph.D. 2
  • 3. What are FPGAs? • Invented in 1985 by Ross Freeman • Re-programmable Digital Device • Equivalent to employing millions of logic gates • Can be configured in the outside in the “field” • Able to implement • very simple logic; or • fully featured systems like multi-core- microprocessors w/ RAM, video, and sound, neural networks, trading systems, … Antonio Roldao, Ph.D. 3
  • 4. How do FPGAs work? Antonio Roldao, Ph.D. 4
  • 5. How do FPGAs work: Look-Up-Tables - 1/2 Antonio Roldao, Ph.D. 5
  • 6. How do FPGAs work: Look-Up-Tables - 2/2 Antonio Roldao, Ph.D. 6
  • 7. How do FPGAs work: Routing Fabric Antonio Roldao, Ph.D. 7
  • 8. How do FPGAs work: Embedded Cores Antonio Roldao, Ph.D. 8
  • 9. Why are FPGAs in-hype? Antonio Roldao, Ph.D. 9
  • 10. Why are FPGA in-hype: CPU’s Trends Antonio Roldao, Ph.D. 10
  • 11. Why are FPGA in-hype: FPGA’s Density Antonio Roldao, Ph.D. 11
  • 12. Why are FPGA in-hype: FPGA’s FP Capability Antonio Roldao, Ph.D. 12
  • 13. Why are FPGA in-hype: Better Tools • Better Integrated Development Environments that detect and flag most common mistakes • Integrated simulation including embeddable probes that can read live FPGA signals • Faster synthesis and bitstream generation that can infer standard embedded blocks (e.g. RAMs/Multipliers) • Comprehensive libs of commonly used logic blocks with highly optimized designs targeting the specific FPGA • Hybrid software and hardware methodologies using embedded microprocessors (e.g. ARM) Antonio Roldao, Ph.D. 13
  • 14. Why are FPGA in-hype: Gartner Hype Cycle Antonio Roldao, Ph.D. 14
  • 15. Where are FPGAs being used? Antonio Roldao, Ph.D. 15
  • 16. • Spatial Parallelizable - Monte Carlo, Map-Reduce, Neural Nets • Temporal Parallelizable - Monte Carlo, Streaming, Encryption • Deterministic - Precise Measurements and Actuators • Low Power - Space Probes, Cameras, Hard-disks • High Integration - Satellite Systems, F1 Steering Wheel • High Throughput - Video Encoding/Decoding, Filtering • Customizable - Data-paths, Number Systems, ISAs • Wire-Speed / Low-Latency - Software Defined Networks, HFTs • Architecture Independence General FPGA Applications Antonio Roldao, Ph.D. 16
  • 17. FPGAs in Finance Antonio Roldao, Ph.D. 17
  • 18. Why is speed so important? • Winner eventually takes it all due to: Positive Network Effects Antonio Roldao, Ph.D. 18
  • 19. Simplified Client-side HFT Platform Antonio Roldao, Ph.D. 19
  • 21. 1/5 - FPGA Applications in Finance • Off-loading Network Computations • Checksum calculations • ARP Requests/Replies • UDP packet crafting • TCP packet crafting, sliding window, re-request, throttling • Protocol Normalization / Transliteration / Tagging / Stamping • Time stamping • Tagging with Unique Identifier • Transliteration from {b}FIX/ProtoBuffers/SBE/ASCII into binary and vice-versa • Façade for different protocols Antonio Roldao, Ph.D. 21
  • 22. • Wire Speed Logging • Capture traffic in real-time • Non-intrusive logging • Reactive alerts • Network traffic replay and compare • Market Data Processing • Manage connections to various data sources • Transliterate and compact data • A/B arbitration • Aggregate multiple feeds into a unified stream • Filter non-subscribed names Antonio Roldao, Ph.D. 22 2/5 - FPGA Applications in Finance
  • 23. • Rules and Limit Checks • Verify an order is within market ranges • Track and limit exposure in real-time • Verify order complies with regulatory requirements • Smart Order Routing • Route by lowest latency • Route by best price • Route by increased liquidity • Route by lowest transaction fees • Route by highest historical success rate • Route by a combination of above or any other factor Antonio Roldao, Ph.D. 23 3/5 - FPGA Applications in Finance
  • 24. • Risk Calculation • Instrument pricing • Scenario building and evaluation • Monte-Carlo simulations • Real-time portfolio / PnL calculations • Risk Mitigation and Hedging Strategies • Generate alerts for certain risky conditions • React to certain risky conditions • Generate alerts when opportunities for hedging arise • Automatically rebalance portfolios to minimize exposure Antonio Roldao, Ph.D. 24 4/5 - FPGA Applications in Finance
  • 25. • Matching and Crossing Engines • Sort orders in parallel • Match many and rank trades in parallel • Preemptively calculate crossing results • Generate market data feeds • Reply to both sides of the trade in parallel • Money Making Engines • Arbitrage strategies • Momentum strategies • Market making strategies • Statistical and pattern-based strategies Antonio Roldao, Ph.D. 25 5/5 - FPGA Applications in Finance
  • 26. Summary BASIC INTERMEDIATE COMPLEX Packet Filtering Various Checks Real-time PnL Protocol Transliteration Market Data Consolidation Trading Strategies A/B Arbitration Order Routing Matching Engines UDP/TCP Off-Loading Risk Calculation • There are a wide range of applications • Winner eventually takes it all due to: Positive Network Effects Antonio Roldao, Ph.D. 26