SlideShare a Scribd company logo
InsideHPC
Dr Nick New, CEO and Founder
Our vision is to revolutionise big
data processing and model
generation through optical
processing technology
© 2017 OPTALYSYS LTD 3
Trends towards specialist processors in
the global computing community
Demand for faster processing
outstrips HPC’s capability to deliver
“Science is evolving faster than IT can refresh
infrastructure, design patterns and operational
practices”
Chris Dagdigian from Bioteam’s “Trend from the Trenches” talk
at BioIT World April 2016
“Moore’s Law is dead says Gordon Moore”
Techworld, 2010
“Specialisation proliferates again. Custom
architectures will reemerge in the market”
Top 5 trends for next 10 years in HPC,
Intersect360 Research Feb 2017
“Accelerated computing has reached the tipping
point in HPC”
Intersect360 Research, 2015
“There’s a concerted move to throughput
processors in the datacenter”
Michael Feldman, Top500, Mar 2017
Move towards specialist systems
designed to perform specific tasks
Optical Fourier Processing
5
Optical Processing and Liquid Crystal SLMs
Input Optics
f f
S(u,v)
l Output
s(x,y)
• Liquid Crystal Spatial Light Modulators (SLMs) used to enter numerical data
• Parallel, speed of light processing
• Highy scalable - Process time not dependent on resolution
• Low power laser light
• Order 1 operation
6
Correlation Pattern Recognition and Equation SolvingLASER
CAMERA
Input data Reference data Output data
Laser Input SLM FT lens Reference SLM FT lens
FT IFTMM
𝑔 𝑥 ∗ 𝑟 𝑥 = 𝐼𝐹𝑇[𝐺 𝑢 𝑅 𝑢 ∗
]
© 2017 OPTALYSYS LTD 7
Presenting Optalysys’s highly scalable, high-speed,
low power optical co-processing technology
© 2017 OPTALYSYS LTD 8
• Multiple SLMs and cameras
• Optical Correlation using
diffractive optical process
• Laser light projected
through the system as the
information carrier
• Optical system mounted to
a driver PCIe card
• Open API Interface
• Highly scalable system
SLMs
PCIe interface
Laser
Cameras
Glass
How it works
Application Areas
© 2017 OPTALYSYS LTD 10
Key sectors demanding high performance computing
Our optical approach will herald a step-change, benefitting
multiple applications within each of these fields
Machine Learning Aerodynamics
Signal
Processing
Weather
Forecasting
Financial
Analysis
Bioinformatics
and Medical
Analysis
Defence and
intelligence
Oil and Energy
11
Solutions for targeted key markets with strategic partners
• Genome assembly / alignment
• InnovateUK
• Next generation weather forecasting
• EU Horizon2020
• Equation solving fundamentals
• DARPA ACCESS program
Focus Areas
© 2017 OPTALYSYS LTD 12
DARPA Seedling Project
• Completed in May 2017
• Successfully demonstrated the
principles of an optical Fourier
transform co-processor for
equation solving
• 13 month “Seedling Project”
• Part of DARPA-SN-15-31: Call for
New Innovations towards:
Analog and Continuous-variable
Co-processors for Efficient
Scientific Simulation (ACCESS)
Input
Intensity Phase
OpticalDigital
© 2017 OPTALYSYS LTD 13
“”ECMWF is hugely supportive of this
proposal. It is clearly an exciting novel
technology path to more scalability
A collaborative exploration with the European Centre for
Medium-Range Weather Forecasts (ECWMF)
Objective:
To explore optical methods for next-
generation exascale weather forecasting
Outputs:
• Design for a proof of concept
• Product development roadmap
Notes:
• Major European consortium led by ECMWF
• 3 year project starting Oct 2015 involving our CFD
expert and dedicated PhD student
• Successfully awarded €4m grant from European
Commission’s Horizon 2020 programme Other consortium members include
“GENESYS™”
Optical Genetic Alignment
© 2017 OPTALYSYS LTD 15
Objective:
To produce a sequence alignment device
(GENESYS™ v0.9) with Earlham Institute (formerly The
Genome Analysis Centre)
Outputs:
• A working system in a real-world environment
• 95% energy savings and 20x faster processing
than a single 32 core HPC node.
• Frees up EI’s HPC processor cores
for other work
Notes:
• 2.5 year collaboration ending Dec 2017
• Funded by Innovate UK
“”Optalysys are the only company with
the proven technology and expertise
to realise our vision and we are excited
to be in partnership with them
Dr Tim Stitt, Head of Scientific Computing
GENESYS™ is developed in collaboration
with the Earlham Institute
© 2017 OPTALYSYS LTD 16
Genomic Sequence Alignment and Analysis
Collect, Load and
Sequence
DNA/RNA Samples
Align DNA/RNA Reads against
Reference library
SAMFILE
FASTQ|FASTAFILES
Variant / SNP
Calling
Metagenomics
RNA
Quantification
Sequencing Alignment Post - Processing
First priority
Second priority
Optalysys target applications
© 2017 OPTALYSYS LTD 17
Optical sequence process
The Fasta / Fastq data generated by the sequencing machine is converted
into binary symbols before being addressed to the optical system:
© 2017 OPTALYSYS LTD 18
The correlation process
Once input and reference data is converted into symbol representations, the
data is addressed into the optical system and matches are identified on a
camera sensor:
© 2017 OPTALYSYS LTD 19
Tests indicate the GENESYS™system is as accurate as the
leading software alignment methods
» As accurate as the best
software processes
» x 20 speed increase in
process time
» > 95% energy reduction
© 2017 OPTALYSYS LTD 20
Current Prototype
© 2017 OPTALYSYS LTD 21
Optical Sequence Alignment
© 2017 OPTALYSYS LTD 22
» $3.95million from undisclosed angel investors
» Funds for the manufacture of the first commercially
available Optical Co-processor for HPC and Desktop
Architectures
» Latest Prototype demonstrated in late August 2017 at
the Genome 10K Conference in Norwich, UK
» Series A Funding Round to begin in Fall 2017
Closure of $4million Seed Funding Round
Thank you
info@optalysys.com
www.optalysys.com
@Optalysys

More Related Content

PDF
OPEN RAN by Pourya Alinezhad
PDF
iptables 101- bottom-up
PDF
Advice for Healthcare IT Startups NYC Executive Informational and Networking ...
PPT
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
PDF
Future-Proofing Asset Failures with Cognitive Predictive Maintenance
PDF
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
PPTX
Seamless Dataflow with a Clinical Metadata Repository
PPTX
Insights into Real World Data Management Challenges
OPEN RAN by Pourya Alinezhad
iptables 101- bottom-up
Advice for Healthcare IT Startups NYC Executive Informational and Networking ...
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Future-Proofing Asset Failures with Cognitive Predictive Maintenance
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
Seamless Dataflow with a Clinical Metadata Repository
Insights into Real World Data Management Challenges

Similar to Optalysys Optical Processing for HPC (20)

PDF
The Possibility of Technology Due Diligence based on IP Panoramic View Analytics
PDF
Getting to Approval Faster Through Technology Innovation
PDF
Building Data Science Ecosystems for Smart Cities and Smart Commerce
PDF
Future of Data Strategy (ASEAN)
PDF
MISE2015
PDF
4 Smart_Connected_Factory_Slides Francois Briandet PDF.pdf
PDF
Postgres Vision 2018: Data as the New Oil
 
PDF
Big Data LDN 2017: The New Dominant Companies Are Running on Data
PDF
Big Data LDN 2017: The New Dominant Companies Are Running on Data
PPTX
The new dominant companies are running on data
PPTX
Research Data Shared Service Webinar #1
PPTX
Oracle big data publix sector 1
PDF
HPC DAY 2017 | The network part in accelerating Machine-Learning and Big-Data
PPTX
Insights into Real-world Data Management Challenges
PPTX
Mmea program - from sensors to services. Keynote from Dr. Tero Eklin
PDF
Domain Specific Languages for Parallel Graph AnalytiX (PGX)
PDF
Digital Transformation Journey
PPTX
Research Data Shared Service
PDF
Trends und Anwendungsbeispiele im Life Science Bereich
PDF
CTO Radshow Hamburg17 - Keynote - The CxO responsibilities in Big Data and AI...
The Possibility of Technology Due Diligence based on IP Panoramic View Analytics
Getting to Approval Faster Through Technology Innovation
Building Data Science Ecosystems for Smart Cities and Smart Commerce
Future of Data Strategy (ASEAN)
MISE2015
4 Smart_Connected_Factory_Slides Francois Briandet PDF.pdf
Postgres Vision 2018: Data as the New Oil
 
Big Data LDN 2017: The New Dominant Companies Are Running on Data
Big Data LDN 2017: The New Dominant Companies Are Running on Data
The new dominant companies are running on data
Research Data Shared Service Webinar #1
Oracle big data publix sector 1
HPC DAY 2017 | The network part in accelerating Machine-Learning and Big-Data
Insights into Real-world Data Management Challenges
Mmea program - from sensors to services. Keynote from Dr. Tero Eklin
Domain Specific Languages for Parallel Graph AnalytiX (PGX)
Digital Transformation Journey
Research Data Shared Service
Trends und Anwendungsbeispiele im Life Science Bereich
CTO Radshow Hamburg17 - Keynote - The CxO responsibilities in Big Data and AI...
Ad

More from inside-BigData.com (20)

PDF
Major Market Shifts in IT
PDF
Preparing to program Aurora at Exascale - Early experiences and future direct...
PPTX
Transforming Private 5G Networks
PDF
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
PDF
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
PDF
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
PDF
HPC Impact: EDA Telemetry Neural Networks
PDF
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
PDF
Machine Learning for Weather Forecasts
PPTX
HPC AI Advisory Council Update
PDF
Fugaku Supercomputer joins fight against COVID-19
PDF
Energy Efficient Computing using Dynamic Tuning
PDF
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
PDF
State of ARM-based HPC
PDF
Versal Premium ACAP for Network and Cloud Acceleration
PDF
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently
PDF
Scaling TCO in a Post Moore's Era
PDF
CUDA-Python and RAPIDS for blazing fast scientific computing
PDF
Introducing HPC with a Raspberry Pi Cluster
PDF
Overview of HPC Interconnects
Major Market Shifts in IT
Preparing to program Aurora at Exascale - Early experiences and future direct...
Transforming Private 5G Networks
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
HPC Impact: EDA Telemetry Neural Networks
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
Machine Learning for Weather Forecasts
HPC AI Advisory Council Update
Fugaku Supercomputer joins fight against COVID-19
Energy Efficient Computing using Dynamic Tuning
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
State of ARM-based HPC
Versal Premium ACAP for Network and Cloud Acceleration
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently
Scaling TCO in a Post Moore's Era
CUDA-Python and RAPIDS for blazing fast scientific computing
Introducing HPC with a Raspberry Pi Cluster
Overview of HPC Interconnects
Ad

Recently uploaded (20)

PDF
Assigned Numbers - 2025 - Bluetooth® Document
PDF
Encapsulation theory and applications.pdf
PDF
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
PPTX
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
PDF
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
PPTX
Big Data Technologies - Introduction.pptx
PDF
Electronic commerce courselecture one. Pdf
PDF
Empathic Computing: Creating Shared Understanding
PPTX
20250228 LYD VKU AI Blended-Learning.pptx
PDF
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
PDF
Network Security Unit 5.pdf for BCA BBA.
PPT
Teaching material agriculture food technology
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
PDF
Diabetes mellitus diagnosis method based random forest with bat algorithm
PDF
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
PDF
Dropbox Q2 2025 Financial Results & Investor Presentation
PDF
Machine learning based COVID-19 study performance prediction
PPTX
Spectroscopy.pptx food analysis technology
PPTX
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
PDF
Building Integrated photovoltaic BIPV_UPV.pdf
Assigned Numbers - 2025 - Bluetooth® Document
Encapsulation theory and applications.pdf
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Big Data Technologies - Introduction.pptx
Electronic commerce courselecture one. Pdf
Empathic Computing: Creating Shared Understanding
20250228 LYD VKU AI Blended-Learning.pptx
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Network Security Unit 5.pdf for BCA BBA.
Teaching material agriculture food technology
Reach Out and Touch Someone: Haptics and Empathic Computing
Diabetes mellitus diagnosis method based random forest with bat algorithm
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
Dropbox Q2 2025 Financial Results & Investor Presentation
Machine learning based COVID-19 study performance prediction
Spectroscopy.pptx food analysis technology
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
Building Integrated photovoltaic BIPV_UPV.pdf

Optalysys Optical Processing for HPC

  • 1. InsideHPC Dr Nick New, CEO and Founder
  • 2. Our vision is to revolutionise big data processing and model generation through optical processing technology
  • 3. © 2017 OPTALYSYS LTD 3 Trends towards specialist processors in the global computing community Demand for faster processing outstrips HPC’s capability to deliver “Science is evolving faster than IT can refresh infrastructure, design patterns and operational practices” Chris Dagdigian from Bioteam’s “Trend from the Trenches” talk at BioIT World April 2016 “Moore’s Law is dead says Gordon Moore” Techworld, 2010 “Specialisation proliferates again. Custom architectures will reemerge in the market” Top 5 trends for next 10 years in HPC, Intersect360 Research Feb 2017 “Accelerated computing has reached the tipping point in HPC” Intersect360 Research, 2015 “There’s a concerted move to throughput processors in the datacenter” Michael Feldman, Top500, Mar 2017 Move towards specialist systems designed to perform specific tasks
  • 5. 5 Optical Processing and Liquid Crystal SLMs Input Optics f f S(u,v) l Output s(x,y) • Liquid Crystal Spatial Light Modulators (SLMs) used to enter numerical data • Parallel, speed of light processing • Highy scalable - Process time not dependent on resolution • Low power laser light • Order 1 operation
  • 6. 6 Correlation Pattern Recognition and Equation SolvingLASER CAMERA Input data Reference data Output data Laser Input SLM FT lens Reference SLM FT lens FT IFTMM 𝑔 𝑥 ∗ 𝑟 𝑥 = 𝐼𝐹𝑇[𝐺 𝑢 𝑅 𝑢 ∗ ]
  • 7. © 2017 OPTALYSYS LTD 7 Presenting Optalysys’s highly scalable, high-speed, low power optical co-processing technology
  • 8. © 2017 OPTALYSYS LTD 8 • Multiple SLMs and cameras • Optical Correlation using diffractive optical process • Laser light projected through the system as the information carrier • Optical system mounted to a driver PCIe card • Open API Interface • Highly scalable system SLMs PCIe interface Laser Cameras Glass How it works
  • 10. © 2017 OPTALYSYS LTD 10 Key sectors demanding high performance computing Our optical approach will herald a step-change, benefitting multiple applications within each of these fields Machine Learning Aerodynamics Signal Processing Weather Forecasting Financial Analysis Bioinformatics and Medical Analysis Defence and intelligence Oil and Energy
  • 11. 11 Solutions for targeted key markets with strategic partners • Genome assembly / alignment • InnovateUK • Next generation weather forecasting • EU Horizon2020 • Equation solving fundamentals • DARPA ACCESS program Focus Areas
  • 12. © 2017 OPTALYSYS LTD 12 DARPA Seedling Project • Completed in May 2017 • Successfully demonstrated the principles of an optical Fourier transform co-processor for equation solving • 13 month “Seedling Project” • Part of DARPA-SN-15-31: Call for New Innovations towards: Analog and Continuous-variable Co-processors for Efficient Scientific Simulation (ACCESS) Input Intensity Phase OpticalDigital
  • 13. © 2017 OPTALYSYS LTD 13 “”ECMWF is hugely supportive of this proposal. It is clearly an exciting novel technology path to more scalability A collaborative exploration with the European Centre for Medium-Range Weather Forecasts (ECWMF) Objective: To explore optical methods for next- generation exascale weather forecasting Outputs: • Design for a proof of concept • Product development roadmap Notes: • Major European consortium led by ECMWF • 3 year project starting Oct 2015 involving our CFD expert and dedicated PhD student • Successfully awarded €4m grant from European Commission’s Horizon 2020 programme Other consortium members include
  • 15. © 2017 OPTALYSYS LTD 15 Objective: To produce a sequence alignment device (GENESYS™ v0.9) with Earlham Institute (formerly The Genome Analysis Centre) Outputs: • A working system in a real-world environment • 95% energy savings and 20x faster processing than a single 32 core HPC node. • Frees up EI’s HPC processor cores for other work Notes: • 2.5 year collaboration ending Dec 2017 • Funded by Innovate UK “”Optalysys are the only company with the proven technology and expertise to realise our vision and we are excited to be in partnership with them Dr Tim Stitt, Head of Scientific Computing GENESYS™ is developed in collaboration with the Earlham Institute
  • 16. © 2017 OPTALYSYS LTD 16 Genomic Sequence Alignment and Analysis Collect, Load and Sequence DNA/RNA Samples Align DNA/RNA Reads against Reference library SAMFILE FASTQ|FASTAFILES Variant / SNP Calling Metagenomics RNA Quantification Sequencing Alignment Post - Processing First priority Second priority Optalysys target applications
  • 17. © 2017 OPTALYSYS LTD 17 Optical sequence process The Fasta / Fastq data generated by the sequencing machine is converted into binary symbols before being addressed to the optical system:
  • 18. © 2017 OPTALYSYS LTD 18 The correlation process Once input and reference data is converted into symbol representations, the data is addressed into the optical system and matches are identified on a camera sensor:
  • 19. © 2017 OPTALYSYS LTD 19 Tests indicate the GENESYS™system is as accurate as the leading software alignment methods » As accurate as the best software processes » x 20 speed increase in process time » > 95% energy reduction
  • 20. © 2017 OPTALYSYS LTD 20 Current Prototype
  • 21. © 2017 OPTALYSYS LTD 21 Optical Sequence Alignment
  • 22. © 2017 OPTALYSYS LTD 22 » $3.95million from undisclosed angel investors » Funds for the manufacture of the first commercially available Optical Co-processor for HPC and Desktop Architectures » Latest Prototype demonstrated in late August 2017 at the Genome 10K Conference in Norwich, UK » Series A Funding Round to begin in Fall 2017 Closure of $4million Seed Funding Round

Editor's Notes

  • #4: Our vision has not changed….
  • #14: Who - an independent intergovernmental organisation supported by 34 states. Why are they important…the largest supercomputer centre for weather forecasting in the world (#28 and #29 in the top500 fastest supercomputers in the world). The Centre's computer system contains the world's largest archive of numerical weather prediction data.