SlideShare a Scribd company logo
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Michael Raposa
Head of Platform Engineering, AQR Capital
John Kain
Head of Business and Market Development, Capital Markets, AWS
FSI303
How AQR Capital Leverages AWS to
Research New Investment Signals
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Grid computing trends in Financial Services
Widespread
deployment
Leading
edge
Traditional
(Data Synapse/
Platform Symphony)
Spark/Container
Orchestration &
Schedulers
Serverless
Current platform of
choice
AWS
Batch
AWS
Lambda
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Introducing AWS Batch
AWS Batch
Managed
No software to install or
servers to manage
Integrated with AWS
Natively integrated with the
AWS platform
Cost-optimized Automatically
provisions compute resources
How AQR Capital Leverages AWS to
Research New Investment Signals
Michael Raposa
July 17, 2018
Head of Platform Engineering, AQR
Not intended for the sale or marketing of financial products or services.
Agenda
• About AQR
• Business Problem
• Solution Summary
• Lessons Learned
• Takeaways
5
Our Firm
AQR is a global investment management firm built at the intersection of financial theory and practical application. We strive to deliver superior,
long-term results for our clients by looking past market noise to identify and isolate what matters most, and by developing ideas that stand up to
rigorous testing. Our focus on practical insights and analysis has made us leaders in alternative and traditional strategies since 1998.
At a Glance
• AQR takes a systematic, research-driven approach to managing alternative and traditional strategies
• We apply quantitative tools to process fundamental information and manage risk
• Our clients include institutional investors, such as pension funds, defined contribution plans, insurance companies, endowments, foundations, family offices and
sovereign wealth funds, as well as RIAs, private banks and financial advisors
• The firm has 36 principals and 914 employees; over half of employees hold advanced degrees
• AQR is based in Greenwich, Connecticut, with offices in Boston, Chicago, Hong Kong, London, Los Angeles, and Sydney
• Approximately $225 billion in assets under management as of March 31, 2018*
*Approximate as of 3/31/2018, includes assets managed by AQR and its advisory affiliates. 6
Problem Statement
Background
• Quantitative Asset Manager
• Investment Decisions based on numerical models and systematic trading
• Researchers develops models and “back test” ideas over many years
• Never-ending appetite for more data and non-obvious data sets
7
Source: AQR. For illustrative purposes only.
Researcher Workflow
Idea
Gather
Data
Build
Model
Backtest
Analyze
Result
8
Source: AQR. For illustrative purposes only.
Problem Statement
Background
• Quantitative Asset Manager
• Investment Decisions based on numerical models and systematic trading
• Researchers develops models and “back test” ideas over many years
• Never-ending appetite for more data and non-obvious data sets
Problem
• On premise compute grid can’t keep up with demand
• CAPEX locked into grid
• Researchers wait for grid resources
• Researchers need job results as quickly as possible
• New experimental use cases, e.g. GPU, require significant time and money
upfront investment
9
Source: AQR. For illustrative purposes only.
Design Considerations
1. Scalable both in compute and memory
2. Fast without long queue times
3. Don’t want to manage a Job Scheduler, e.g. Condor or Sun Grid Engine
4. Easy to use
5. Secure
10
Solution Summary
Burstability
Leverage building block application
services, e.g. S3, EC2, DynamoDB
Reduce costs – Spot
Leverage Cloud
11
Solution Summary
Based on Sun Grid Engine
Submit jobs via CLI or API –
Unlimited compute at fingertips
Short feedback loop – immediate
results to researcher
Backend grid matches researcher
workstation – No “It works in DEV
but not in the GRID”
Seamless Interface
12
Solution Summary
Get out of researchers way –
Automate everything
“Unlimited” compute
Short start times – No “queues”
Fast
13
Solution Summary
Encryption Everywhere
Leverage AWS Security Tools
Security Cloud Engineering
problem – Not researcher’s
Secure
14
Solution Summary
Step 1
15
Solution Summary
Step 2
16
Solution Summary
Step 3
17
Solution Summary
Step 4
18
Solution Summary – Job Submission (CLI)
19
Solution Summary – Job Submission (CLI)
20
Solution Summary – Job Submission (CLI)
21
Solution Summary – Job Submission (CLI)
22
Solution Summary – Job Submission (CLI)
24
Solution Summary – Job Submission (CLI)
25
Solution Summary – Job Submission (API)
Step 1
29
Solution Summary – Job Submission (API)
Step 2
30
Solution Summary – Job Submission (API)
Step 3
31
Solution Summary – Job Submission (API)
Step 4
32
Solution Summary – DAGs
33
Client
Job
Child jobs called by other jobs
Lessons Learned - General
Use Spot
Use as many instance types as you
can
Use as many AZs as you can
Drive the Lowest Cost - $15/1000 vCPU/Hour
34
Spot Instance
Availability Zones
Instance Families
Lessons Learned - General
ECS Logs
Job Output
Host Logs
…etc.
Log Everything!
35
Lessons Learned - General
Job Run Time
Job Start-up Times
Job Cost by User
vCPU Consumption by User
High Priority Queue Consumption
by User
Monitor Everything!
36
CloudWatch Dashboard
37
CloudWatch Dashboard
38
Lessons Learned
As we added more Users…
1. TooManyRequestsException
1. DescribeJobs API Call Failing
2. Providing Governance “Guard Rails”
39
Lessons Learned – Event Based Pipeline
40
Lessons Learned – Event Based Pipeline
41
Lessons Learned – Event Based Pipeline
42
Lessons Learned – Event Based Pipeline
43
Lessons Learned
As we added more compute…
1. TooManyRequestsException … API calls in the container
2. Job State storage woes
3. Job Start times
4. Job Costs
44
Lessons Learned – API in Containers
45
Parameter
Store
ECS
container
Get Secret
Lessons Learned – API in Containers
At Scale
46
Exception:
Too many requestsParameter
Store
ECS
container
ECS
container
ECS
container
ECS
container
ECS
container
Lessons Learned – API in Containers
At Scale
47
ECS
container
ECS
container
ECS
container
ECS
container
ECS
container
Amazon
S3
What is Shared Job State
Shared Memory Across Containers
Job Assignment & Orchestration
Job Input and Output
48
Lessons Learned – Job State Backend
49
Amazon
EFS
Lessons Learned – Job State Backend
50
Amazon
EFS
Redis
Lessons Learned – Job State Backend
51
Amazon
EFS
Amazon
S3
Redis
Lessons Learned – Job Start Times
52
Lessons Learned – Set Limits
53
Takeaways
Spot
Multi-AZ
Log Everything
Monitor Everything
Follow Best Practices
54
Spot
Instance
Availability Zones
Takeaways
Eliminate API Calls in your
containers. Only use services that
scale, e.g. DynamoDB
Switch to Event/Message Based
Status from Poll Based
Choose a job state backend that fits
your use case
Watch for Scale Issues
55
Takeaways
Write a light-weight wrapper around
AWS Batch for non-technical users
Have a “Governator” function to
ensure compliance
Make it Easy and Secure
56
Takeaways
Packer and bake AMIs
Pre-warm the cluster during active
times of the day
Give yourself an SLA – 75% of jobs
start in 10 mins and 90% in 15 mins
Reduce Your Start Times
57
Takeaways
Alarm for large long running jobs ->
$$$
Set limits on your Compute
Environment
Control Runaway Costs
58
Q&A
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Thank you!

More Related Content

What's hot

Value Realization with SAP Ariba Solutions Approach, Measurement, and Success
Value Realization with SAP Ariba Solutions Approach, Measurement, and SuccessValue Realization with SAP Ariba Solutions Approach, Measurement, and Success
Value Realization with SAP Ariba Solutions Approach, Measurement, and Success
SAP Ariba
 
S/4 HANA presentation at INDUS
S/4 HANA presentation at INDUSS/4 HANA presentation at INDUS
S/4 HANA presentation at INDUS
INDUSCommunity
 
SAP HANA SPS10- SAP HANA Modeling
SAP HANA SPS10- SAP HANA ModelingSAP HANA SPS10- SAP HANA Modeling
SAP HANA SPS10- SAP HANA Modeling
SAP Technology
 
Apache Kafka for Real-time Supply Chain in the Food and Retail Industry
Apache Kafka for Real-time Supply Chainin the Food and Retail IndustryApache Kafka for Real-time Supply Chainin the Food and Retail Industry
Apache Kafka for Real-time Supply Chain in the Food and Retail Industry
Kai Wähner
 
SAP S/4HANA - transition brochure
SAP S/4HANA - transition brochureSAP S/4HANA - transition brochure
SAP S/4HANA - transition brochure
Andreas Born
 
Web 3.0 The Semantic Web
Web 3.0 The Semantic WebWeb 3.0 The Semantic Web
Web 3.0 The Semantic Web
Hatem Mahmoud
 
Achieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
Achieve Digital Transformation with SAP Hybris Commerce Travel AcceleratorAchieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
Achieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
SAP Customer Experience
 
Managing the Complexity of Microservices Deployments
Managing the Complexity of Microservices DeploymentsManaging the Complexity of Microservices Deployments
Managing the Complexity of Microservices Deployments
Apigee | Google Cloud
 
Sap Analytics Cloud
Sap Analytics CloudSap Analytics Cloud
Sap Analytics Cloud
Srinath Reddy
 
Sap fico demo presentation
Sap fico demo presentationSap fico demo presentation
Sap fico demo presentation
Intelligroup, Inc.
 
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
WSO2
 
SAP HANA Migration Deck.pptx
SAP HANA Migration Deck.pptxSAP HANA Migration Deck.pptx
SAP HANA Migration Deck.pptx
SingbBablu
 
Simplify your Retail business processes with SAP that helps drive business tr...
Simplify your Retail business processes with SAP that helps drive business tr...Simplify your Retail business processes with SAP that helps drive business tr...
Simplify your Retail business processes with SAP that helps drive business tr...
Vct Sales
 
Sap fashion-management
Sap fashion-managementSap fashion-management
Sap fashion-management
suryanarayana tata
 
B1AIP30 - Kickoff Meeting Template.pptx
B1AIP30 - Kickoff Meeting Template.pptxB1AIP30 - Kickoff Meeting Template.pptx
B1AIP30 - Kickoff Meeting Template.pptx
ssuser97c2dc
 
SUP_EDE_0140_1712_DAY1_Q123.pdf
SUP_EDE_0140_1712_DAY1_Q123.pdfSUP_EDE_0140_1712_DAY1_Q123.pdf
SUP_EDE_0140_1712_DAY1_Q123.pdf
Franklin Sojo
 
Sap fico blueprint 2017 final Anilkumar chowdary
Sap fico blueprint 2017 final Anilkumar chowdarySap fico blueprint 2017 final Anilkumar chowdary
Sap fico blueprint 2017 final Anilkumar chowdary
ANILKUMARPULIPATI1
 
SAP Cloud Platform API Management Technical Brief
SAP Cloud Platform API Management Technical BriefSAP Cloud Platform API Management Technical Brief
SAP Cloud Platform API Management Technical Brief
SAP Cloud Platform
 
“Migration to Suite of HANA”
“Migration to Suite of HANA”“Migration to Suite of HANA”
“Migration to Suite of HANA”
Wise Men
 
Guided Buying: Exploration and Best Practices
Guided Buying: Exploration and Best PracticesGuided Buying: Exploration and Best Practices
Guided Buying: Exploration and Best Practices
SAP Ariba
 

What's hot (20)

Value Realization with SAP Ariba Solutions Approach, Measurement, and Success
Value Realization with SAP Ariba Solutions Approach, Measurement, and SuccessValue Realization with SAP Ariba Solutions Approach, Measurement, and Success
Value Realization with SAP Ariba Solutions Approach, Measurement, and Success
 
S/4 HANA presentation at INDUS
S/4 HANA presentation at INDUSS/4 HANA presentation at INDUS
S/4 HANA presentation at INDUS
 
SAP HANA SPS10- SAP HANA Modeling
SAP HANA SPS10- SAP HANA ModelingSAP HANA SPS10- SAP HANA Modeling
SAP HANA SPS10- SAP HANA Modeling
 
Apache Kafka for Real-time Supply Chain in the Food and Retail Industry
Apache Kafka for Real-time Supply Chainin the Food and Retail IndustryApache Kafka for Real-time Supply Chainin the Food and Retail Industry
Apache Kafka for Real-time Supply Chain in the Food and Retail Industry
 
SAP S/4HANA - transition brochure
SAP S/4HANA - transition brochureSAP S/4HANA - transition brochure
SAP S/4HANA - transition brochure
 
Web 3.0 The Semantic Web
Web 3.0 The Semantic WebWeb 3.0 The Semantic Web
Web 3.0 The Semantic Web
 
Achieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
Achieve Digital Transformation with SAP Hybris Commerce Travel AcceleratorAchieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
Achieve Digital Transformation with SAP Hybris Commerce Travel Accelerator
 
Managing the Complexity of Microservices Deployments
Managing the Complexity of Microservices DeploymentsManaging the Complexity of Microservices Deployments
Managing the Complexity of Microservices Deployments
 
Sap Analytics Cloud
Sap Analytics CloudSap Analytics Cloud
Sap Analytics Cloud
 
Sap fico demo presentation
Sap fico demo presentationSap fico demo presentation
Sap fico demo presentation
 
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
[WSO2 Summit EMEA 2020] Building an Interactive API Marketplace
 
SAP HANA Migration Deck.pptx
SAP HANA Migration Deck.pptxSAP HANA Migration Deck.pptx
SAP HANA Migration Deck.pptx
 
Simplify your Retail business processes with SAP that helps drive business tr...
Simplify your Retail business processes with SAP that helps drive business tr...Simplify your Retail business processes with SAP that helps drive business tr...
Simplify your Retail business processes with SAP that helps drive business tr...
 
Sap fashion-management
Sap fashion-managementSap fashion-management
Sap fashion-management
 
B1AIP30 - Kickoff Meeting Template.pptx
B1AIP30 - Kickoff Meeting Template.pptxB1AIP30 - Kickoff Meeting Template.pptx
B1AIP30 - Kickoff Meeting Template.pptx
 
SUP_EDE_0140_1712_DAY1_Q123.pdf
SUP_EDE_0140_1712_DAY1_Q123.pdfSUP_EDE_0140_1712_DAY1_Q123.pdf
SUP_EDE_0140_1712_DAY1_Q123.pdf
 
Sap fico blueprint 2017 final Anilkumar chowdary
Sap fico blueprint 2017 final Anilkumar chowdarySap fico blueprint 2017 final Anilkumar chowdary
Sap fico blueprint 2017 final Anilkumar chowdary
 
SAP Cloud Platform API Management Technical Brief
SAP Cloud Platform API Management Technical BriefSAP Cloud Platform API Management Technical Brief
SAP Cloud Platform API Management Technical Brief
 
“Migration to Suite of HANA”
“Migration to Suite of HANA”“Migration to Suite of HANA”
“Migration to Suite of HANA”
 
Guided Buying: Exploration and Best Practices
Guided Buying: Exploration and Best PracticesGuided Buying: Exploration and Best Practices
Guided Buying: Exploration and Best Practices
 

Similar to How AQR Capital Uses AWS to Research New Investment Signals

Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
Amazon Web Services
 
Migrating & Operating Microsoft Applications in AWS
Migrating & Operating Microsoft Applications in AWSMigrating & Operating Microsoft Applications in AWS
Migrating & Operating Microsoft Applications in AWS
Amazon Web Services
 
Amazon Web Services User Group Sydney - March 2018
Amazon Web Services User Group Sydney - March 2018Amazon Web Services User Group Sydney - March 2018
Amazon Web Services User Group Sydney - March 2018
PolarSeven Pty Ltd
 
Mythbusting the Federal Cloud Journey
Mythbusting the Federal Cloud JourneyMythbusting the Federal Cloud Journey
Mythbusting the Federal Cloud Journey
Amazon Web Services
 
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
Amazon Web Services
 
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
Ed Sattar
 
Scaling Security Operations and Automating Governance: Which AWS Services Sho...
Scaling Security Operations and Automating Governance: Which AWS Services Sho...Scaling Security Operations and Automating Governance: Which AWS Services Sho...
Scaling Security Operations and Automating Governance: Which AWS Services Sho...
Amazon Web Services
 
AWS Well-Architected Workshop
AWS Well-Architected WorkshopAWS Well-Architected Workshop
AWS Well-Architected Workshop
Amazon Web Services
 
MLOps and Reproducible ML on AWS with Kubeflow and SageMaker
MLOps and Reproducible ML on AWS with Kubeflow and SageMakerMLOps and Reproducible ML on AWS with Kubeflow and SageMaker
MLOps and Reproducible ML on AWS with Kubeflow and SageMaker
Provectus
 
Well-Architected Bootcamp
Well-Architected BootcampWell-Architected Bootcamp
Well-Architected Bootcamp
Amazon Web Services
 
Multi cloud governance best practices - AWS, Azure, GCP
Multi cloud governance best practices - AWS, Azure, GCPMulti cloud governance best practices - AWS, Azure, GCP
Multi cloud governance best practices - AWS, Azure, GCP
Faiza Mehar
 
Develop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
Develop an Enterprise-wide Cloud Adoption Strategy – Chris MerriganDevelop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
Develop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
Amazon Web Services
 
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
Amazon Web Services
 
Pivoting to Cloud: How an MSP Brokers Cloud Services
Pivoting to Cloud: How an MSP Brokers Cloud Services Pivoting to Cloud: How an MSP Brokers Cloud Services
Pivoting to Cloud: How an MSP Brokers Cloud Services
RightScale
 
Outpost24 webinar : how to secure your data in the cloud - 06-2018
Outpost24 webinar : how to secure your data in the cloud - 06-2018Outpost24 webinar : how to secure your data in the cloud - 06-2018
Outpost24 webinar : how to secure your data in the cloud - 06-2018
Outpost24
 
Benefits of Cloud Computing
Benefits of Cloud ComputingBenefits of Cloud Computing
Benefits of Cloud Computing
Amazon Web Services
 
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
MongoDB
 
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
MongoDB
 
reInvent reCap 2022
reInvent reCap 2022reInvent reCap 2022
reInvent reCap 2022
CloudHesive
 
AWS User Group Sydney - Meetup #60
AWS User Group Sydney - Meetup #60AWS User Group Sydney - Meetup #60
AWS User Group Sydney - Meetup #60
PolarSeven Pty Ltd
 

Similar to How AQR Capital Uses AWS to Research New Investment Signals (20)

Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
Intro to AWS Batch & How AQR Capital leverages AWS to Identify New Investment...
 
Migrating & Operating Microsoft Applications in AWS
Migrating & Operating Microsoft Applications in AWSMigrating & Operating Microsoft Applications in AWS
Migrating & Operating Microsoft Applications in AWS
 
Amazon Web Services User Group Sydney - March 2018
Amazon Web Services User Group Sydney - March 2018Amazon Web Services User Group Sydney - March 2018
Amazon Web Services User Group Sydney - March 2018
 
Mythbusting the Federal Cloud Journey
Mythbusting the Federal Cloud JourneyMythbusting the Federal Cloud Journey
Mythbusting the Federal Cloud Journey
 
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
Distributed Solar Systems at EDF Renewables and AWS IoT: A Natural Fit (PUT30...
 
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
Cloud Based Cognitive Learning & IT Project Performance Platform (CLIPP Platf...
 
Scaling Security Operations and Automating Governance: Which AWS Services Sho...
Scaling Security Operations and Automating Governance: Which AWS Services Sho...Scaling Security Operations and Automating Governance: Which AWS Services Sho...
Scaling Security Operations and Automating Governance: Which AWS Services Sho...
 
AWS Well-Architected Workshop
AWS Well-Architected WorkshopAWS Well-Architected Workshop
AWS Well-Architected Workshop
 
MLOps and Reproducible ML on AWS with Kubeflow and SageMaker
MLOps and Reproducible ML on AWS with Kubeflow and SageMakerMLOps and Reproducible ML on AWS with Kubeflow and SageMaker
MLOps and Reproducible ML on AWS with Kubeflow and SageMaker
 
Well-Architected Bootcamp
Well-Architected BootcampWell-Architected Bootcamp
Well-Architected Bootcamp
 
Multi cloud governance best practices - AWS, Azure, GCP
Multi cloud governance best practices - AWS, Azure, GCPMulti cloud governance best practices - AWS, Azure, GCP
Multi cloud governance best practices - AWS, Azure, GCP
 
Develop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
Develop an Enterprise-wide Cloud Adoption Strategy – Chris MerriganDevelop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
Develop an Enterprise-wide Cloud Adoption Strategy – Chris Merrigan
 
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
AWS re:Invent 2016: Fueling Migration: Shortcutting your Application Portfoli...
 
Pivoting to Cloud: How an MSP Brokers Cloud Services
Pivoting to Cloud: How an MSP Brokers Cloud Services Pivoting to Cloud: How an MSP Brokers Cloud Services
Pivoting to Cloud: How an MSP Brokers Cloud Services
 
Outpost24 webinar : how to secure your data in the cloud - 06-2018
Outpost24 webinar : how to secure your data in the cloud - 06-2018Outpost24 webinar : how to secure your data in the cloud - 06-2018
Outpost24 webinar : how to secure your data in the cloud - 06-2018
 
Benefits of Cloud Computing
Benefits of Cloud ComputingBenefits of Cloud Computing
Benefits of Cloud Computing
 
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
MongoDB World 2019: A MongoDB Journey: Moving From a Relational Database to M...
 
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
MongoDB .local Chicago 2019: A MongoDB Journey: Moving from a relational data...
 
reInvent reCap 2022
reInvent reCap 2022reInvent reCap 2022
reInvent reCap 2022
 
AWS User Group Sydney - Meetup #60
AWS User Group Sydney - Meetup #60AWS User Group Sydney - Meetup #60
AWS User Group Sydney - Meetup #60
 

More from Amazon Web Services

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Amazon Web Services
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Amazon Web Services
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
Amazon Web Services
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
Amazon Web Services
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
Amazon Web Services
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
Amazon Web Services
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Amazon Web Services
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
Amazon Web Services
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Amazon Web Services
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
Amazon Web Services
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
Amazon Web Services
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Amazon Web Services
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
Amazon Web Services
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Amazon Web Services
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWSAmazon Web Services
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckAmazon Web Services
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without serversAmazon Web Services
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...Amazon Web Services
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
Amazon Web Services
 

More from Amazon Web Services (20)

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
 
Fundraising Essentials
Fundraising EssentialsFundraising Essentials
Fundraising Essentials
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
 

How AQR Capital Uses AWS to Research New Investment Signals

  • 1. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Michael Raposa Head of Platform Engineering, AQR Capital John Kain Head of Business and Market Development, Capital Markets, AWS FSI303 How AQR Capital Leverages AWS to Research New Investment Signals
  • 2. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Grid computing trends in Financial Services Widespread deployment Leading edge Traditional (Data Synapse/ Platform Symphony) Spark/Container Orchestration & Schedulers Serverless Current platform of choice AWS Batch AWS Lambda
  • 3. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Introducing AWS Batch AWS Batch Managed No software to install or servers to manage Integrated with AWS Natively integrated with the AWS platform Cost-optimized Automatically provisions compute resources
  • 4. How AQR Capital Leverages AWS to Research New Investment Signals Michael Raposa July 17, 2018 Head of Platform Engineering, AQR Not intended for the sale or marketing of financial products or services.
  • 5. Agenda • About AQR • Business Problem • Solution Summary • Lessons Learned • Takeaways 5
  • 6. Our Firm AQR is a global investment management firm built at the intersection of financial theory and practical application. We strive to deliver superior, long-term results for our clients by looking past market noise to identify and isolate what matters most, and by developing ideas that stand up to rigorous testing. Our focus on practical insights and analysis has made us leaders in alternative and traditional strategies since 1998. At a Glance • AQR takes a systematic, research-driven approach to managing alternative and traditional strategies • We apply quantitative tools to process fundamental information and manage risk • Our clients include institutional investors, such as pension funds, defined contribution plans, insurance companies, endowments, foundations, family offices and sovereign wealth funds, as well as RIAs, private banks and financial advisors • The firm has 36 principals and 914 employees; over half of employees hold advanced degrees • AQR is based in Greenwich, Connecticut, with offices in Boston, Chicago, Hong Kong, London, Los Angeles, and Sydney • Approximately $225 billion in assets under management as of March 31, 2018* *Approximate as of 3/31/2018, includes assets managed by AQR and its advisory affiliates. 6
  • 7. Problem Statement Background • Quantitative Asset Manager • Investment Decisions based on numerical models and systematic trading • Researchers develops models and “back test” ideas over many years • Never-ending appetite for more data and non-obvious data sets 7 Source: AQR. For illustrative purposes only.
  • 9. Problem Statement Background • Quantitative Asset Manager • Investment Decisions based on numerical models and systematic trading • Researchers develops models and “back test” ideas over many years • Never-ending appetite for more data and non-obvious data sets Problem • On premise compute grid can’t keep up with demand • CAPEX locked into grid • Researchers wait for grid resources • Researchers need job results as quickly as possible • New experimental use cases, e.g. GPU, require significant time and money upfront investment 9 Source: AQR. For illustrative purposes only.
  • 10. Design Considerations 1. Scalable both in compute and memory 2. Fast without long queue times 3. Don’t want to manage a Job Scheduler, e.g. Condor or Sun Grid Engine 4. Easy to use 5. Secure 10
  • 11. Solution Summary Burstability Leverage building block application services, e.g. S3, EC2, DynamoDB Reduce costs – Spot Leverage Cloud 11
  • 12. Solution Summary Based on Sun Grid Engine Submit jobs via CLI or API – Unlimited compute at fingertips Short feedback loop – immediate results to researcher Backend grid matches researcher workstation – No “It works in DEV but not in the GRID” Seamless Interface 12
  • 13. Solution Summary Get out of researchers way – Automate everything “Unlimited” compute Short start times – No “queues” Fast 13
  • 14. Solution Summary Encryption Everywhere Leverage AWS Security Tools Security Cloud Engineering problem – Not researcher’s Secure 14
  • 19. Solution Summary – Job Submission (CLI) 19
  • 20. Solution Summary – Job Submission (CLI) 20
  • 21. Solution Summary – Job Submission (CLI) 21
  • 22. Solution Summary – Job Submission (CLI) 22
  • 23. Solution Summary – Job Submission (CLI) 24
  • 24. Solution Summary – Job Submission (CLI) 25
  • 25. Solution Summary – Job Submission (API) Step 1 29
  • 26. Solution Summary – Job Submission (API) Step 2 30
  • 27. Solution Summary – Job Submission (API) Step 3 31
  • 28. Solution Summary – Job Submission (API) Step 4 32
  • 29. Solution Summary – DAGs 33 Client Job Child jobs called by other jobs
  • 30. Lessons Learned - General Use Spot Use as many instance types as you can Use as many AZs as you can Drive the Lowest Cost - $15/1000 vCPU/Hour 34 Spot Instance Availability Zones Instance Families
  • 31. Lessons Learned - General ECS Logs Job Output Host Logs …etc. Log Everything! 35
  • 32. Lessons Learned - General Job Run Time Job Start-up Times Job Cost by User vCPU Consumption by User High Priority Queue Consumption by User Monitor Everything! 36
  • 35. Lessons Learned As we added more Users… 1. TooManyRequestsException 1. DescribeJobs API Call Failing 2. Providing Governance “Guard Rails” 39
  • 36. Lessons Learned – Event Based Pipeline 40
  • 37. Lessons Learned – Event Based Pipeline 41
  • 38. Lessons Learned – Event Based Pipeline 42
  • 39. Lessons Learned – Event Based Pipeline 43
  • 40. Lessons Learned As we added more compute… 1. TooManyRequestsException … API calls in the container 2. Job State storage woes 3. Job Start times 4. Job Costs 44
  • 41. Lessons Learned – API in Containers 45 Parameter Store ECS container Get Secret
  • 42. Lessons Learned – API in Containers At Scale 46 Exception: Too many requestsParameter Store ECS container ECS container ECS container ECS container ECS container
  • 43. Lessons Learned – API in Containers At Scale 47 ECS container ECS container ECS container ECS container ECS container Amazon S3
  • 44. What is Shared Job State Shared Memory Across Containers Job Assignment & Orchestration Job Input and Output 48
  • 45. Lessons Learned – Job State Backend 49 Amazon EFS
  • 46. Lessons Learned – Job State Backend 50 Amazon EFS Redis
  • 47. Lessons Learned – Job State Backend 51 Amazon EFS Amazon S3 Redis
  • 48. Lessons Learned – Job Start Times 52
  • 49. Lessons Learned – Set Limits 53
  • 50. Takeaways Spot Multi-AZ Log Everything Monitor Everything Follow Best Practices 54 Spot Instance Availability Zones
  • 51. Takeaways Eliminate API Calls in your containers. Only use services that scale, e.g. DynamoDB Switch to Event/Message Based Status from Poll Based Choose a job state backend that fits your use case Watch for Scale Issues 55
  • 52. Takeaways Write a light-weight wrapper around AWS Batch for non-technical users Have a “Governator” function to ensure compliance Make it Easy and Secure 56
  • 53. Takeaways Packer and bake AMIs Pre-warm the cluster during active times of the day Give yourself an SLA – 75% of jobs start in 10 mins and 90% in 15 mins Reduce Your Start Times 57
  • 54. Takeaways Alarm for large long running jobs -> $$$ Set limits on your Compute Environment Control Runaway Costs 58
  • 55. Q&A
  • 56.
  • 57. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Thank you!