SlideShare une entreprise Scribd logo
1  sur  61
DevOps for Data Science
by Stepan Pushkarev
CTO of Hydrosphere.io
DevOps is a catchy buzzword to optimise things
© Josh Wills
http://www.slideshare.net/g33ktalk/dataengconf-sf16-bridging-the-gap-between-data-science-and-data-engineerin
g
Is there life after marriage data science?
Dating, Flowers,
Dreams
Marriage
Happily lived
forever?
Collect & prepare
data
Build ML Model
This talk is for people who are married aware of
“other 99% of data science”
Dating, Flowers,
Dreams
Marriage
Happily lived
forever?
Collect &
prepare data
Build ML Model
This talk is NOT about
- Setting up Apache Spark/Hadoop cluster
- Configuring CI/CD tools like Jenkins
- Configuring monitoring tools & dashboards
- Agile/DevOps brainwashing & consulting story
Agenda
- Challenges in deploying analytics into
production
- Deploying analytics as a service
- Feedback loops: testing, monitoring,
analytics of analytics
Why do companies hire data scientists?
Why do companies hire data scientists?
To make products smarter.
What is a deliverable of data scientist and data
engineer?
What is a deliverable of data scientist?
Academic
paper?
ML Model? R/Python
script?
Jupiter
Notebook?
BI
Dashboard?
What has to be a deliverable of data scientist?
Data pipelines and machine
learning models that deployed as
pluggable, testable, supportable,
monitorable analytics services.
Option 1: Engineer to implement academic paper
Option 2: Engineer to re-implement R/Python script
Option 3: Run notebook as it is using cron
Option 3: Run notebook as it is using cron
Option 4: Build software to eat the world Data Science
Eating data science
© Daniel Tunkelang - Where should you put your data scientists? -
www.slideshare.net/dtunkelang/where-should-you-put-your-data-scientists
Step 1 (management): Integrate data scientists into
cross-functional teams
Eating data science
Step 2 (operations): Make environments scalable
and elastic. Finally.
Eating data science
Step 3: Make data scientists to write less code
Eating data science
Step 4: Deploy analytics as services
Step 5: Use feedback loops: testing, monitoring,
analytics for analytics
Build ML Model
Test
Monitor,
maintain,
analyze
Deploy as a service
Collect & prepare
data
Agenda
- Challenges in deploying analytics into
production
- Deploying analytics as a service
- Feedback loops: testing, monitoring,
analytics of analytics
Deploying analytics as a service
- Defines deliverable for Data Scientist / Data Engineer.
- Plugs analytics into end-to-end products through API.
- With the right tooling allows Data Scientist to deploy it in self
serve
Look around - proprietary ML based APIs
- Alchemy API
- Google Prediction API
- Cloud Vision API
- Azure ML
Can we do our own on top of Apache Spark?
Bad Practice #1. Business logic in Spark? WTF?
Bad Practice #2. Database as API
Execute reporting job
Mark Job as complete &
save result
Poll for new tasks
Poll for resultSet a flag to build a report
Bad Practice #3. Low level HTTP API
When Data Scientists
design an API...
Hydrosphere Mist - a service for exposing analytics
jobs and machine learning models as web services
Types of analytics services
- Enterprise Analytics services
- Reactive or Streaming services
- Realtime ML services
Enterprise analytics services
- Could not be
pre-calculated
- On-demand
parametrized jobs
- Requires large scale
processing
- Reporting
- Simulation (pricing, bank
stress testing, taxi rides)
- Forecasting (ad
campaign, energy
savings, others)
- Ad-hoc analytics tools
for business users
Demo #1
Reactive or Streaming services
Reactive or Streaming Reporting services
Demo #2
Realtime Machine Learning Services
Train models in Apache Spark and deploy it for realtime
low latency serving/scoring with high throughput
PMML is not an option
Spark ML, TensorFlow, H2O, Vowpal Wabbit, and every new ML
library invents uses own serialisation format
Format is not an issue if we re-define a deliverable for
ML model
xml, json, parquet, pojo, other
Single row Serving / Scoring
layer
Large Scale,
Batch
processing
engine
Monitoring,
testing
integration
Deliverable artifact for Machine Learning Model
Repository
Zooming out
MLLib model TensorFlow model Other model
Unified Serving/Scoring API
Demo #3
Agenda
- Challenges in deploying analytics into
production
- Deploying analytics as a service
- Feedback loops: testing,
monitoring, analytics of analytics
Testing, monitoring, analytics of analytics
- Poorly discussed in community.
- We are in production, baby!
- Regression.
- State matters. Model lifetime is limited.
- Data drifts, pipelines and model fail silently.
● Saves time
● Saves money
● Saves lifes
TDD world
TDD world does not work here
Pff… easy:
- Unit tests - by platform developers
- Integration tests - often impossible
Not clear who and not clear how:
- Regression
- Data Validation
- Production testing
- Data and ML pipelines quality monitoring
Need either “Data QA” & “Data Ops” people
or … AI
(formula for the next 10 000 startups - take something and add AI)
Smart data structures and dumb code works a lot
better than the other way around
Can we develop DSL and Data Structure which is
smart enough to learn from data patterns, trends and
anomalies to be self-QAed?
QA view: a universe vs. big data analytics system
People observe and monitor signals from stars to
check that universe is not broken today
Monitoring! Grafana, Kibana
And Marijuana to make sense out of it
Metrics processing, monitoring, correlation insights...
...Isn’t it a big data analytics task on its own?
ML pipeline Kafka
Analytics jobs
for metrics
Emit Metrics
Stream it back
into Spark
Context
Use insights to
make our data
structures
smart
Solution: loop of analytics for analytics
Benefits
● Don’t need to talk to Ops! :)
● Already have Apache Spark and Kafka in place
● Data Scientist in the loop!
● Unlimited flexibility in analytics, correlation and
using ML for ML
● Models could feeded back into Smart self
QA-ed data structures.
Hydrosphere Swirl - a system that creates a swirl of
analytics for analytics
Original ML
pipeline
Kafka
Streaming or
Batch Swirl
jobs
Hydrosphere
Swirl
Plug, modify,
deploy, run jobs &
consume results
Metrics
definition,
Notebook
integration
Hydrosphere
Mist
(1) Emit metrics
Hydrosphere Swirl: Vision
Demo #4
Classify by
sentiment
Twitter
Prepare
Data
Serve ads to
user
Hydrosphere Swirl
Invalid records 10/sec 2k/sec0.8Ratio Clicks
Swirl Demo: Serve Ads to users with positive Tweets
Classify by
sentiment
Twitter
Ingest &
transform
Serve ads to
user
Hydrosphere Swirl
Invalid records 20k/sec 10/sec0.2 Clicks
Data pipeline
is broken
Ratio
Swirl Demo: Serve Ads to users with positive Tweets
Twitter
Ingest &
transform
Serve ads to
user
Hydrosphere Swirl
Invalid records 10/sec 10/sec0.2 Clicks
New ML model
deployment
Deployed
bug in ML
code
Ratio
Swirl Demo: Serve Ads to users with positive Tweets
Thank you
Looking for
- Feedback
- Advisors, mentors & partners
- Pilots and early adopters
Stay in touch
- @hydrospheredata
- https://github.com/Hydrospheredata
- http://hydrosphere.io/
- spushkarev@hydrosphere.io

Contenu connexe

Tendances

Monitoring kubernetes with prometheus
Monitoring kubernetes with prometheusMonitoring kubernetes with prometheus
Monitoring kubernetes with prometheusBrice Fernandes
 
Metrics to Power DevOps
Metrics to Power DevOpsMetrics to Power DevOps
Metrics to Power DevOpsCollabNet
 
DevOps and Splunk
DevOps and SplunkDevOps and Splunk
DevOps and SplunkSplunk
 
MLOps with Azure DevOps
MLOps with Azure DevOpsMLOps with Azure DevOps
MLOps with Azure DevOpsMarco Parenzan
 
"Platform Engineering in practice — Why and How to start", Serg Hospodarets
"Platform Engineering in practice — Why and How to start", Serg Hospodarets "Platform Engineering in practice — Why and How to start", Serg Hospodarets
"Platform Engineering in practice — Why and How to start", Serg Hospodarets Fwdays
 
Fundamentals of DevOps and CI/CD
Fundamentals of DevOps and CI/CDFundamentals of DevOps and CI/CD
Fundamentals of DevOps and CI/CDBatyr Nuryyev
 
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation Slides
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation SlidesDevops On Cloud Powerpoint Template Slides Powerpoint Presentation Slides
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation SlidesSlideTeam
 
Introduction to kubernetes
Introduction to kubernetesIntroduction to kubernetes
Introduction to kubernetesRishabh Indoria
 
Ml ops intro session
Ml ops   intro sessionMl ops   intro session
Ml ops intro sessionAvinash Patil
 
Devops Devops Devops, at Froscon
Devops Devops Devops, at FrosconDevops Devops Devops, at Froscon
Devops Devops Devops, at FrosconKris Buytaert
 
DevOps- exec level briefing
DevOps-  exec level briefingDevOps-  exec level briefing
DevOps- exec level briefingRavi Tadwalkar
 
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...Chris Fregly
 
Kubeflow at Spotify (For the Kubeflow Summit)
Kubeflow at Spotify (For the Kubeflow Summit)Kubeflow at Spotify (For the Kubeflow Summit)
Kubeflow at Spotify (For the Kubeflow Summit)Josh Baer
 

Tendances (20)

MLOps for production-level machine learning
MLOps for production-level machine learningMLOps for production-level machine learning
MLOps for production-level machine learning
 
Monitoring kubernetes with prometheus
Monitoring kubernetes with prometheusMonitoring kubernetes with prometheus
Monitoring kubernetes with prometheus
 
Devops
DevopsDevops
Devops
 
Metrics to Power DevOps
Metrics to Power DevOpsMetrics to Power DevOps
Metrics to Power DevOps
 
DevOps and Splunk
DevOps and SplunkDevOps and Splunk
DevOps and Splunk
 
MLOps with Azure DevOps
MLOps with Azure DevOpsMLOps with Azure DevOps
MLOps with Azure DevOps
 
"Platform Engineering in practice — Why and How to start", Serg Hospodarets
"Platform Engineering in practice — Why and How to start", Serg Hospodarets "Platform Engineering in practice — Why and How to start", Serg Hospodarets
"Platform Engineering in practice — Why and How to start", Serg Hospodarets
 
Fundamentals of DevOps and CI/CD
Fundamentals of DevOps and CI/CDFundamentals of DevOps and CI/CD
Fundamentals of DevOps and CI/CD
 
DevOps Foundation
DevOps FoundationDevOps Foundation
DevOps Foundation
 
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation Slides
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation SlidesDevops On Cloud Powerpoint Template Slides Powerpoint Presentation Slides
Devops On Cloud Powerpoint Template Slides Powerpoint Presentation Slides
 
Introduction to kubernetes
Introduction to kubernetesIntroduction to kubernetes
Introduction to kubernetes
 
Ml ops intro session
Ml ops   intro sessionMl ops   intro session
Ml ops intro session
 
Devops Devops Devops, at Froscon
Devops Devops Devops, at FrosconDevops Devops Devops, at Froscon
Devops Devops Devops, at Froscon
 
DevOps
DevOpsDevOps
DevOps
 
DevOps- exec level briefing
DevOps-  exec level briefingDevOps-  exec level briefing
DevOps- exec level briefing
 
Past, Present and Future of DevOps Infrastructure
Past, Present and Future of DevOps InfrastructurePast, Present and Future of DevOps Infrastructure
Past, Present and Future of DevOps Infrastructure
 
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...
KubeFlow + GPU + Keras/TensorFlow 2.0 + TF Extended (TFX) + Kubernetes + PyTo...
 
Kubeflow at Spotify (For the Kubeflow Summit)
Kubeflow at Spotify (For the Kubeflow Summit)Kubeflow at Spotify (For the Kubeflow Summit)
Kubeflow at Spotify (For the Kubeflow Summit)
 
What is MLOps
What is MLOpsWhat is MLOps
What is MLOps
 
DevOps
DevOpsDevOps
DevOps
 

Similaire à DevOps for DataScience

Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning ModelsApache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning ModelsAnyscale
 
Serverless machine learning operations
Serverless machine learning operationsServerless machine learning operations
Serverless machine learning operationsStepan Pushkarev
 
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...All Things Open
 
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)dtz001
 
Architecting an Open Source AI Platform 2018 edition
Architecting an Open Source AI Platform   2018 editionArchitecting an Open Source AI Platform   2018 edition
Architecting an Open Source AI Platform 2018 editionDavid Talby
 
mlflow: Accelerating the End-to-End ML lifecycle
mlflow: Accelerating the End-to-End ML lifecyclemlflow: Accelerating the End-to-End ML lifecycle
mlflow: Accelerating the End-to-End ML lifecycleDatabricks
 
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....Databricks
 
Machine Learning Models in Production
Machine Learning Models in ProductionMachine Learning Models in Production
Machine Learning Models in ProductionDataWorks Summit
 
EPAM ML/AI Accelerator - ODAHU
EPAM ML/AI Accelerator - ODAHUEPAM ML/AI Accelerator - ODAHU
EPAM ML/AI Accelerator - ODAHUDmitrii Suslov
 
Hydrosphere.io Platform for AI/ML Operations Automation
Hydrosphere.io Platform for AI/ML Operations AutomationHydrosphere.io Platform for AI/ML Operations Automation
Hydrosphere.io Platform for AI/ML Operations AutomationRustem Zakiev
 
Spark and machine learning in microservices architecture
Spark and machine learning in microservices architectureSpark and machine learning in microservices architecture
Spark and machine learning in microservices architectureStepan Pushkarev
 
IBM Strategy for Spark
IBM Strategy for SparkIBM Strategy for Spark
IBM Strategy for SparkMark Kerzner
 
Machine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabsMachine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabszekeLabs Technologies
 
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...James Anderson
 
Feature Store as a Data Foundation for Machine Learning
Feature Store as a Data Foundation for Machine LearningFeature Store as a Data Foundation for Machine Learning
Feature Store as a Data Foundation for Machine LearningProvectus
 
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache Spark
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache SparkData-Driven Transformation: Leveraging Big Data at Showtime with Apache Spark
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache SparkDatabricks
 
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...ScyllaDB
 
Continuous delivery for machine learning
Continuous delivery for machine learningContinuous delivery for machine learning
Continuous delivery for machine learningRajesh Muppalla
 
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...Karthik Murugesan
 

Similaire à DevOps for DataScience (20)

Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning ModelsApache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
Apache ® Spark™ MLlib 2.x: How to Productionize your Machine Learning Models
 
03_aiops-1.pptx
03_aiops-1.pptx03_aiops-1.pptx
03_aiops-1.pptx
 
Serverless machine learning operations
Serverless machine learning operationsServerless machine learning operations
Serverless machine learning operations
 
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
Deployment Design Patterns - Deploying Machine Learning and Deep Learning Mod...
 
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
AllThingsOpen 2018 - Deployment Design Patterns (Dan Zaratsian)
 
Architecting an Open Source AI Platform 2018 edition
Architecting an Open Source AI Platform   2018 editionArchitecting an Open Source AI Platform   2018 edition
Architecting an Open Source AI Platform 2018 edition
 
mlflow: Accelerating the End-to-End ML lifecycle
mlflow: Accelerating the End-to-End ML lifecyclemlflow: Accelerating the End-to-End ML lifecycle
mlflow: Accelerating the End-to-End ML lifecycle
 
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
How to Productionize Your Machine Learning Models Using Apache Spark MLlib 2....
 
Machine Learning Models in Production
Machine Learning Models in ProductionMachine Learning Models in Production
Machine Learning Models in Production
 
EPAM ML/AI Accelerator - ODAHU
EPAM ML/AI Accelerator - ODAHUEPAM ML/AI Accelerator - ODAHU
EPAM ML/AI Accelerator - ODAHU
 
Hydrosphere.io Platform for AI/ML Operations Automation
Hydrosphere.io Platform for AI/ML Operations AutomationHydrosphere.io Platform for AI/ML Operations Automation
Hydrosphere.io Platform for AI/ML Operations Automation
 
Spark and machine learning in microservices architecture
Spark and machine learning in microservices architectureSpark and machine learning in microservices architecture
Spark and machine learning in microservices architecture
 
IBM Strategy for Spark
IBM Strategy for SparkIBM Strategy for Spark
IBM Strategy for Spark
 
Machine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabsMachine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabs
 
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...
GDG Cloud Southlake #16: Priyanka Vergadia: Scalable Data Analytics in Google...
 
Feature Store as a Data Foundation for Machine Learning
Feature Store as a Data Foundation for Machine LearningFeature Store as a Data Foundation for Machine Learning
Feature Store as a Data Foundation for Machine Learning
 
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache Spark
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache SparkData-Driven Transformation: Leveraging Big Data at Showtime with Apache Spark
Data-Driven Transformation: Leveraging Big Data at Showtime with Apache Spark
 
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...
Simplifying the Creation of Machine Learning Workflow Pipelines for IoT Appli...
 
Continuous delivery for machine learning
Continuous delivery for machine learningContinuous delivery for machine learning
Continuous delivery for machine learning
 
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...
Uber - Building Intelligent Applications, Experimental ML with Uber’s Data Sc...
 

Plus de Stepan Pushkarev

AI for the Human Retina to Protect Newborn Vision
AI for the Human Retina to Protect Newborn VisionAI for the Human Retina to Protect Newborn Vision
AI for the Human Retina to Protect Newborn VisionStepan Pushkarev
 
Automating machine learning lifecycle with kubeflow
Automating machine learning lifecycle with kubeflowAutomating machine learning lifecycle with kubeflow
Automating machine learning lifecycle with kubeflowStepan Pushkarev
 
Handling inference in anomalous ever changing environment
Handling inference in anomalous ever changing environmentHandling inference in anomalous ever changing environment
Handling inference in anomalous ever changing environmentStepan Pushkarev
 
Data ops: Machine Learning in production
Data ops: Machine Learning in productionData ops: Machine Learning in production
Data ops: Machine Learning in productionStepan Pushkarev
 
Multi runtime serving pipelines for machine learning
Multi runtime serving pipelines for machine learningMulti runtime serving pipelines for machine learning
Multi runtime serving pipelines for machine learningStepan Pushkarev
 

Plus de Stepan Pushkarev (7)

AI for the Human Retina to Protect Newborn Vision
AI for the Human Retina to Protect Newborn VisionAI for the Human Retina to Protect Newborn Vision
AI for the Human Retina to Protect Newborn Vision
 
Automating machine learning lifecycle with kubeflow
Automating machine learning lifecycle with kubeflowAutomating machine learning lifecycle with kubeflow
Automating machine learning lifecycle with kubeflow
 
Handling inference in anomalous ever changing environment
Handling inference in anomalous ever changing environmentHandling inference in anomalous ever changing environment
Handling inference in anomalous ever changing environment
 
Monitoring AI with AI
Monitoring AI with AIMonitoring AI with AI
Monitoring AI with AI
 
Data ops: Machine Learning in production
Data ops: Machine Learning in productionData ops: Machine Learning in production
Data ops: Machine Learning in production
 
Multi runtime serving pipelines for machine learning
Multi runtime serving pipelines for machine learningMulti runtime serving pipelines for machine learning
Multi runtime serving pipelines for machine learning
 
Spark ML Pipeline serving
Spark ML Pipeline servingSpark ML Pipeline serving
Spark ML Pipeline serving
 

Dernier

introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdf
introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdfintroduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdf
introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdfVishalKumarJha10
 
Pharm-D Biostatistics and Research methodology
Pharm-D Biostatistics and Research methodologyPharm-D Biostatistics and Research methodology
Pharm-D Biostatistics and Research methodologyAnusha Are
 
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxBUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxalwaysnagaraju26
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerThousandEyes
 
Chinsurah Escorts ☎️8617697112 Starting From 5K to 15K High Profile Escorts ...
Chinsurah Escorts ☎️8617697112  Starting From 5K to 15K High Profile Escorts ...Chinsurah Escorts ☎️8617697112  Starting From 5K to 15K High Profile Escorts ...
Chinsurah Escorts ☎️8617697112 Starting From 5K to 15K High Profile Escorts ...Nitya salvi
 
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfonteinmasabamasaba
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsArshad QA
 
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...Jittipong Loespradit
 
A Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docxA Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docxComplianceQuest1
 
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Steffen Staab
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisamasabamasaba
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVshikhaohhpro
 
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️Delhi Call girls
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...panagenda
 
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456KiaraTiradoMicha
 
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
Direct Style Effect Systems -The Print[A] Example- A Comprehension AidDirect Style Effect Systems -The Print[A] Example- A Comprehension Aid
Direct Style Effect Systems - The Print[A] Example - A Comprehension AidPhilip Schwarz
 
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrainmasabamasaba
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...ICS
 

Dernier (20)

introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdf
introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdfintroduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdf
introduction-to-automotive Andoid os-csimmonds-ndctechtown-2021.pdf
 
Pharm-D Biostatistics and Research methodology
Pharm-D Biostatistics and Research methodologyPharm-D Biostatistics and Research methodology
Pharm-D Biostatistics and Research methodology
 
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptxBUS PASS MANGEMENT SYSTEM USING PHP.pptx
BUS PASS MANGEMENT SYSTEM USING PHP.pptx
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
 
Chinsurah Escorts ☎️8617697112 Starting From 5K to 15K High Profile Escorts ...
Chinsurah Escorts ☎️8617697112  Starting From 5K to 15K High Profile Escorts ...Chinsurah Escorts ☎️8617697112  Starting From 5K to 15K High Profile Escorts ...
Chinsurah Escorts ☎️8617697112 Starting From 5K to 15K High Profile Escorts ...
 
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
%in kaalfontein+277-882-255-28 abortion pills for sale in kaalfontein
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
 
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
MarTech Trend 2024 Book : Marketing Technology Trends (2024 Edition) How Data...
 
Microsoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdfMicrosoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdf
 
A Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docxA Secure and Reliable Document Management System is Essential.docx
A Secure and Reliable Document Management System is Essential.docx
 
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
Shapes for Sharing between Graph Data Spaces - and Epistemic Querying of RDF-...
 
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa%in tembisa+277-882-255-28 abortion pills for sale in tembisa
%in tembisa+277-882-255-28 abortion pills for sale in tembisa
 
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTV
 
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
 
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456LEVEL 5   - SESSION 1 2023 (1).pptx - PDF 123456
LEVEL 5 - SESSION 1 2023 (1).pptx - PDF 123456
 
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
Direct Style Effect Systems -The Print[A] Example- A Comprehension AidDirect Style Effect Systems -The Print[A] Example- A Comprehension Aid
Direct Style Effect Systems - The Print[A] Example - A Comprehension Aid
 
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
%in Bahrain+277-882-255-28 abortion pills for sale in Bahrain
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
 

DevOps for DataScience

  • 1. DevOps for Data Science by Stepan Pushkarev CTO of Hydrosphere.io
  • 2. DevOps is a catchy buzzword to optimise things
  • 4. Is there life after marriage data science? Dating, Flowers, Dreams Marriage Happily lived forever? Collect & prepare data Build ML Model
  • 5. This talk is for people who are married aware of “other 99% of data science” Dating, Flowers, Dreams Marriage Happily lived forever? Collect & prepare data Build ML Model
  • 6. This talk is NOT about - Setting up Apache Spark/Hadoop cluster - Configuring CI/CD tools like Jenkins - Configuring monitoring tools & dashboards - Agile/DevOps brainwashing & consulting story
  • 7. Agenda - Challenges in deploying analytics into production - Deploying analytics as a service - Feedback loops: testing, monitoring, analytics of analytics
  • 8. Why do companies hire data scientists?
  • 9. Why do companies hire data scientists? To make products smarter.
  • 10. What is a deliverable of data scientist and data engineer?
  • 11. What is a deliverable of data scientist? Academic paper? ML Model? R/Python script? Jupiter Notebook? BI Dashboard?
  • 12. What has to be a deliverable of data scientist? Data pipelines and machine learning models that deployed as pluggable, testable, supportable, monitorable analytics services.
  • 13. Option 1: Engineer to implement academic paper
  • 14. Option 2: Engineer to re-implement R/Python script
  • 15. Option 3: Run notebook as it is using cron
  • 16. Option 3: Run notebook as it is using cron
  • 17. Option 4: Build software to eat the world Data Science
  • 18. Eating data science © Daniel Tunkelang - Where should you put your data scientists? - www.slideshare.net/dtunkelang/where-should-you-put-your-data-scientists Step 1 (management): Integrate data scientists into cross-functional teams
  • 19. Eating data science Step 2 (operations): Make environments scalable and elastic. Finally.
  • 20. Eating data science Step 3: Make data scientists to write less code
  • 21. Eating data science Step 4: Deploy analytics as services
  • 22. Step 5: Use feedback loops: testing, monitoring, analytics for analytics Build ML Model Test Monitor, maintain, analyze Deploy as a service Collect & prepare data
  • 23. Agenda - Challenges in deploying analytics into production - Deploying analytics as a service - Feedback loops: testing, monitoring, analytics of analytics
  • 24. Deploying analytics as a service - Defines deliverable for Data Scientist / Data Engineer. - Plugs analytics into end-to-end products through API. - With the right tooling allows Data Scientist to deploy it in self serve
  • 25. Look around - proprietary ML based APIs - Alchemy API - Google Prediction API - Cloud Vision API - Azure ML Can we do our own on top of Apache Spark?
  • 26. Bad Practice #1. Business logic in Spark? WTF?
  • 27. Bad Practice #2. Database as API Execute reporting job Mark Job as complete & save result Poll for new tasks Poll for resultSet a flag to build a report
  • 28. Bad Practice #3. Low level HTTP API When Data Scientists design an API...
  • 29. Hydrosphere Mist - a service for exposing analytics jobs and machine learning models as web services
  • 30. Types of analytics services - Enterprise Analytics services - Reactive or Streaming services - Realtime ML services
  • 31. Enterprise analytics services - Could not be pre-calculated - On-demand parametrized jobs - Requires large scale processing - Reporting - Simulation (pricing, bank stress testing, taxi rides) - Forecasting (ad campaign, energy savings, others) - Ad-hoc analytics tools for business users
  • 34. Reactive or Streaming Reporting services
  • 36. Realtime Machine Learning Services Train models in Apache Spark and deploy it for realtime low latency serving/scoring with high throughput
  • 37. PMML is not an option Spark ML, TensorFlow, H2O, Vowpal Wabbit, and every new ML library invents uses own serialisation format
  • 38. Format is not an issue if we re-define a deliverable for ML model xml, json, parquet, pojo, other Single row Serving / Scoring layer Large Scale, Batch processing engine Monitoring, testing integration Deliverable artifact for Machine Learning Model
  • 39. Repository Zooming out MLLib model TensorFlow model Other model Unified Serving/Scoring API
  • 41. Agenda - Challenges in deploying analytics into production - Deploying analytics as a service - Feedback loops: testing, monitoring, analytics of analytics
  • 42. Testing, monitoring, analytics of analytics - Poorly discussed in community. - We are in production, baby! - Regression. - State matters. Model lifetime is limited. - Data drifts, pipelines and model fail silently. ● Saves time ● Saves money ● Saves lifes
  • 44. TDD world does not work here Pff… easy: - Unit tests - by platform developers - Integration tests - often impossible Not clear who and not clear how: - Regression - Data Validation - Production testing - Data and ML pipelines quality monitoring
  • 45. Need either “Data QA” & “Data Ops” people or … AI (formula for the next 10 000 startups - take something and add AI)
  • 46. Smart data structures and dumb code works a lot better than the other way around
  • 47. Can we develop DSL and Data Structure which is smart enough to learn from data patterns, trends and anomalies to be self-QAed?
  • 48. QA view: a universe vs. big data analytics system People observe and monitor signals from stars to check that universe is not broken today
  • 50. And Marijuana to make sense out of it
  • 51. Metrics processing, monitoring, correlation insights... ...Isn’t it a big data analytics task on its own?
  • 52.
  • 53. ML pipeline Kafka Analytics jobs for metrics Emit Metrics Stream it back into Spark Context Use insights to make our data structures smart Solution: loop of analytics for analytics
  • 54. Benefits ● Don’t need to talk to Ops! :) ● Already have Apache Spark and Kafka in place ● Data Scientist in the loop! ● Unlimited flexibility in analytics, correlation and using ML for ML ● Models could feeded back into Smart self QA-ed data structures.
  • 55. Hydrosphere Swirl - a system that creates a swirl of analytics for analytics
  • 56. Original ML pipeline Kafka Streaming or Batch Swirl jobs Hydrosphere Swirl Plug, modify, deploy, run jobs & consume results Metrics definition, Notebook integration Hydrosphere Mist (1) Emit metrics Hydrosphere Swirl: Vision
  • 58. Classify by sentiment Twitter Prepare Data Serve ads to user Hydrosphere Swirl Invalid records 10/sec 2k/sec0.8Ratio Clicks Swirl Demo: Serve Ads to users with positive Tweets
  • 59. Classify by sentiment Twitter Ingest & transform Serve ads to user Hydrosphere Swirl Invalid records 20k/sec 10/sec0.2 Clicks Data pipeline is broken Ratio Swirl Demo: Serve Ads to users with positive Tweets
  • 60. Twitter Ingest & transform Serve ads to user Hydrosphere Swirl Invalid records 10/sec 10/sec0.2 Clicks New ML model deployment Deployed bug in ML code Ratio Swirl Demo: Serve Ads to users with positive Tweets
  • 61. Thank you Looking for - Feedback - Advisors, mentors & partners - Pilots and early adopters Stay in touch - @hydrospheredata - https://github.com/Hydrospheredata - http://hydrosphere.io/ - spushkarev@hydrosphere.io