SlideShare a Scribd company logo
1 of 21
Optimization at Alibaba
Zhenliang Zhang
iDST at Alibaba
Alibaba Ecosystem
- --
-- $ -
- D F
)
F
1B F B
DB: EE B
1 DH DBH DE
F
( F DF F
,B EF E
-B DBIE D
,B F B E1B -
Optimization Problems Found Everywhere
Data Center
Optimization
Resource
Optimization
Large-scale
Nonconvex
Optimization
E-commerce Logistics Smart City Digital Entertainment
Computing
Infrastructure
Computing
Platform
Machine Learning
Platform
Business
E-commerce
• Connecting consumers with sellers/products
– Efficiency is the key of any e-commerce platform
– Alibaba: 0.5 Bn users, 1 Bn products, GMV 459Bn
• Two key problems
– Prediction: accurately predict users' needs
– Matching: effectively match users' needs with products
Predicting User Needs
Trans.
Behavior
LBS
User Profiles
Real-time data Contextual data
Prediction
Effectively Matching User Needs
Ease case: supplies ³ demands
• A simple greedy matching works best
• Match each user with the best prediction
Difficult case: supplies < demands
• How to decide which users will get which items ?
• Greedy algorithm does not work, leading to assignment problem
Effectively Matching User Needs
Assignment Problem
• Given a bipartite graph
– m agents & n tasks
– An award ci,j is given when agent i is
assigned to perform task j
• Optimal task assignment
– Goal: maximize overall awards
– Constraints: each agent/task can be
assigned to a limited number of
tasks/agents
0.8
0.1
ci,j
Applications in Alibaba
• Online advertisement
– Match users with different ads
– Awards: the number of clicks
– Constraints: budgeted number of
impressions for individual ads
• Significance of optimization
– Improved revenue by ~50 m RMB per
day
Users Ads
Ask-All ( LM)
Background
• Allow customers to raise questions about products
before making purchases
• System distributes questions to users who could
provide answers within an hour
Goal
• Maximize the number of questions answered
Constraints
• For better experience, each user will only be asked
for a small number of questions QuestionsUsers
Example of Ask-All
• Mobile message push
• Online traffic allocation
• Online distribution of coupons
• Stock aware online recommendation
• Online assignment of delivery requests
• …
Other Applications in Alibaba
Cloud Computing
• The largest cloud computing company in China
– 1Million users, ~100% YoY revenue growth
• Need to improve resource management
– Reliability: service-level agreement
– Pricing & cost reduction
Computing Infrastructure at Alibaba
• 6 data centers in mainland China
• 8 data centers globally, 3 more are coming
• Power management
– US data centers consumed ~70
billion kilowatt-hours of electricity
• Network planning
– Minimize the latency
Example: Scheduling Problem
• Online multi-resource allocation problem, e.g. CPU, Mem, Disk
• Generally NP hard
• Need to make online decision
Compute
Resources
Scheduler Pending Jobs
Example: Traffic Balancing
• Assign projects to clusters with minimal cost
– Dependent projects need to communicate,
– But traffic between clusters are expensive
• Constraints
– Network band restriction
– Some project migration are infeasible
Logistics in Alibaba
• Cainiao network
– 80 major warehouse networks, 4000 Cainiao partners
– 0.5 Bn parcels/day, next day delivery service in 150 cities
• Key areas
Supply chain Warehouse
management
Transportation
management
Applications of Optimization in Logistics
• 3D bin packing
– Find smallest number of boxes to
include all items from an order
• Warehouse management
– New retailer Hema
• Vehicle route planning
– Planning for order pickup
– Planning for delivery service
& B( B D
,B EF E:D EFD F D
A A
&B C F F B D EB D F
E
. EE F
A A
BI D F F F D FIBD
C
0B F I D B E F
C H DB F : F B F B
C
Focus of Our Team
Optimization at Alibaba
- --
-- $ -
- D F
)
F
1B F B
DB: EE B
1 DH DBH DE
F
( F DF F
,B EF E
-B DBIE D
,B F B E1B -
n .CF F B E
( (0 (0(
n & E
n F DF F EE
n . BCF F B
n , D E F
n E EE B E

More Related Content

Similar to Deep learning optimization at alibaba by zhenliang zhang from Alibaba

Dubbo and Weidian's practice on micro-service architecture
Dubbo and Weidian's practice on micro-service architectureDubbo and Weidian's practice on micro-service architecture
Dubbo and Weidian's practice on micro-service architectureHuxing Zhang
 
e-Minggu 1 Pengantar e-Business & e-Commerce.ppt
e-Minggu 1 Pengantar e-Business & e-Commerce.ppte-Minggu 1 Pengantar e-Business & e-Commerce.ppt
e-Minggu 1 Pengantar e-Business & e-Commerce.pptErwinEruWidodo
 
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...NoSQLmatters
 
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...IBM India Smarter Computing
 
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...IBM India Smarter Computing
 
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼Elasticsearch
 
Monitoring aws finance spend
Monitoring aws finance spendMonitoring aws finance spend
Monitoring aws finance spendJohn Malone
 
Enterprise architectsview 2015-apr
Enterprise architectsview 2015-aprEnterprise architectsview 2015-apr
Enterprise architectsview 2015-aprMongoDB
 
Accounts Receivable Management Software Buyers Guide
Accounts Receivable Management Software Buyers GuideAccounts Receivable Management Software Buyers Guide
Accounts Receivable Management Software Buyers GuideE2B Teknologies
 
Digital Strategy for future business
Digital Strategy for future businessDigital Strategy for future business
Digital Strategy for future businessAshish Bhasin
 
Enabling real interactive BI on Hadoop
Enabling real interactive BI on HadoopEnabling real interactive BI on Hadoop
Enabling real interactive BI on HadoopDataWorks Summit
 
GraphTour - Popular Use Cases
GraphTour - Popular Use CasesGraphTour - Popular Use Cases
GraphTour - Popular Use CasesNeo4j
 
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_final
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_finalEpgp(one year) 2009-10_it strategy_groupassignment_case-dell_final
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_finalRajendra Inani
 
Hadoop Boosts Profits in Media and Telecom Industry
Hadoop Boosts Profits in Media and Telecom IndustryHadoop Boosts Profits in Media and Telecom Industry
Hadoop Boosts Profits in Media and Telecom IndustryDataWorks Summit
 
Project presentation.pptx
Project presentation.pptxProject presentation.pptx
Project presentation.pptxMRABD3
 
How to Achieve Digital World Class Performance with AR Automation
How to Achieve Digital World Class Performance with AR AutomationHow to Achieve Digital World Class Performance with AR Automation
How to Achieve Digital World Class Performance with AR AutomationEmagia
 
Zeng Ming of Alibaba: "Making Sense of Big Data"
Zeng Ming of Alibaba: "Making Sense of Big Data"Zeng Ming of Alibaba: "Making Sense of Big Data"
Zeng Ming of Alibaba: "Making Sense of Big Data"sprie-stanford
 

Similar to Deep learning optimization at alibaba by zhenliang zhang from Alibaba (20)

Dubbo and Weidian's practice on micro-service architecture
Dubbo and Weidian's practice on micro-service architectureDubbo and Weidian's practice on micro-service architecture
Dubbo and Weidian's practice on micro-service architecture
 
e-Minggu 1 Pengantar e-Business & e-Commerce.ppt
e-Minggu 1 Pengantar e-Business & e-Commerce.ppte-Minggu 1 Pengantar e-Business & e-Commerce.ppt
e-Minggu 1 Pengantar e-Business & e-Commerce.ppt
 
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...
Kai Wähner – Real World Use Cases for Realtime In-Memory Computing - NoSQL ma...
 
Yodle Ecosystem Trends
Yodle Ecosystem Trends Yodle Ecosystem Trends
Yodle Ecosystem Trends
 
Smart DataCenters
Smart DataCentersSmart DataCenters
Smart DataCenters
 
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
 
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
SAP and IBM Demonstrate Capability of Handling High Billing Volume in a Telec...
 
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼
Customer Story: Elastic Stack을 이용한 게임 서비스 통합 로깅 플랫폼
 
Monitoring aws finance spend
Monitoring aws finance spendMonitoring aws finance spend
Monitoring aws finance spend
 
Enterprise architectsview 2015-apr
Enterprise architectsview 2015-aprEnterprise architectsview 2015-apr
Enterprise architectsview 2015-apr
 
Accounts Receivable Management Software Buyers Guide
Accounts Receivable Management Software Buyers GuideAccounts Receivable Management Software Buyers Guide
Accounts Receivable Management Software Buyers Guide
 
Digital Strategy for future business
Digital Strategy for future businessDigital Strategy for future business
Digital Strategy for future business
 
Enabling real interactive BI on Hadoop
Enabling real interactive BI on HadoopEnabling real interactive BI on Hadoop
Enabling real interactive BI on Hadoop
 
GraphTour - Popular Use Cases
GraphTour - Popular Use CasesGraphTour - Popular Use Cases
GraphTour - Popular Use Cases
 
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_final
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_finalEpgp(one year) 2009-10_it strategy_groupassignment_case-dell_final
Epgp(one year) 2009-10_it strategy_groupassignment_case-dell_final
 
Hadoop Boosts Profits in Media and Telecom Industry
Hadoop Boosts Profits in Media and Telecom IndustryHadoop Boosts Profits in Media and Telecom Industry
Hadoop Boosts Profits in Media and Telecom Industry
 
Project presentation.pptx
Project presentation.pptxProject presentation.pptx
Project presentation.pptx
 
Big data in telecom
Big data in telecomBig data in telecom
Big data in telecom
 
How to Achieve Digital World Class Performance with AR Automation
How to Achieve Digital World Class Performance with AR AutomationHow to Achieve Digital World Class Performance with AR Automation
How to Achieve Digital World Class Performance with AR Automation
 
Zeng Ming of Alibaba: "Making Sense of Big Data"
Zeng Ming of Alibaba: "Making Sense of Big Data"Zeng Ming of Alibaba: "Making Sense of Big Data"
Zeng Ming of Alibaba: "Making Sense of Big Data"
 

More from Bill Liu

Walk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectWalk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectBill Liu
 
Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Bill Liu
 
Productizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeProductizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeBill Liu
 
Transformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroTransformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroBill Liu
 
Deep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsDeep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsBill Liu
 
Metaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixMetaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixBill Liu
 
Practical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScalePractical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScaleBill Liu
 
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBuilding large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBill Liu
 
Deep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsDeep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsBill Liu
 
Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Bill Liu
 
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsHighly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsBill Liu
 
Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Bill Liu
 
Causal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningCausal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningBill Liu
 
Weekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileWeekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileBill Liu
 
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningWeekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningBill Liu
 
AISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsAISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsBill Liu
 
AISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldAISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldBill Liu
 
AISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeAISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeBill Liu
 
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...Bill Liu
 
Toronto meetup 20190917
Toronto meetup 20190917Toronto meetup 20190917
Toronto meetup 20190917Bill Liu
 

More from Bill Liu (20)

Walk Through a Real World ML Production Project
Walk Through a Real World ML Production ProjectWalk Through a Real World ML Production Project
Walk Through a Real World ML Production Project
 
Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...Redefining MLOps with Model Deployment, Management and Observability in Produ...
Redefining MLOps with Model Deployment, Management and Observability in Produ...
 
Productizing Machine Learning at the Edge
Productizing Machine Learning at the EdgeProductizing Machine Learning at the Edge
Productizing Machine Learning at the Edge
 
Transformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to HeroTransformers in Vision: From Zero to Hero
Transformers in Vision: From Zero to Hero
 
Deep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps WorkflowsDeep AutoViML For Tensorflow Models and MLOps Workflows
Deep AutoViML For Tensorflow Models and MLOps Workflows
 
Metaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at NetflixMetaflow: The ML Infrastructure at Netflix
Metaflow: The ML Infrastructure at Netflix
 
Practical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at ScalePractical Crowdsourcing for ML at Scale
Practical Crowdsourcing for ML at Scale
 
Building large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudiBuilding large scale transactional data lake using apache hudi
Building large scale transactional data lake using apache hudi
 
Deep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its ApplicationsDeep Reinforcement Learning and Its Applications
Deep Reinforcement Learning and Its Applications
 
Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19Big Data and AI in Fighting Against COVID-19
Big Data and AI in Fighting Against COVID-19
 
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world ApplicationsHighly-scalable Reinforcement Learning RLlib for Real-world Applications
Highly-scalable Reinforcement Learning RLlib for Real-world Applications
 
Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...Build computer vision models to perform object detection and classification w...
Build computer vision models to perform object detection and classification w...
 
Causal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine LearningCausal Inference in Data Science and Machine Learning
Causal Inference in Data Science and Machine Learning
 
Weekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on MobileWeekly #106: Deep Learning on Mobile
Weekly #106: Deep Learning on Mobile
 
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine LearningWeekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
Weekly #105: AutoViz and Auto_ViML Visualization and Machine Learning
 
AISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with MicroeconomicsAISF19 - On Blending Machine Learning with Microeconomics
AISF19 - On Blending Machine Learning with Microeconomics
 
AISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First WorldAISF19 - Travel in the AI-First World
AISF19 - Travel in the AI-First World
 
AISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the EdgeAISF19 - Unleash Computer Vision at the Edge
AISF19 - Unleash Computer Vision at the Edge
 
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
AISF19 - Building Scalable, Kubernetes-Native ML/AI Pipelines with TFX, KubeF...
 
Toronto meetup 20190917
Toronto meetup 20190917Toronto meetup 20190917
Toronto meetup 20190917
 

Recently uploaded

08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksSoftradix Technologies
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphNeo4j
 
Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Hyundai Motor Group
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAndikSusilo4
 
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptx
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptxMaking_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptx
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptxnull - The Open Security Community
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraDeakin University
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsHyundai Motor Group
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 

Recently uploaded (20)

08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other Frameworks
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
 
Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2Next-generation AAM aircraft unveiled by Supernal, S-A2
Next-generation AAM aircraft unveiled by Supernal, S-A2
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & Application
 
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptx
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptxMaking_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptx
Making_way_through_DLL_hollowing_inspite_of_CFG_by_Debjeet Banerjee.pptx
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning era
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 

Deep learning optimization at alibaba by zhenliang zhang from Alibaba

  • 1.
  • 2. Optimization at Alibaba Zhenliang Zhang iDST at Alibaba
  • 3. Alibaba Ecosystem - -- -- $ - - D F ) F 1B F B DB: EE B 1 DH DBH DE F ( F DF F ,B EF E -B DBIE D ,B F B E1B -
  • 4. Optimization Problems Found Everywhere Data Center Optimization Resource Optimization Large-scale Nonconvex Optimization E-commerce Logistics Smart City Digital Entertainment Computing Infrastructure Computing Platform Machine Learning Platform Business
  • 5. E-commerce • Connecting consumers with sellers/products – Efficiency is the key of any e-commerce platform – Alibaba: 0.5 Bn users, 1 Bn products, GMV 459Bn • Two key problems – Prediction: accurately predict users' needs – Matching: effectively match users' needs with products
  • 6. Predicting User Needs Trans. Behavior LBS User Profiles Real-time data Contextual data Prediction
  • 7. Effectively Matching User Needs Ease case: supplies ³ demands • A simple greedy matching works best • Match each user with the best prediction
  • 8. Difficult case: supplies < demands • How to decide which users will get which items ? • Greedy algorithm does not work, leading to assignment problem Effectively Matching User Needs
  • 9. Assignment Problem • Given a bipartite graph – m agents & n tasks – An award ci,j is given when agent i is assigned to perform task j • Optimal task assignment – Goal: maximize overall awards – Constraints: each agent/task can be assigned to a limited number of tasks/agents 0.8 0.1 ci,j
  • 10. Applications in Alibaba • Online advertisement – Match users with different ads – Awards: the number of clicks – Constraints: budgeted number of impressions for individual ads • Significance of optimization – Improved revenue by ~50 m RMB per day Users Ads
  • 11. Ask-All ( LM) Background • Allow customers to raise questions about products before making purchases • System distributes questions to users who could provide answers within an hour Goal • Maximize the number of questions answered Constraints • For better experience, each user will only be asked for a small number of questions QuestionsUsers
  • 13. • Mobile message push • Online traffic allocation • Online distribution of coupons • Stock aware online recommendation • Online assignment of delivery requests • … Other Applications in Alibaba
  • 14. Cloud Computing • The largest cloud computing company in China – 1Million users, ~100% YoY revenue growth • Need to improve resource management – Reliability: service-level agreement – Pricing & cost reduction
  • 15. Computing Infrastructure at Alibaba • 6 data centers in mainland China • 8 data centers globally, 3 more are coming • Power management – US data centers consumed ~70 billion kilowatt-hours of electricity • Network planning – Minimize the latency
  • 16. Example: Scheduling Problem • Online multi-resource allocation problem, e.g. CPU, Mem, Disk • Generally NP hard • Need to make online decision Compute Resources Scheduler Pending Jobs
  • 17. Example: Traffic Balancing • Assign projects to clusters with minimal cost – Dependent projects need to communicate, – But traffic between clusters are expensive • Constraints – Network band restriction – Some project migration are infeasible
  • 18. Logistics in Alibaba • Cainiao network – 80 major warehouse networks, 4000 Cainiao partners – 0.5 Bn parcels/day, next day delivery service in 150 cities • Key areas Supply chain Warehouse management Transportation management
  • 19. Applications of Optimization in Logistics • 3D bin packing – Find smallest number of boxes to include all items from an order • Warehouse management – New retailer Hema • Vehicle route planning – Planning for order pickup – Planning for delivery service
  • 20. & B( B D ,B EF E:D EFD F D A A &B C F F B D EB D F E . EE F A A BI D F F F D FIBD C 0B F I D B E F C H DB F : F B F B C Focus of Our Team
  • 21. Optimization at Alibaba - -- -- $ - - D F ) F 1B F B DB: EE B 1 DH DBH DE F ( F DF F ,B EF E -B DBIE D ,B F B E1B - n .CF F B E ( (0 (0( n & E n F DF F EE n . BCF F B n , D E F n E EE B E