Driving Digital Transformation through Big Data Analytics and Machine Learning
Download as PPTX, PDF1 like1,650 views
The document outlines Verizon's capabilities in digital transformation through big data analytics, machine learning, and cloud infrastructure. It emphasizes the importance of customer experience, the rise of customer expectations, and the need for personalized interactions. The document also details Verizon's technology infrastructure, including their big data and AI platform and security measures for identity and access management.
Driving Digital Transformation through Big Data Analytics and Machine Learning
- 1. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. Prachi Govalkar Senior Dev Ops Engineer Dinesh Lokhande Distinguish Engineer, Software Development Sept. 27, 2017 Driving digital transformation through big data analytics and machine learning
- 2. ● Cloud Infrastructure ● API and micro services driven framework ● Identity and Access Management ● Security ● Analytics ● ... Digital transformation playbook © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 3. “Software is eating the world” – Marc Andreessen, August 20,2011 Digital transformation © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 4. ● U.S. coverage ● 98% wireless network coverage ● Broadband ● 100% fiber optics network ● Video & Advertising ● 150M hours of video streamed monthly ● Security ● 200+ data centers in 24 countries Verizon - Delivering the connected world © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 5. “You’ve got to start with the customer experience and work backwards to the technology” - Steve Jobs Customer experience © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 6. The digital landscape evolution Source - http://purplegoldfish.com/amazon-effect-rising-customer-expectations/ Customers’ expectations are rising at a dramatic pace. They expect a personalized, quick and consistent experience. • 76% of customers expect brands to understand their individual needs • 81% of customers demand improved response time • 68% anticipate organizations will harmonize customer experiences across channels Empowered customers are starting to take for granted that a company will know and understand their individual needs. They expect that companies will know what they searched for and their past ordering history. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 7. “© Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.” Verizon Big Data & Analytics The BDAI team focuses on developing advanced, external-facing data products that generate actionable insights on datasets through scalable machine learning These insights can lead to: • Data monetization • Cost mitigation • Enhanced customer experiences based on preferences
- 8. Verizon Big Data and AI platform Visualization APIs Network data Third-party data Extraction Loading Advertising Marketing Consumer lifestyle segmentation services Telematics analytics Transformation Data sources Big data and AI platform with privacy controls Applications IoT services Video analytics Cybersecurity Artificial Intelligence Large scale machine learning Deep learning Bots technology Natural language Automated reasoning Predictive algorithms BI and reporting Optimization Recommendation engine anomaly detection Domain-specific rules engine IoT Platform Fixed network Customer Wireless network IoT sensor Third party Public Telematics © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 9. “© Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.” Platform capabilities 60B Records processed per day by a single platform 20 PB 150TB 16000 Cores Easily scalable >65M Subscribers 32M Variables each 1M Records streamed per second 30+ Days of traffic ~5 sec To receive results x x = Source: Data points based on Verizon experience Features • Enterprise specific data and algorithm pipelines with batch and streaming services • Powerful suite of APIs to build custom solutions • Rich geo-spatial, temporal and comparative visualizations • Enterprise grade AI and ML at scale for advertising, marketing, IoT and other markets • Multi-tenancy and high availability • Ease of software development • Prebuilt ad network integration
- 10. “© Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.” Data Center
- 11. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. BDA client (HTTPS) API access (HTTPS) File send File streaming Kafka streaming infrastructure Spark stream processing Hadoop/ Yarn Solr | HDFS | Spark | HBase Raw data Compound data Monitoring Log indexing/alerting Firewall Firewall Firewall Secure VLAN Identity Manag- ement API BDA web- services Users REST APIs Stream/ file pull Portal Firewall Secure VLAN Data Ingestion Services PII data hashing Batch ETL Secure VLAN Secure VLAN Secure VLAN Software architecture Portal access (SSO) HAProxy Data analysis (IP filtering) Secure VLAN Privacy Controls
- 12. Secure and scalable pipeline Collect Secure Organize Categorize Learn • Standardized carrier-based data interfaces • Data ingestion • Multi-dimensional indexing • Near real-time model building • Predictive and machine learning algorithms • Data integration and indexing • Feature extraction • Traffic classification • Keyword extraction • Hashing • Parsing and cleansing Publish • Web platform • API © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 13. Verizon insights © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 14. Verizon insights © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 15. “© Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.” Combining measurement and prediction for >65M subscribers across 2.5M variables, such as: • TLDs • Apps • Keywords • ZIP codes Pushes compute time down from few minutes to a few seconds. Verizon insights
- 16. ● User self registration ● UserId and email verification ● Enforce consistent and standard password policies ● Self profile creation ● Multi-factor authentication ● Security questions ● One time passcode ● Secure and unsecure network verification ● Public and private computer User onboarding © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 17. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. Managing customers accounts with multiple applications Scaling the infrastructure Robust security Identity and access management API as a service Requirements Challenges High Availability
- 18. ● Standard based Authentication ● SAML, OAuth ● Single Sign On ● Use existing user credentials for new applications ● Onboard new users ● Federation Authentication ● Use Verizon SSO platform for authentication and authorization ● Authorization ● User Role and Permission Management ● Operational Analytics Security: Identity and Access Management © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 19. Aa Single sign-on and federated identity management Verizon SSO External ● Minimize integration ● Encrypted database ● Open source Trapezium Authentication and authorization framework Applications Customer insights IOT analytics Video analytics Location analytics Applications © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. IS analyticsVerizon SSO Internal
- 20. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. APIs as a service Analytics Databases REST APIs Insights Data Sources Authentication Security Subscription Rate limiting Elasticity Prediction API Time Series API Location API Keywords API Counts API Analytics API
- 21. API subscription - scaling infrastructure REST client Algorithms Trending APIs Prediction APIs Aggregation Location Web client Elastic Computing by API Subscription © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. Databases Memory Keyword search
- 22. “© Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.” API thresholding - consistency and reliability 20 PB 150TB 16000 Cores Easily scalable 32M Variables each 1M Records streamed per second x x = Source: Data points based on Verizon experience Rate limiting • Uniform usage in a multi-tenant setup • Fair allocation of the resources • Consistent SLA • Burst Control • Prevent Total Service Failure
- 23. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. High availability and microservices framework Customer 1 Customer 2 WSO2 Identity server WSO2 API gateway SSO Microservices 1 Microservices N • Active -- Active setup • Scale micro services on-demand
- 24. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. API analytics - platform infrastructure insights WSO2 API manager Microservices 1 Microservices N WSO2 API analytics Clients • Know your users • Know API usages by type, time • Know context -- API sequences • Know latency • Know capacity, API Quota • Detect Abnormalities, Failures
- 25. ● System Health Monitoring ● Abnormalities ● API latency ● API Failures ● Authentication Failures ● Fault Detection and Isolation ● Alerting and Notification ● Email ● Integration with Alerting systems Analytics : Operational © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 26. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. Identity management analytics dashboard
- 27. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. API latencyFault invocations API analytics
- 28. ● Robust security and privacy ● Peace of mind ● APIs ● Enables creativity, innovation and integration of multiple platforms ● Analytics and artificial intelligence ● Measure, gain and predict insights ● Mobile reach ● Increase customer reach with mobile networks Improving customer digital experience © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.
- 29. Digital transformation@Verizon © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. ● Wireless network ○ Highest in Wireless Network quality and performance ● Connected Devices ○ Internet of Things ( IOT) Platform ( thingspace.verizon.com) ● Next-generation connected vehicles ○ Verizon Telematics ( verizontelematics.com) ● Digital Media Services ○ oath.com ● Analytics, artificial intelligence and machine learning at scale
- 30. © Verizon 2017, All Rights Reserved. Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners. Thank You