IoT, 디바이스부터머신러닝까지 놓치지않을꺼에욧
Microsoft Data platform End-to-end 서비스 디바이스및센서데이터연결및처리데이터저장및성능분석프레젠테이션및활용 Internal only
Microsoft IoT platform End-to-end 서비스 디바이스및센서 데이터연결및처리데이터저장및성능분석프레젠테이션및활용 각종소형디바이스및센서 IoT 허브 ( 대량 Telemetry 수집 ) SQL Database (PaaS 형태의 SQL Server 서비스 ) 머신러닝 ( 고급분석및기계학습 ) Power BI ( 데이터분석 & 시각화플랫폼 ) 대형기계장치및그센서 Windows 10 IoT core Event Hub ( 대량이벤트수집 ) Table/Blob 저장소 ( 클라우드저장소 ) 스트림분석 ( 실시간분석 ) 서비스버스 ( 메시지연결 ) { } DocumentDB (NoSQL 문서 DB 서비스 ) HDInsight ( 클라우드상의 Hadoop 시스템 ) 데이터카탈로그 ( 데이터원본검색 ) 데이터레이크 ( 빅데이터를위한저장소및분석 ) Revoluton Analytics ( 고성능 R 통계분석 ) Data Factory ( 클라우드상의 ETL) SQL DW (DW 특화서비스 ) Internal only Redis Cache ( 클라우드앱을위한 Redis cache)
Devices Windows 10 IoT Core
Phone Phablet Small Tablet Large Tablet 2-in-1s (Tablet or Laptop) Classic Laptop Desktops & All-in-Ones Windows 10 Surface Hub Xbox Holographic IoT http://windows.microsoft.com
Adaptive User Interface Natural User Inputs One SDK + Tooling One Store + One Dev Center Reuse Existing Code One Universal Windows Platform http://windows.microsoft.com
https://msdn.microsoft.com/enus/library/windows/hardware/dn914597(v=vs.85).aspx
DEMO Client Programming (http://github.com /KoreaEva/IoT)
Platform Services Security & Ma nagement Portal Cloud Services Service Fabric Web Apps API Apps SQL Database Data Warehouse DocumentDB Hybrid Operations Azure AD Health Monitoring Azure Active Directory Azure AD B2C Batch RemoteApp Mobile Apps Logic Apps Redis Cache Azure Search Storage Tables AD Privileged Identity Management Domain Services Multi-Factor Authentication Automation Storage Queues BizTalk Services API Management Notification Hubs Backup Scheduler Hybrid Connections Service Bus HDInsight Machine Learning Stream Analytics Data Lake Operational Analytics Key Vault Visual Studio Azure SDK Data Factory Event Hubs Data Catalog Import/Export Store/ Marketplace VM Image Gallery & VM Depot Media Services Content Delivery Network (CDN) VS Online App Insights Infrastructure Services IoT Hub Mobile Engagement Azure Site Recovery StorSimple
Services IoT Hub Stream Analytics
일반적인 IoT 아키텍처 데이터의흐름 통신프로토콜 디바이스및센서 디바이스와저장소간의연결 데이터처리및저장 프레젠테이션및활용영역 MQTT, 웹소켓또는 Device Device Device 사용자프로토콜 AMQP HTTPS, MQTT, OPC Batch processing : Azure Data Factory, Azure HDInsight Hot Path Analytics : Azure Stream Analytics, Azure HDInsight Storm 데이터저장소및처리 프레젠테이션또는 비즈니스에연결 CoAP, AllJoyn, OPC Device Device AMQP HTTPS, MQTT, OPC IoT Hub 저상소및분석 앱서비스배포, 웹게시 CoAP, AllJoyn, OPC Device VPN LWM2M, HTTP, CoAP Real-time Analysis Data-At-Rest Analysis Machine Learning (R) 고급분석 CRM, ERP 등사내시스템과의연동및알림제공 Internal only
Serverless Architecture Cortana Cognitive Services Devices Management SQL Database Machine Learning Sensors on IoT Hubs Stream Analytics Power BI Windows 10 IoT Core Blob / Table
Device Explorer https://github.com/azure/azure-iot-sdks /tools/deviceexplorer Internal only
DEMO IoT Hub
DEMO SELECT * INTO OUTPUT FROM INPUT Real time analysis SQL Database Machine Learning Sensors on Windows 10 IoT Core IoT Hubs INPUT Stream Analytics QUERY Power BI OUTPUT Blob / Table
DEMO Stream Analytics Job
DEMO SELECT DeviceID, AVG(Temperature)as Temperature, AVG(Humidity) as Humidity, AVG(Dust) as Dust INTO OUTPUTS FROM INPUT GROUP BY DeviceID, TumblingWindow(Minute, 1) Real time analysis OUTPUTS SQL Database Machine Learning Sensors on Windows 10 IoT Core IoT Hubs INPUT Stream Analytics QUERY Power BI OUTPUT Blob / Table
DEMO SQL Server
클라우드데이터시각화및분석플랫폼 (Power BI) 프레젠테이션및활용 개요 Power BI 는클라우드상에서 SaaS 형태로서비스되는데이터시각화및분석플랫폼 특징 Office 제품, 특히 Excel 과의리포트연동기능 화려한대시보드구성을지원하는툴지원 (Power BI Desktop) 거의모든형식의데이터원본지원 데이터원본에의라이브연결지원 실시간대시보드지원 Oracle, IBM 과같은기타 DB 와도스케쥴링된데이터새로고침지원 대시보드및리포트, 데이터모델등을미리구성하여조직에배포할수있는콘텐츠팩기능지원 조직의콘텐츠팩뿐만아니라많이사용하는 SaaS 솔루션들이제공하는콘텐츠팩이용가능 데이터셋을통해리포트또는대시보드를만들수있으며기본제공되는차트이외에사용자가 D3.js 를통해개발한시각화요소 ( 그래프, 차트 ) 를생성가능 Andorid, ios 등에모바일 Native App 지원 KPI 관련알림기능, 코멘트기능등을포함 현재실시간연결대시보드연결은 Power BI REST API 또는 Azure Stream Analytics 를통해가능 라이브연결지원은 SQL Server Analysis Services 를통해지원 Internal only 한계 사용예시 기업의 KPI 모음대시보드 IT 관리대시보드 부서별매출분석대시보드 마케팅트위터 /Facebook 분석대시보드
DEMO Power BI
Common Classes of Algorithms (Supervised Unsupervised) Classification Clustering Regression Anomaly Detection
Known data Model Unknown data 1990 2000 2010 2020 50 F 30 F 68 F 95 F Weather forecast sample 48 F 29 F 70 F 98 F 49 F 27 F 67 F 96 F???? Using known data, develop a model to predict unknown data.
Model (Regression) 90 F 1990 2000 2010 50 F 30 F 68 F 95 F Predict 2020 Summer 48 F 29 F 70 F 98 F 49 F 27 F 67 F 96 F -26 F Using known data, develop a model to predict unknown data.
Azure Machine Learning Ecosystem Provision Workspace Build ML Model Deploy as Web Service Publish an App Get/Prepare Data Get Azure Subscription Create Workspace Evaluate Model Results Build/Edit Experiment Publish Web Service Azure Data Marketplace Create/Update Model
Data I/O Taking Data & preparing for Analysis Dimensionality reduction. E.g. Kinect measures 1000 points, 6 are relevant Fitting Model selection; calibration; assessment R free scripts/graphics, many packages based on Vector Data. Metrics to allow us to describe the data. E.g. Mean, Correlation Tools used for Text Input. E.g. What is the theme of this essay?
DEMO Azure ML