azure paas and saas - assetsprod.microsoft.com · #iotinactionms azure paas and saas microsoft’s...
TRANSCRIPT
#IoTinActionMS
Azure PaaSand SaaSMicrosoft’s two approaches to building IoT solutions
Hector Garcia TelladoProgram Manager Lead,
Azure IoT Suite
#IoTinActionMS
Customers using IoT today
Microsoft SaaS IoT offering
Microsoft PaaS IoT offerings
IoT Analytics – What's new!
Agenda
#IoTinActionMS
Improving processes, efficiency, and human decision making
with predictive maintenance and tool-level data
Predictive maintenance brings Sandvik to the cutting edge of digital manufacturing
#IoTinActionMS
Norwegian developers at Kongsberg Maritime map unpredictable harbor floor with IoT Hub
Increasing ROI by determining optimal shipping
loads & improving navigation safety
“I had no previous experience with Microsoft. I knew almost
nothing about Azure, the cloud, or IoT. It only took a day or
two to get into it, after which it wasn’t that hard.”
Terje Nilsen, Manager of Disruptive Technology
Options to deploy cloud IoT solutions
Microsoft IoT CentralQuickly create solutions in
a managed environment
Azure IoT SuiteCustomize to your needs with full control
Remote Monitoring | Predictive Maintence | Connected Factory
SaaSAzure IoT Hub
Azure Stream Analytics
Azure Time Series Insights
Azure Machine Learning
Azure Logic Apps
MorePaaS
No cloud development
expertise required
Fully hosted and
managed by Microsoft
Risk-free trial with
simplified pricing
Reducing the complexity of IoT through managed services
Microsoft IoT Central
Microsoft IoT Central
Microsoft
IoT Central
Risk-free trial with simplified pricing
Analytics, dashboards and visualization
User roles and permissions
Monitoring rules and triggered actions
Device connectivity and management
Time-series Insights
Analytics & dashboards
Device management
Alerts and actions
Template Management
Rules Workflows
Device settings
Product Modeler
Builder Operator
Azure IoT Suite
Combining IoT Services Into An Extensible IoT Solution
Easily build a PoC
Your solution,
in minutes
Customize, extend
and scale
Azure IoT Suite Preconfigured Solutions Features
PaaS
Azure
IoT Suite
Data Ingestion and Command & Control
Stream Processing & Predictive Analytics
Workflow Automation and Integration
Dashboards and Visualization
.NET & Java
Open-sourced, microservices-based
architecture
• Device management, dashboards, commands
• Rules and actions, backend integration
• Add your devices and begin tailor to your needs
Partners accelerate time to value
Start quickly for
common IoT scenarios
• Customize to your assets and rules
• Highly visual for your real-time operational data
• Integrate with back-end systems
Finish with your IoT
application
Components of a preconfigured solution
Microservices
VM
Devices
Back end
systems and
processes
Cosmos DB
Web App
Logic
AppsIoT Hub
Simulator
Active
Directory
Orchestrator
Microservices
VM
Microservices
VM
Microservices
VM
Remote monitoring | Predictive maintenance | Connected factory | Device Simulation
Azure ML
Azure IoT Edge
Enabling the Intelligent Edge to achieve more
Configure, update and monitor from the cloud
Compatible with popular operating systems
Code symmetry between cloud and edge for easy
development and testing
Secure solution from chipset to cloud
Build once, deploy anywhere
Seamless deployment of AI and
advanced analytics
Azure IoT Edge
Azure
IoT
Devi
ces
Azure IoT Edge Runtime
Azure IoT Edge Architecture
Bridging cloud and devices to provide a cohesive end to end IoT solution
Security Multiplexing Store and forward
Modules (Container)
Managing leaf
devices
Azure Machine
Learning
Cognitive
ServicesAzure Functions Custom Code
Azure Stream
Analytics
#IoTinActionMS
What's new in IoT Analytics?
Krishna MamidipakaSenior Program Manager,
Azure Big Data
#IoTinActionMS
Companies that invest in IoT & data analytics
operating margin (18% vs. 10%)technology spend of revenue
Sources: Keystone Research
Unlocking Insights with Real-time analytics
Insights are Perishable
Window of opportunity for insights to be actionable
Time to Insight is Critical
Reducing decision latency can unlock business value
Query data still while it is still in motion
Can’t wait for data to get to rest before running computation
#IoTinActionMS
Mission critical
reliability
Lowest
TCO
Fully
managed
Ease of getting
started
Programmer
Productivity
Declarative SQL
like language
Source/sink
integrations
No cluster
provisioning
Pay as
you goEnterprise
grade SLA
Azure Stream Analytics
IoT Hubs
Archiving for long term storage/ batch analytics
Real-time dashboard
Stream Analytics
Automation to kick-off workflows
Machine LearningReference Data
Event Hubs
Blobs
Devices & Gateways
Presentation & Action
Storage &Batch Analysis
StreamAnalytics
Event Queuing & Stream Ingestion
Event production
Applications
#IoTinActionMS
Making buildings smarter
Benefits ▪ Greener Buildings ▪ Comfortable occupants
“The queries we need to run are quite complicated.... We are able to do this much quicker with Azure Stream Analytics, and with very low overhead.”
- Arvind Shetty, Technology Specialist
#IoTinActionMS
Enabling better business outcomes
“Plant equipment heat and vibration readings are passed along to asset management teams to ensure our equipment is being maintained correctly. Production output can be tracked and provided to our regulator to ensure compliance, and our commercial teams use this telemetry for billing purposes.”
- Kent Weare, Lead Architect
Benefits ▪ Lower equipment failures and downtime ▪ Secure infrastructure ▪ Lower operational costs
Inline Anomaly Detection
Pre-trained ML model
Easily called within our SQL-like query language
Can configure the size of the history window, used to compute martingale values over the look-back history
Simple usage to detect anomalies over one hour of time series data
select id, val, ANOMALYDETECTION(val) OVER(LIMIT DURATION(hour, 1)) FROM input
Usage with partitioning
select id, val, ANOMALYDETECTION(val) OVER(PARTITION BY id LIMIT DURATION(hour, 1)) FROM input
Usage with partitioning and "when"
select id, val, ANOMALYDETECTION(val) OVER(PARTITION BY id LIMIT DURATION(hour, 1) WHEN id != 2) FROM input
Usage showing the extraction of scores:
select id, val FROM input WHERE (GetRecordPropertyValue(ANOMALYDETECTION(val) OVER(LIMIT DURATION(hour, 1)), 'BiLevelChangeScore')) < -1.0
Azure IoT Edge
Azure
IoT
Devi
ces
Azure IoT Edge Runtime
Azure IoT Edge ArchitectureBridging cloud and devices to provide a cohesive end to end IoT solution
Security Multiplexing Store and forward
Modules (Container)
Managing leaf
devices
Azure Machine
Learning
Cognitive
ServicesAzure Functions Custom Code
Azure Stream
Analytics
Analytics closer to devices is key for many IoT scenarios
Ultra low-latency needs
Intermittent connectivity
Bandwidth economics
Compliance requirements
Same language for both
Cloud and Edge “jobs”
Azure Time Series Insights
Oil & Gas
Manufacturing
Smart Energy
Time-series data heavy apps
Smart BuildingInteractive
Analytics
Indexing and
Scalable storage
Visualization
and APIs
Fully-integrated
time series data pipeline
Data parsing and
metadata enrichment
Power & Utility