DEVOPS
IN
IOT
WHO AM I
• Solutions Architect
• DevOps
• Serverless
• IoT
• Big Data
• Blockchain
• AI
• ML
dlavrin@softserveinc.com
AGENDA
1. IoT Services
2. DevOps Implementation as a Service
3. IoT Projects Landscape
4. Programmable Edge
5. IoT (Hybrid) Infrastructure
6. Event Flow
7. Data Processing on IoT Edge
8. Enterprise IoT
9. Enterprise Hybrid Cloud
10. IoT Edge Auto-Discovery
11. Visibility
12. Real-Life examples
IOT SERVICES
SOLUTIONS ARCHITECTURE
AND DEVELOPMENT
in Hybrid-Cloud environment
BUILD CONTINUOUS DEVOPS
EXPERIENCE
across IoT Edge & Clouds
QA/QC PROCESS SETUP AND
AUTOMATION
in Micro-Services environment
MICROSERVICES ADOPTION
For IoT
PROVIDING ESSENTIAL
SECURITY
for customer’s solutions, threat detection,
visibility and control
OPEN SOURCE COMPONENTS
CUSTOMIZATIONS
to fit requirements
• IoT & Hybrid Infrastructure
• Securing components
• Facilitates the growths of the IoT
DEVOPS IMPLEMENTATION AS A SERVICE
Business
• KPI’s, CSF’s, how they align with
Hybrid Cloud
• Allign current business process
with HC strategy
Discovery
• Define use cases
• FR & NFR ( like capacity,
availability,security, ops, scale,
assurance )
Architecture Design
• Define context
• Allign high-level Architecture with
HC principles
• Document Architecture decisions
• Ops Model design
• Components considering
Roadmap
• Define timeline, phases, milestones
• Deliverables aligning
• Define scope
• Define inclusions and exclusions
PROGRAMMABLE
EDGE
IOT PROJECT LANDSCAPE
HYBRID
CLOUDS/HYBRID
INFRASTRUCTURE
IOT &
SERVERLESS
DATA
PROCESSING
SECURITY
• IoT Edge
• SP Edge
• Identity management
• Data processing and
anomaly detection
PROGRAMMABLE EDGE
What for
•Performing data processing at the edge
•Reducing the communications bandwidth
•Analytics and knowledge generation
•Machine Learning predictions.
•Data Reduction & Data Aggregation
•Manage Time Critical Workload via AI
Value
•Cost Saving
•Efficiency Resource Utilization
•Faster Data Processing
IOT HYBRID INFRASTRUCTURE
• Cloud & On-Premise
• Data Aggregation
• Data Reduction
• Event Processing
• Etc.
• IoT Edge
• Data Processing on the fly
• Predictive analytics
• Programmable Edge
• Event Driven Flow
• IoT Devices
Fn
Fn
Fn
Fn
SDK
EVENT FLOW
• Ability to define «workflow»
• Ability to define «build blocks» – tasks
• Ability to create different scenarios
• Ability to assign triggers
• React to events
• Manage workload
DATA PROCESSING ON IOT EDGE
• Continuous data loading
• Massively parallel processing
• Data consolidation
• Dimensional processing
• Data normalization &
denormalization (depends on
tech stack)
• Structured & dimensional data
models
• Hybrid distributed
warehouses
• Data WH
• Processing
• Analytics
• Visualization
• Machine Learning
• Data Virtualization
• Data Ingestion
• Identity
ENTERPRISE IOT
• Security
• AuthN/AuthZ
• RBAC
• Policies
• Assurance
• Monitoring
• Telemetry
• Logging
• Tracing
• Orchestration
• Service Identity
• Intelligence
ENTERPRISE HYBRID CLOUD
OVERVIEW
TIGHT INTEGRATION WITH
DIFFERENT VENDORS
to expose key features like programmability,
security, and scale across all resource types
PLATFORMS MANAGEMENT
AUTOMATION
(deployment, updates, control)
SERVICE CATALOG ASSURANCE
USAGE METERING POLICIES
INFRASTRUCTURE CAPACITY
REPORTING
SELF-HEALING PIPELINE
(analytics, visibility, tracing, real-time control )
SECURITY
IOT EDGE DISCOVERY
• API-Gateways
• Load Balancers
• Traffic Management Services
• High Availability management
• Canary deployments
• Access control
• Geo-distribution
• Auto-Discovery
VISIBILITY
• Open Tracing (tracing in general )
• Enterprise solutions
• Telemetry
• Monitoring
• Logging
IOT: WAREHOUSE MICROCLIMATE
REDUCTION IN RISKS OF
FOODS’
DETERIORATION
CONTROLLING
MICROCLIMATE IN THE
WHOLE SUPPLY CHAIN
FAST DETECTION
OF SYSTEM
BREAKDOWNS
OIL WELLEarly prevention system to detect oil production anomalies in real time
9,000+
DISTRIBUTED
ACOUSTIC SENSORS
(DAS)
1,200,000 events per min
30GB per min
10,000+
DISTRIBUTED TEMPERATURE
SENSORS (DTS)
30Gb per day
• Solution architecture
• Reference Architecture
• Real-Time Analytics
• Retrospective Analytics
• Capture and transform data on the fly
• Apply rules to data in motion
• Securely deliver data
• Distributed Big Data stack
SMART BACK WALLRevenue growth and cost-saving solution for a B2B customer in retail
INCREASED PRODUCT
AVAILABILITY ON THE
SHELVES
COST SAVING (FOR
EXTRA DELIVERIES)
TOBACCO BACK WALLS
MEET REQUIREMENTS SET
BY CIGARETTE VENDORS
• Special camera installation to recognize empty slot tags
by SKU in the tobacco back wall and to control
equipment technical conditions;
• The software platform to collect and process data,
transfer them to server and online reports;
• Different views and drill-down options for different users
inside the company;
• The system of notifications being sent to the proper
employees depending on the problem occurred.
R&D: SMART COFFEE MACHINE
• IoT + Blockchain
• Biosensors
• Apple Watch
IOT DATA BEACON TRACKER
IOT AND ACOUSTIC DATA PROCESSING
Reference & solution architecture
Speed Layer
(Real-Time Analytics)
Storage Layer
Sensors
Streaming Layer Processing Layer
Visualization
Data Collector
Message
Queue
Data
Pipeline
Batch Layer
(Retrospective Analytics)
IoT
Data
Real-Time Stream
Processing
Real-Time
Data Storage
Real-Time Data
Visualization
Batch
Processing
Historical Data
Storage
Retrospective Data
Visualization
SOFTSERVE AND NATIONAL
UNIVERSITY LVIV POLYTECHNICS
Program features
• 23 SoftServe employees took part in program
development
• 70% - renovated courses
• SoftServe IT specialists are involved as instructors
• SoftServe representatives supervise and consult final
bachelor program execution
BACHELOR’S PROGRAM
KHARKIV DEVOPS COMMUNITY
& UKROPS
https://www.facebook.com/KharkivDevOpsCommunity/
https://www.facebook.com/KyivDevOpsCommunity/
http://ukrops.club
FOR
THE
FUTURE

DevOps in IoT

  • 1.
  • 2.
    WHO AM I •Solutions Architect • DevOps • Serverless • IoT • Big Data • Blockchain • AI • ML [email protected]
  • 3.
    AGENDA 1. IoT Services 2.DevOps Implementation as a Service 3. IoT Projects Landscape 4. Programmable Edge 5. IoT (Hybrid) Infrastructure 6. Event Flow 7. Data Processing on IoT Edge 8. Enterprise IoT 9. Enterprise Hybrid Cloud 10. IoT Edge Auto-Discovery 11. Visibility 12. Real-Life examples
  • 4.
    IOT SERVICES SOLUTIONS ARCHITECTURE ANDDEVELOPMENT in Hybrid-Cloud environment BUILD CONTINUOUS DEVOPS EXPERIENCE across IoT Edge & Clouds QA/QC PROCESS SETUP AND AUTOMATION in Micro-Services environment MICROSERVICES ADOPTION For IoT PROVIDING ESSENTIAL SECURITY for customer’s solutions, threat detection, visibility and control OPEN SOURCE COMPONENTS CUSTOMIZATIONS to fit requirements • IoT & Hybrid Infrastructure • Securing components • Facilitates the growths of the IoT
  • 5.
    DEVOPS IMPLEMENTATION ASA SERVICE Business • KPI’s, CSF’s, how they align with Hybrid Cloud • Allign current business process with HC strategy Discovery • Define use cases • FR & NFR ( like capacity, availability,security, ops, scale, assurance ) Architecture Design • Define context • Allign high-level Architecture with HC principles • Document Architecture decisions • Ops Model design • Components considering Roadmap • Define timeline, phases, milestones • Deliverables aligning • Define scope • Define inclusions and exclusions
  • 6.
    PROGRAMMABLE EDGE IOT PROJECT LANDSCAPE HYBRID CLOUDS/HYBRID INFRASTRUCTURE IOT& SERVERLESS DATA PROCESSING SECURITY • IoT Edge • SP Edge • Identity management • Data processing and anomaly detection
  • 7.
    PROGRAMMABLE EDGE What for •Performingdata processing at the edge •Reducing the communications bandwidth •Analytics and knowledge generation •Machine Learning predictions. •Data Reduction & Data Aggregation •Manage Time Critical Workload via AI Value •Cost Saving •Efficiency Resource Utilization •Faster Data Processing
  • 8.
    IOT HYBRID INFRASTRUCTURE •Cloud & On-Premise • Data Aggregation • Data Reduction • Event Processing • Etc. • IoT Edge • Data Processing on the fly • Predictive analytics • Programmable Edge • Event Driven Flow • IoT Devices Fn Fn Fn Fn SDK
  • 9.
    EVENT FLOW • Abilityto define «workflow» • Ability to define «build blocks» – tasks • Ability to create different scenarios • Ability to assign triggers • React to events • Manage workload
  • 10.
    DATA PROCESSING ONIOT EDGE • Continuous data loading • Massively parallel processing • Data consolidation • Dimensional processing • Data normalization & denormalization (depends on tech stack) • Structured & dimensional data models • Hybrid distributed warehouses • Data WH • Processing • Analytics • Visualization • Machine Learning • Data Virtualization • Data Ingestion • Identity
  • 11.
    ENTERPRISE IOT • Security •AuthN/AuthZ • RBAC • Policies • Assurance • Monitoring • Telemetry • Logging • Tracing • Orchestration • Service Identity • Intelligence
  • 12.
    ENTERPRISE HYBRID CLOUD OVERVIEW TIGHTINTEGRATION WITH DIFFERENT VENDORS to expose key features like programmability, security, and scale across all resource types PLATFORMS MANAGEMENT AUTOMATION (deployment, updates, control) SERVICE CATALOG ASSURANCE USAGE METERING POLICIES INFRASTRUCTURE CAPACITY REPORTING SELF-HEALING PIPELINE (analytics, visibility, tracing, real-time control ) SECURITY
  • 13.
    IOT EDGE DISCOVERY •API-Gateways • Load Balancers • Traffic Management Services • High Availability management • Canary deployments • Access control • Geo-distribution • Auto-Discovery
  • 14.
    VISIBILITY • Open Tracing(tracing in general ) • Enterprise solutions • Telemetry • Monitoring • Logging
  • 15.
    IOT: WAREHOUSE MICROCLIMATE REDUCTIONIN RISKS OF FOODS’ DETERIORATION CONTROLLING MICROCLIMATE IN THE WHOLE SUPPLY CHAIN FAST DETECTION OF SYSTEM BREAKDOWNS
  • 16.
    OIL WELLEarly preventionsystem to detect oil production anomalies in real time 9,000+ DISTRIBUTED ACOUSTIC SENSORS (DAS) 1,200,000 events per min 30GB per min 10,000+ DISTRIBUTED TEMPERATURE SENSORS (DTS) 30Gb per day • Solution architecture • Reference Architecture • Real-Time Analytics • Retrospective Analytics • Capture and transform data on the fly • Apply rules to data in motion • Securely deliver data • Distributed Big Data stack
  • 17.
    SMART BACK WALLRevenuegrowth and cost-saving solution for a B2B customer in retail INCREASED PRODUCT AVAILABILITY ON THE SHELVES COST SAVING (FOR EXTRA DELIVERIES) TOBACCO BACK WALLS MEET REQUIREMENTS SET BY CIGARETTE VENDORS • Special camera installation to recognize empty slot tags by SKU in the tobacco back wall and to control equipment technical conditions; • The software platform to collect and process data, transfer them to server and online reports; • Different views and drill-down options for different users inside the company; • The system of notifications being sent to the proper employees depending on the problem occurred.
  • 18.
    R&D: SMART COFFEEMACHINE • IoT + Blockchain • Biosensors • Apple Watch
  • 19.
  • 20.
    IOT AND ACOUSTICDATA PROCESSING Reference & solution architecture Speed Layer (Real-Time Analytics) Storage Layer Sensors Streaming Layer Processing Layer Visualization Data Collector Message Queue Data Pipeline Batch Layer (Retrospective Analytics) IoT Data Real-Time Stream Processing Real-Time Data Storage Real-Time Data Visualization Batch Processing Historical Data Storage Retrospective Data Visualization
  • 21.
    SOFTSERVE AND NATIONAL UNIVERSITYLVIV POLYTECHNICS Program features • 23 SoftServe employees took part in program development • 70% - renovated courses • SoftServe IT specialists are involved as instructors • SoftServe representatives supervise and consult final bachelor program execution BACHELOR’S PROGRAM
  • 22.
    KHARKIV DEVOPS COMMUNITY &UKROPS https://www.facebook.com/KharkivDevOpsCommunity/ https://www.facebook.com/KyivDevOpsCommunity/ http://ukrops.club
  • 23.