Page 1
© ABB Group
Towards the Automation Cloud Architectural Challenges for a Novel Smart Ecosystem
Dr.-Ing. Heiko Koziolek, Principal Scientist, ABB Forschungszentrum, Ladenburg, 2013-07-02
http
://w
ww
.rcr
wire
less
.com
/am
eric
as/2
0110
816/
feat
ured
/clo
ud-c
ompu
ting-
is-in
-its-
infa
ncy-
gart
ner-
vp-s
ays/
Page 2
SCADA SystemsABB SCADAVantage for Oil & Gas Industries
© ABB Group
Page 3
SCADA SystemsABB SCADAVantage for Oil & Gas Industries
© ABB Group
Page 4
SCADA SystemsSystem Architecture
© ABB Group
Page 5
Automation PyramidClassical Structuring
© ABB Group
HMI PLC PLC HMI…
… ……
ERP
MES APCSCADA
Enterprise Resource Planning
Manufacturing Execution System
Supervisory Control and Data Aquisition
Advanced Process Control
Human-Machine Interface
Programmable Logic Controller
DataAcquisition Planning
Field devices: sensors, actuators...
Page 6
Cloud Computing
“Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”
5 essential characteristics
On-demand self-service
Broad network access
Resource pooling
Rapid elasticity or expansion
Measured service
NIST Definition
© ABB Group
[http://www.nist.gov/itl/cloud/]
Page 7
CloudificationClassical Automation Pyramid (AP)
© ABB Group
HMI PLC PLC HMI…
… ……
ERP
MES APCSCADA
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 8
Cloudification Step 1Few Cyber Physical Systems, Independent Clouds
HMI PLC PLC HMI…
… ……
© ABB Group
ERP
MES APCSCADA
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 9
HMI PLC PLC HMI…
… ……
PLC
ERP
MES APCSCADA
HMI PLC PLC HMI…
… ……
PLC
ERP
MES APCSCADA
Cloudification Step 2Few Cyber Physical Systems, Connected Cloud
© ABB Group
…
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 10
Cloudification Step 3Mainly Cyber Physical Systems on upper AP Levels
HMI PLC PLC HMI…
…
MES… …
ERP
……
PLC
© ABB Group
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 11
Cloudification Step 4Upper AP Levels partly replaced by Cloud Services
© ABB Group
…
HMI PLC PLC HMI…
…
MES…
ERP
……
PLC
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 12
Cloudification Step 5Cloud Services dominating upper AP levels
…HMI PLC PLC HMI…
…
MES…
ERP
……
PLC
© ABB Group
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 13
HMI PLC PLC HMI…
MES…
ERP
PLC…
… ……
Cloudification Step 6Cloud Services replacing upper AP levels
© ABB Group
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 14
Cloudification Step 7Automation Pyramid replaced by Automation Cloud
© ABB Group
…
…
……
Source: „Cyber-Physical Systems: Chancen und Nutzen aus Sicht der Automation“,
April 2013, GMA Fachausschuss 7.20 „Cyber Physical Systems“, Gesellschaft Mess- und Automatisierungstechnik im VDI/VDE
Page 15
Automation Cloud
Latency Security Multitenancy
Availability Elasticity Portability
Usability Interoperabili
ty
Selected Architectural Challenges
March 1, 2013 | Slide 15
© ABB Group
Page 16
© ABB Group
Criticality
Sampling time1 ms 10 ms 100 ms 1 s 10 s 1 d 30 d 1 y1 min 10 min 1 h 10 h
Real-time boiler control
Loop auditing
HVAC / buildings
Chemicals / refining MPC
?Service data
Network round-trip time for a cloud data center in Europe
(distance < 5000 km, 66 % speed of light)
Rolling millsCementMinerals
Power gen.
Solar panel / wind turbine
Low
High
Architectural ChallengesLatency
Page 17
© ABB Group
Architectural ChallengesLatency
• Characteristics: high latencies (seconds)• Well suited for cloud latency. Future: fleet
management, big data intelligence
SCADA, MES, PIMS, (ERP)
• Characteristics: latencies not critical• Today: VMs with engineering tools• Future: web-based, social design of
control applications
Engineering (e.g., device
configuration)
• Characteristics: high latencies, high computational power
• Conflicts with a typical cloud business model („low end“)
Advanced Process Control
(APC)
• Characteristics: short latencies, low computational power
• Impossible to run many control tasks remotely due to insufficient latencies
Basic Process Control
Page 18
ABB partnered with cloud provider GlobaLogix to provide a hosted version of SCADAVantage (SaaS)
RTUs triggering fast, basic control on-site
High latency SCADA functionality hosted in 53 data centers in North America, regional proximity
But: no horizontal scaling, no elasticity
Architecture ChallengesLatency
[http://www.abb.com/cawp/seitp202/cf46b46446b6f83985257b7a00488357.aspx]
Page 19
© ABB Group
Architecture ChallengesSecurity
Stuxnet
Duqu
Page 20
© ABB Group
GlobaLogix data centers hosting ABB‘s SCADAVantage
2048 bit encryption (exceeding DoD standards)
Compliance with the most stringent Tier 4 data center standards from the Telecommunications Industry Association (TIA) and American National Standards Institute (ANSI)
Citrix authentication on client laptops and tables
Password protected web access to read-only data
Architecture ChallengesSecurity
[http://www.abb.com/cawp/seitp202/cf46b46446b6f83985257b7a00488357.aspx]
March 1, 2013 | Slide 20
Page 21
Architecture ChallengesMultitenancy
© ABB Group
[http://goo.gl/FlrES/]
Page 22
Architecture Challenges
Embedded Systems
PLC, RTU, IED (for low-level, fast control)
Cloud-managed firmware updates
Information Systems
SCADA, MES, ERP, etc.
Cloud with comon data storage enables fleet management, social engineering, preventive maintenance
Mobile Systems
Mobile apps for maintenance personal and operators
APIs for third party extensions
Automation Cloud as Smart Ecosystem
© ABB Group
SCADAVantage on the iPhone
…
…
……
Page 23
Architecture Challenges
Cloud pattern catalogues Architecture decision sets, ontologies,
domain-specific patterns, …
Architecture description languages Cloud elements as first-class entities,
domain-specific abstractions, …
Architecture evaluation ATAM templates for cloud platforms Model-based predictions
Cloud benchmarks Reference workloads,
tooling, comparisons, …
Methods for Ultra-large Scale Systems Smart Grid & Automation Cloud as ULSS Systems of systems
Directions for Academic Research
© ABB Group
[Koziolek, Proc. WICSA’11]
Page 24
Automation Pyramid affected by cloud computing capabilities
Architecture Challenges
What control task can be executed remotely?
How to secure a SCADA system in the Internet?
How to effectively apply multi-tenancy?
Future: emergence of a novel smart ecosystem
ConclusionsTowards the Automation Cloud
© ABB Group
…
…
……
…
…
……