The Future for Smart Technology Architects

The use, disclosure, reproduction, modification, transfer, or transmittal of this work without the written permission of IASA is strictly prohibited. © IASA 2010

Smart

Cloud and the

Internet of T

hings

A Day in the Life

IASA is

a non-profit professional association

run by architects

for all IT architects

centrally governed and locally run

technology and vendor agnostic

The use, disclosure, reproduction, modification, transfer, or transmittal of this work without the written permission of IASA is strictly prohibited. © IASA 2009

Topics

What is Cloud and IoT?

What is the relationship between Cloud and IoT?

Where is the ‘Smart’ in Smart Cloud and Smart IoT?

What is valuable about Cloud and IoT?

How to include smart system thinking into designs

How to get started with smart tools like inferencing, fuzzy, neural and other technologies

When to think smart and when to avoid

Possible outcomes to strive for today in preparing your architecture for the age of smart systems.

Cloud and IoT

’

Cloud

The umbrella term for anything available over a network

Relevant attributes which typify and classify architectures include Public or private Virtualized or non-virtualized Service oriented or person oriented Hardware oriented or platform oriented or software oriented Organizationally oriented or personally oriented Secure or unsecure Paid or free Paid by quality attribute or paid by operational attribute Guaranteed or unguaranteed

Internet of Things

Identifying all physical and virtual objects on a network

Relevant attributes which will typify and classify architectures may include Type of IoT identity (hardware, network, software, service,

invoker, agent, intelligent agent, independent intelligent agent, provocateur)

Size or scope of object (molecular -> planetary) Data type/volume consumption/production Power consumption/production Location and Mobility Object interaction power in virtual, physical or both Intention and Autonomy

Proposed Hierarchy of IoT Identities Provocateur - Intelligent agent with intention (human level)

Independent Intelligent Agent - Intelligent agent acting without permission

Intelligent Agent – Agent with a degree of reasoning capacity

Agent – Invoker which changes addresses in some way

Invoker – Service which calls other services

Service – Software object which returns a complex response

Software – Network object which returns a simple response

Network – An object which is addressable over a network

Hardware – An object which is identifiable over a network

Concepts and Relationships

Cloud is the raw network access mechanism

IoT is the type of things accessible

Understanding these relationships requires a much more sophisticated ontology and series of reference points

Google AcquiresDeep Mind

Why Is Smart Required for IoT and Cloud?

How is Smart Implemented Now

Advanced Search – Genetic, Graph Theory

Inferencing (Deductive, Inductive)

Fuzzy Reasoning

Optimization

Learning

Interpreting and Language

Negotiation

Searching for Information

Our lives and companies are run with information

Information has to be constructed from data and context

There is more data and information in the world than we can process

Intelligent search is key to our ability to make use of information

Common applications: business intelligence, lifestyle optimization, interest optimization

The Rules We Live By

Most companies have large numbers of commonly modified rules

Inferencing allows us to deduce new information within context (forward-chaining) induce information from existing data (backward-chaining)

Common Applications: Insurance rates and converage, retail pricing and discounts, purchase decisions, lifestyle choices “If the train is late let me sleep in”

Fuzzy Reasoning and Controllers

Humans and business work on ‘fuzzy definitions’ which is simply that most things are both true and not true “It is cold in Sweden” may be true to a Texan but not an

Eskimo! “A cup is also a bowl” can be more or less true “That hotel is extremely expensive” for me but Bill Gates?

Allows our devices to be more precise and selective in decision making and reasoning “Pre-heat the car when it is very cold” “We buy very high quality business supplies”

Common Applications: Energy utilization, mechanical controllers, human definitional input

Optimization

Business processes, graph navigation, optimal path traversal, and business integration all involve process optimizations

Multi-processes integration beyond the simplicity of a single service (physical or virtual) control much of our lives

Utilization of embedded process engines and optimization allows for maximum flexibility of physical and virtual agents

Common Applications: multi-partner business transactions, automated delivery systems, personal travel itineraries, multi-device automation

Learning

More and more data and choice is available to system software

As automation and autonomy become ubiquitous training in desired outcomes is necessary for personal and business

The vast amount of data and information requires grouping, characterizing and classifying

Neural networks and decision trees

Common applications: Food, travel and personal preferences, natural language processing, optimal energy input/output, security threat detection

Thing to Thing Communication

Language, dialect, grammar, vocabulary and pronunciation are all relevant in IoT communications and configuration

Knowledge and language ontology and dictionary will be essential to self-configuration (and therefore adoption)

This may be the single most difficult task in the IoT Even humans struggle with this constantly ‘Molecular’ data element combinations are not solidified (what is

an address, a name, a birthday)

Common applications: Thing configuration and communication, business analytics, service orchestration, personal identity management (pay for use)

Negotiation

As systems begin to represent us there is more and more conflict “What is the best price we can get for pencils for

employees”

Using negotiation techniques to avoid conflict with game theory

Common applications: Device resource allocation and utilization, purchasing

Considering Value and Risk

Value to Who? Individuals Governments and NGOs Vendors and Service

Integrators For Profit – non-vendor

What type of Value Lifestyle|Social Value Financial Value Customer|Operational

Value Societal|Human Value

Risk to Who? Individual Corporation Governments

What type of Risk? Physical Financial Societal

How Smart Becomes Value

There is a world of ‘new’ objects to sell to the world

There is an unlimited number of ways to incorporate new inventions into multiple channels, services and ‘products’ Learning about your customers and partners Dynamically allocating resources and processes Optimized pathing Planning and forecasting Configuration management and ease of use Human interaction and reasoning

Architecture Value

• Profitability• Constituent Value• Reuse• Grow Market Size• Grow Market Quality

What is “creates value”?What is Good?

suitable or efficient for a purposebeneficial or advantageous

Value Questions

Financial Value How do our customers buy from us? When does a person ‘have’ to be involved? How do our partners supply us? When do our customers have to think? When do our employees have to use a best guess or

experience? Are there times we ‘diagnose’ a problem? How can our systems interact on long-lasting complex

transactions?

What does Smart Mean Tomorrow

We must begin to consider systems as more than software services Autonomy – the degree to which systems can act without

permission Power (to influence) – the amount of influence or size of

outcomes a system can achieve Resources (to command and use) – the size and makeup of

objects a system may use Motivation – as systems gain more power and autonomy we

will need to understand Combat – when systems with autonomy, power and

resources disagree about outcomes

Resources

Books Designing the Internet of Things Practical Artificial Intelligence Programming with Java Rethinking the Internet of Things IoT – Global Technological and Societal Trends

Tools/Frameworks Drools Weka JFuzzyLogic Fuzzylite Gambit

Skill Taxonomy

The use, disclosure, reproduction, modification, transfer, or transmittal of this work without the written permission of IASA is strictly prohibited. © IASA 2009

Iasa ResearchUniversity

Partners

Grants and Awards

Questions