Christof Bornhoevd Brian Mo Thomas Odenwald Presenter: Asuman Suenbuel SAP Research Labs, Palo Alto Towards a Viable Sensor Network Industrial Ecosystem.

Christof Bornhoevd

Brian Mo

Thomas OdenwaldPresenter: Asuman Suenbuel

SAP Research Labs, Palo Alto

Towards a Viable Sensor Network

Industrial Ecosystem

SAP AG 2004, Title of Presentation / Speaker Name / 2

Five Key Elements of Enterprise Services Architecture

ANALYTICS/REPORTING

Bus.Partner

Legacy

PEOPLE PRODUCTIVITY

SAP NetWeaver

SERVICE COMPOSITION

Business Objects, Components, and Engines

PartnerServices

EnterpriseServices

RepositorySERVICE

ENABLEMENT

LIFE-CYCLELIFE-CYCLE MGMT

SAP’s Enterprise Services


SAP’s Blueprint for Sustainable Success

Business solutions for 27 industries

mySAP Business Suite, SAP xApps

SAP NetWeaver

Adaptive computing infrastructure

Enterprise Services Architecture

is a Web-services based architecture for adaptive business solutions

Smart Items Research Program – Vision

A holistic service-oriented architecture for the seamless integration of real world data and

events into enterprise software that efficiently exploits the capabilities of current, emerging and

future ubiquitous computing technologies like embedded systems and wireless sensor

networks


CRM PLMSRMSCM CRMPLMERPLeg-acy

InformationA2A

People

collaboration

B2B

roles

Devices

embeddedanalytics

alerts

roles

Real-Time Information for the Real-Time Enterprise

Media Breaks

PhysicalWorld

SmartItems

Sensor Network

EmbeddedSystemRFID

MobileSensors


How the Internet came about?

I I I I I

1960Foundation of the Internet,Sending/compression of data packages

Used NCP as transmission protocol1982 replaced by TCP/IP;First large-scale Internet was created as a set of interconnected US-Military computers,Services: email etc.

1979/1980USENET

1968/69 ARPANET

Spirit of freedom of speech, information sharing, introduction of PCs

1991WWW

standard linked information system accessible across the range of different computers in use.

1993the first proper web-browser, Mosaic

341,634% annual growth rate! Important sites like the White House and Pizza Hut appeared. Online shopping sites showed up.

1982TCP/IP

1984DNS 1995

Java


Development of the Internet

Technology basis

Prototype systems and applications

Definition of Standards

Widespread business use


What can we learn from this development for sensor networks?

So, the first large-scale Internet was created (1969 to 1982)— a set of

interconnected US military computers (e.g. for e-mail exchange).

Why did it take more than 25 years to revolutionize the world and

And what factors made the revolution finally happen?

The first applications stem from defense research projects with

government. Outside defense, the industrial applications are slowly

moving forward. Why slowly?


Revolution barriers of sensor networks

Lack of tools and testbets for application development

Lack of standards at software and hardware level: industry is reluctant to invest or to do development

a) for non standardized languages, tools, applications and operating systems.

b) in non standardized languages other than their own one

c) Lack of basic service standards (security, location, name service)

Sometimes nonproprietary nature: The risks for a software company is that nonproprietary products are in the public domain and anyone can produce or distribute them. This also means: there is no one to blame for if hardware level alterations need to be performed

Lack of hardware for reasonable prices & reasonable quality: Wireless nodes and sensors are still pricey, leaving the early implementation to those who can afford the initial outlay.


Revolution barriers of sensor networks

Lack of industrial application. The technology is too new to refer to

industrial success stories. Pioneering spirit required from initial

users/customers. Low return on investment in immediate future due to high

technology introduction costs. Lack of industry partners.

Lack of separation of business logic from device logic and network

logic no proper abstraction layers, too low level programming required

business application developers are not used to deal with low level

programming

Lack of consumer products as a result of reluctance of industry to adopt

the new technology reduced demands for sensor devices


Learn Lessons from the Internet History?

With sensor networks, we are in a similar situation like the internet in the early stages. In order to accelerate the technology maturation, we should learn from the history of the internet. Key factors for success for both internet and sensor networks are

standards,

tools, and

affordable hardware

marketable business applications and adaptions

Christof Bornhoevd Brian Mo Thomas Odenwald Presenter: Asuman Suenbuel SAP Research Labs, Palo Alto Towards a Viable Sensor Network Industrial Ecosystem.

Documents

sap ag

title of presentation

lack of hardware

wireless sensor networks

business logic

sensor devices

largescale internet

java slide