Mobile and Ubiquitous Computing A background and overview - PowerPoint PPT Presentation

1 / 34
About This Presentation
Title:

Mobile and Ubiquitous Computing A background and overview

Description:

... dialog box pops up on Jane's screen suggesting that she go to ... Client may guide the application to reduce fidelity. Use reservation based Quality of Service ... – PowerPoint PPT presentation

Number of Views:62
Avg rating:3.0/5.0
Slides: 35
Provided by: hkha
Category:

less

Transcript and Presenter's Notes

Title: Mobile and Ubiquitous Computing A background and overview


1
Mobile and Ubiquitous Computing A background and
overview
  • Vision of Mark Weiser
  • Challenges and Issues of Mobile Computing
  • Challenges and Issues of Ubiquitous Computing

2
Reference Papers
  • 1 The Computer for the 21st Century
  • Mark Weiser, Palo Alto Research Center
  • 1991, Scientific American
  • Reprinted in Pervasive Computing
  • January March 2002
  • 2 Fundamental Challenges in Mobile Computing
  • M. Satyanarayanan, Carnegie Mellon University
  • Published in 1996 in ACM (PODC96)
  • 3 Pervasive Computing Vision and Challenges
  • M. Satyanarayanan, Carnegie Mellon University
  • Published in IEEE Personal
    Communications August
    2001

3
Preview
  • The convergence of technologies
  • Issues pointed to by Mark Weiser
  • Problems and issues of Mobile Computing
  • Five major research areas
  • Problems of Ubiquitous Computing
  • Integration of different technologies
  • Major areas of research

4
  • An Interesting Scenario
  • Jane is at Gate 23 at the Pittsburgh airport,
    waiting for her connecting flight
  • She would like to use her wireless connection to
    e-mail a number of large documents
  • Many passengers at Gates 22 and 23 are surfing
    the Web
  • Aura observes that Jane wont be able to finish
    sending her documents before her flight departs
  • After consulting the airports flight schedule
    service, Aura discovers that wireless bandwidth
    is excellent at Gate 15
  • A dialog box pops up on Janes screen suggesting
    that she go to Gate 15, which is only three
    minutes away
  • It also asks her to prioritize her e-mail, so
    that the most critical messages are transmitted
    first
  • Jane accepts Auras advice and walks to Gate 15
  • Aura informs her that it is close to being done
    with her messages, and that she can start walking
    back
  • The last message is transmitted during her walk,
    and she is back at Gate 23 in time for her
    boarding call 3

5
Vision of Mark Weiser
  • Specialized elements of hardware and
    software, connected by wires, radio waves and
    infrared, will be so ubiquitous that no one will
    notice their presence
  • Ubiquitous
  • Invisible computers embedded in everyday objects
    that would replace PCs
  • Technology should disappear to make computers
    ubiquitous
  • Issues
  • Scale and the knowledge of where the
    user is located to adapt to the surroundings

6
Vision of Mark Weiser
  • Embodied virtuality
  • Modified virtual reality
  • Process of drawing computers out of their
    electronic shells
  • Devices that implement the technology
  • Interaction between devices is more important
    than making them
  • Parts that build a ubiquitous system
  • Low cost and low power computers
  • Software
  • Network that ties the computer together

7
Vision of Mark Weiser
  • The Computer Hardware is satisfactory
  • The Operating System needs a change Microkernel
    approach
  • Network requirements
  • Transparent and wireless link between modules
  • Protocols that are efficient in mobile networks
  • A large number of channels
  • A future where computers disappear

8
Problems and issues of mobile computing
  • Constraints of mobile computing and their impact
    on distributed systems
  • All constraints are intrinsic to mobility
  • Mobile elements are resource poor
  • Mobility is inherently hazardous
  • Mobile connectivity is highly variable
  • Mobile elements rely on a finite energy source
  • Solution lies in the mobile client being
  • ADAPTIVE

9
Problems and issues of mobile computing
  • Adaptive implementation has two extremes
  • Total autonomy Adaptation is the responsibility
    of the individual application
  • Application transparent adaptation
    Responsibility of adaptation lies entirely
    on the system
  • Actual implementation lies between the two
    extremes Application aware adaptation
  • Use an extended client-server model to implement
    mobile application
  • The client may have to take the role of a server

10
Problems and issues of mobile computing
  • Experimental results on Coda file system
  • Experiment conducted to study the effect of
    mobility and application transparent adaptation
  • Disconnected operation is feasible, effective and
    usable
  • Use techniques of hoarding, update logging and
    reintegration on reconnection
  • Optimistic replica control strategy is applicable
  • Use techniques of log-based directory
    resolution, application-specific file
    resolution, containment and manual repair

11
Problems and issues of mobile computing
  • Experimental results on Coda file system.
  • 3. Support weak connectivity to avoid the
    limitations of disconnected operations
  • Use techniques of adaptive transport
    protocols, rapid cache validation mechanism,
    trickle reintegration mechanism and model-based
    cache miss handling
  • 4. Use isolation-only transaction to cope with
    detection and handling of read-write conflicts
    during disconnected operations
  • Use database transactions
  • 5. Server replication can be used to complement
    disconnected operation

12
Areas of research in mobile computing
  • Five areas of research are identified
  • Caching Metrics
  • Problems of common metric - Miss ratio
  • Should be consistent and easy to monitor
  • Open problems
  • What is an appropriate set of caching metrics for
    mobile computing?
  • Under what circumstances does one use each
    metric?
  • How does one efficiently monitor these metrics?
  • What are the implications of these alternative
    metrics for caching algorithms?

13
Areas of research in mobile computing
  • 2. Use cache validation instead of maintaining
    cache coherence to reduce remote communication
  • Use callbacks
  • Maintain cache coherence at multiple levels of
    granularity
  • Use semantic validation to maintain cache
    coherence at multiple levels of granularity
  • Open questions are related to predicate Q that is
    a function of the block value P
  • How useful are semantic validation and callbacks?
  • What forms can P and Q take for data types?

14
Areas of research in mobile computing
  • 3. Algorithms for resource revocation
  • Application aware adaptation complicates the
    problem of resource management
  • Open problems are
  • How does one formulate the resource revocation
    problem?
  • How does one characterize the differential impact
    of revocation on different applications?
  • What strategies does one use if multiple
    resources must be simultaneously revoked?
  • How does one distinguish between resources whose
    revocation is easy to recover from and those
    which are expensive or impossible to recover
    from?
  • How does one handle deadlocks during revocation

15
Areas of research in mobile computing
  • 4. Analysis of adaptation
  • Primary figure of merit for mobile clients
    Agility
  • Ideal mobile client is highly agile and very
    stable
  • Open questions
  • What are the right metrics to measure agility?
  • Are there systematic techniques to improve the
    agility of a system?
  • How does one decide when a mobile system is
    agile enough?
  • What are the right metrics to measure system
    stability?
  • Can one develop design guidelines to ensure
    stability?
  • Can one analytically derive the agility and
    stability properties of an adaptive system
    without building it?

16
Areas of research in mobile computing
  • 5. Global estimation from local observations
  • Adaptive mobile clients should have the
    capability to infer global changes by detecting
    local changes
  • Issues are
  • Are there systematic ways to perform global
    estimation from local observations?
  • Can one bind the error in global estimates?
  • What is the relationship between global
    estimation and agility of adaptation?
  • Can one provide system support to improve global
    estimation?
  • Can one quantify the benefits of out-of-band
    channels?

17
Problems and Issues of Ubiquitous Computing
  • Pervasive Computing
  • Technology that disappears
  • Relationship with distributed systems and mobile
    computing
  • Problems solved in distributed systems
  • Remote communication
  • Fault tolerance
  • High availability
  • Remote information access
  • Security
  • Key constraints of mobility forced the
    development of specialized techniques
  • Four areas of research in ubiquitous computing

18
How Did We Get Here?
Smart spaces Invisibility Localized
scalability Uneven conditioning
19
Problems and Issues of Ubiquitous Computing
  • Areas of research
  • Effective use of smart spaces context-aware
    space
  • Invisibility minimal user distraction
  • Localized scalability
  • Masking uneven conditioning smartness of
    environments vary
  • First requirement for any application is
    proactivity
  • All technologies for implementing the scenario is
    available
  • Whole is much greater than simply the sum of its
    parts
  • Requires seamless integration of component
    technologies
  • Problems lie in the architecture, component
    synthesis and system level engineering

20
Problems and Issues of Ubiquitous Computing
  • User
  • Immersed in a PC environment
  • Adaptable and reconfigurable
  • Client
  • May take the role of a server
  • A multilayered architecture
  • Environment
  • Smart space
  • Server

21
The structure of an Aura Client
22
Problems and Issues of Ubiquitous Computing
  • Problems that need to be solved
  • User intent needs to be established
  • A number of research questions need to be
    answered
  • Deal with the conflicting requirement of higher
    functionality and limited resources
  • Use available resources in the smart space
  • Cyber foraging and surrogate of mobile computing
  • Cyber foraging living off the land exploit
    wired hardware infrastructure
  • Hardware starts playing the role of surrogate of
    the mobile computer

23
Problems and Issues of Ubiquitous Computing
  • Areas to explore
  • Cyber foraging
  • Adaptation strategies
  • High level energy management
  • Client thickness
  • Context awareness
  • Balancing proactivity and transparency
  • Privacy and trust
  • Impact on layering

24
Problems and Issues of Ubiquitous Computing
  • Cyber foraging gives rise to a number of research
    problems
  • Discover the presence of surrogates
  • Level of trust with a surrogate
  • Load balancing
  • Time to stage
  • Effect of scalability
  • System support needed for seamless and minimally
    intrusive surrogate

25
Problems and Issues of Ubiquitous Computing
  • Adaptation strategies
  • Client may guide the application to reduce
    fidelity
  • Use reservation based Quality of Service
  • Client may suggest the user a corrective action
  • Unanswered questions
  • Factors that affect the decision to choose
    between the adaptation strategies
  • Is reservation based QoS the best strategy?
  • Multiple and competing requests for reservations
  • Practicality of adaptation using corrective
    actions
  • Methods of lowering fidelity

26
Problems and Issues of Ubiquitous Computing
  • High level energy management
  • High level system should get involved for energy
    management
  • Energy aware memory management
  • Unanswered questions
  • Methods of managing energy
  • Impact on invisibility
  • Can knowledge of user intent be exploited in
    energy management?
  • Use of smart spaces and surrogates to reduce
    energy demand
  • Remote execution to extend battery life

27
Problems and Issues of Ubiquitous Computing
  • Client thickness
  • The clients need to be as thin as possible but
    should be thick enough to fulfill the
    requirements
  • Research questions are
  • Quantify client thickness and environmental
    conditioning
  • Can the system alert a user when it migrates to a
    less hospitable environment?
  • Is transparent migration from a thinner to a
    thicker client possible?
  • Can computers be reconfigured to serve as optimal
    mobile clients under diverse environmental
    conditions?
  • Can semi-portable infrastructure be carried with
    a user to augment less hospitable environments?

28
Problems and Issues of Ubiquitous Computing
  • Context awareness
  • Required for a minimally intrusive computing
    system
  • Issues
  • How is context represented internally?
  • How frequently does context information need to
    be consulted?
  • Minimal services that the environment needs to
    provide
  • What are the relative merits of various
    location-sensing technologies?

29
Problems and Issues of Ubiquitous Computing
  • Balancing proactivity and transparency
  • Both are important requirements
  • Proactive systems may not always work well
  • Transparency is especially important in
    distributed systems with scarce resources
  • Unanswered questions
  • How are individual user preferences and
    tolerances specified and taken into account?
  • How is it possible to determine the balance?
  • Can the existing balancing mechanisms be used to
    provide a systematic design guideline to
    application designers?

30
Problems and Issues of Ubiquitous Computing
  • Privacy and trust
  • System contains detailed information related to a
    user
  • User trust is achieved by keeping the information
    strictly confidential and private
  • Research issues
  • Balance between seamless system behavior and the
    need to alert users to potential loss of privacy
  • What are the authentication techniques best
    suited to pervasive computing?
  • How is access control enforced using the identity
    of users?

31
Problems and Issues of Ubiquitous Computing
  • Impact on layering
  • Layering techniques help solve large problems
  • Helps hiding some details but exposes others that
    are required to solve a problem
  • Issues are
  • Preserving the benefits of layering while
    accommodating the needs of ubiquitous computing
  • Are existing layers best extended for pervasive
    computing by broadening their primary interfaces
    or creating secondary interfaces?
  • What is the complexity involved in design and
    implementation of layers for ubiquitous computing?

32
Areas explored recent publications
  • Research areas (2005-2006)
  • Internet suspend/resume (ISR)
  • VM technology and Distributed File systems
  • Same hardware architecture at least to ISA level
  • Remote execution
  • Cyber foraging
  • Resource monitoring and predicting resource
    demands

33
Conclusion
  • Challenging research problems exist in the field
    of Mobile and Ubiquitous Computing
  • Integration of technologies is the main issue
  • HCI, Software agents, AI are other areas that are
    integrated in the field
  • A lot has been achieved
  • More to be explored

34
  • Thank you
Write a Comment
User Comments (0)
About PowerShow.com