The document discusses two key challenges for pervasive computing systems: capturing user intent and enabling cyber foraging. For user intent, the document questions how systems can infer a user's goals in order to better support adaptation and proactivity. For cyber foraging, it proposes augmenting mobile devices' resources by exploiting nearby wired infrastructure through techniques like offloading computations or caching data. Several open research questions are also raised regarding discovering available surrogates, establishing trust, handling load balancing, and supporting seamless surrogate access.

1. 1
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (27)
Drilling Down - The Difficult Problems (4)
 4.1) User Intent
 Proactivity requires that a PERV system tracks user intent
 Otherwise, almost impossible to determine which system actions
help rather than hinder the user.
 E.g., suppose a user is viewing video over a network
connection whose bandwidth suddenly drops
 System choices:
 a) reduce the fidelity of the video,
 b) pause briefly to find another higher-bandwidth
connection, or
 c) terminate the task?
 Correct choice depends on what the user is trying to
accomplish
 Today’s systems are poor at capturing and exploiting user intent
 On the one hand: generic applications that have no idea what the
user is attempting to do
 Offer little support for adaptation and proactivity
 On the other hand: applications that try to anticipate user intent but
do so very badly
 Gimmicks like the Microsoft ‘‘paperclip’’ often more annoying than helpful

2. 2
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (28)
Drilling Down - The Difficult Problems (5)
 User intent - some important research questions:
 Can user intent be inferred, or does it have to be explicitly
specified?
 In the latter case, is it statically specified (e.g., from a file) or
obtained on demand through dynamic interactions?
 How is user intent represented internally? How is it used?
 How rich must this information be for it to be useful? When and
how is it updated? How do different layers of a system access this
knowledge?
 How does one characterize accuracy of user intent
knowledge?
 Is incomplete or imprecise knowledge of user intent still useful? At
what level of uncertainty is it better to ignore such knowledge in
making decisions?
 Will the attempt to obtain intent to burdensome on the
user?
 Will it hurt usability and performance unacceptably?
 Is the benefit [of using user intent] worth the cost? How does one
quantify this benefit?

4. 4
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (29)
Drilling Down - The Difficult Problems (6)
 4.2) Cyber Foraging
 Two contradictory requirements for mobile devices:
 Must be lightweight = small, light, with small and light battery
 Compromises computing capabilities
 Must have powerful comp/comm capabilities
 Ever-growing expectations of mobile users
 Well beyond capabilities of a lightweight mobile computer
 How to reconcile contradictory requirements?
 Difficult to reconcile
 Idea: use cyber foraging
 forage [fawr-ij] … 3. the act of searching for provisions of any kind.
[cf. Dictionary.com]
 Cyber foraging:
Dynamically augment the computing resources of a
wireless mobile computer by exploiting wired hardware
infrastructure [LTL: of the surrounding smart space]

5. 5
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (30)
Drilling Down - The Difficult Problems (7)
 4.2. Cyber Foraging – cont.1
 Need infrastructure facilitating cyber foraging
 Need smart public spaces
 Equipped with compute servers or data staging servers
 Much as table lamps [or power outlets] are today
 Connected to the wired Internet through high-bandwidth networks
 ‘‘Wasting’’ smart infrastructure when not in use
 We can afford this ‘‘waste’’ to improve user experience
 Could ensure that never wasted - always in use
 E.g., run SETI program / sell time slices to others
 Surrogate of a mobile computer:
Hardware in the wired infrastructure temporarily
supporting cyber foraging for the mobile computer
 [LTL:] Compare this limited cyber foraging with more
extensive foraging (by employing helpers) in opportunistic
networks (oppnets). After we discuss oppnets (soon), see
how oppnets’ approach differs.

6. 6
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (31)
Drilling Down - The Difficult Problems (8)
 4.2. Cyber Foraging – cont.2
 A typical foraging scenario
 Mobile computer (MC) enters a smart space
 MC detects surrogates and negotiates their use
 Communication with a surrogate via short-range wireless P2P (peer-
to-peer) technology
 Surrogate is MC’s networking gateway to the Internet
 When MC needs to perform an intensive computation accessing a
large volume of data, MC uses the surrogate
 Surrogate performs computations for MC
OR
 Surrogate services cache misses for MC
 Lower latency, avoiding Internet delays
 Surrogate caches data from the Internet on its local disk during
performing the computation
OR
 Surrogate may have staged data ahead of time in anticipation of the
user’s arrival in the neighborhood
 MC leaves the space
 Its surrogate bindings are broken
 Any data cached or staged for MC are discarded

7. 7
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (32)
Drilling Down - The Difficult Problems (9)
 Cyber foraging – some important research questions:
 How does one discover the presence of surrogates?
 Of the many proposed service discovery mechanisms such as
JINI, UPnP, and BlueTooth proximity detection, which is best
suited for this purpose? Can one build a discovery mechanism that
subsumes all of them for greatest flexibility?
 How to handle the issue of trust?
 How does one establish an appropriate level of trust in a
surrogate? What are useful levels of trust in practice? How
applicable and useful is the concept of caching trust [29]? Can
one amortize the cost of establishing trust across many surrogates
in a neighborhood?
 How is load balancing on surrogates done?
 Is surrogate allocation based on an admission control approach,
or a best-effort approach? How relevant is previous work on load
balancing on networks of workstations?

8. 8
© 2007 by Leszek T. Lilien
Based on: M. Satyanarayanan, “Pervasive Computing: Vision and Challenges,” IEEE Personal Communications, 2001
Pervasive Computing vs. Distributed Systems & Mobile Computing (33)
Drilling Down - The Difficult Problems (10)
 Cyber foraging – some important research questions – cont.
 How to handle the issue of staging?
 In typical situations, how much advance notice does a surrogate
need for effective staging with minimal delay?
 Is this on the order of seconds, minutes or tens of minutes?
What implications does this requirement have for the other
components of a pervasive computing system?
 What are the implications of cyber foraging for scalability
of smart fixed infrastructure?
 How dense does the fixed infrastructure have to be to avoid
overloads during periods of peak demand?
 What is the system support needed for seamless and
minimally intrusive surrogate experience for a user?
 Which are the components of this support that must be provided
by MC, and which by the smart space?