Breakout Sessions

1
Break-out Sessions
August 16th 21st, 2009

• Organizer Hans-Arno Jacobsen

Open Collaborate Wiki at http//www.debs.msrg.utor
onto.ca/collaborate
Based on a similar event at DEBS07 in Toronto.
2
Organization

• Meet in small groups during three slots
  throughout the week
• Adjourn in plenum on Friday to present discussion
  and results
• Each group is given some guiding questions, but
  may define their own
• Groups should be self-organizing, i.e., designate
  a scribe and a presenter

3
Loose Definition Scope

• All references to systems in the sequel are
  meant to refer to (non-exclusively)
• Publish/Subscribe systems (pub/sub)
• Event Processing systems (EP)
• Event Stream Processing systems
• Rule-based systems
• Distributed Event-based systems
• We are primarily targeting the scope of Events,
  Publish/Subscribe and Systems

4
Objectives

• Engage and actively participate
• Network and make new friends
• Create awareness for relevant research problems
• Collectively brainstorm on possible solutions
• Work towards a research agenda and road map
• Build on foundations for these discussions

5
Outcome

• The ideal outcome would be
• One coherent deck of slides per group that we
  post on the summer school web
• The tracking of group discussions in the
  Collaborate Wiki
• A common public report for the broader community

6
Suggested Discussion Groups

• Simulation evaluation
• Applications
• Research agenda road map
• Marketing advocacy
• Practical limitations
• Collaborations
• Interoperability
• Management issues

• Others?

7
Simulation Evaluation

• How can experiments be more systematic and
  comparable?
• Where can we obtain realistic workloads and data
  sets?
• What benchmarks exist today and should exist in
  the future?
• How have other communities developed and adopted
  benchmarks and data sets?
• What are realistic models for workload
  generation?
• What are good performance metrics?
• What is a solid evaluation methodology?

8
Applications

• What are the compelling applications, and how can
  we promote event-based application development?
• How can we bring up a world-wide pub/sub network
  or event processing system?
• What applications would drive the use,
  participation, and adoption of such a system?
• Can we run a generic event processing system
  playground across dedicated nodes?
• Who would/should the users be?
• Pub/Sub researchers, other researchers,
  application developers, end-users?

9
Applications cont.d

• What are killer applications for distributed
  event-based systems?
• How can event-based systems help build global
  sensor networks?
• How can small-scale actuator/sensor networks
  (e.g., in an e-home scenario) profit from
  pub/sub?
• What apps are the low hanging fruit for adoption?
• What applications today are naturally modeled
  with events, are easy to implement, are widely
  used, would benefit from events pub/sub and EP?

10
Research Agenda Road Map

• What are the interesting and relevant problems
  questions in event processing research?
• What are exciting research challenges that the
  community should look at?
• What research questions have been neglected in
  the past?
• Unification
• What important synergies with other areas need to
  be explored and developed?
• One-size-fits-all
• Federation, integration, and interoperation of
  systems designed for constrained environments
  (e.g. sensor networks) with systems designed
  without these constraints

11
Marketing Advocacy

• How can we better sell the event-based paradigm?
• What are the key selling points of pub/sub and
  EP?
• Decoupling, push based, asynchronous
  communication?
• What are the competing paradigms?
• Tuple spaces, (active) relational databases,
  streams, CQs?
• What are the difficulties in getting pub/sub EP
  accepted?
• Unfamiliarity with model
• No killer app
• Databases (pull model) can do everything
• No consistent terminology
• Why arent tuplespaces, active databases popular,
  or are they?

12
Practical Limitations

• What are the limitations of pub/sub and EP when
  it comes to implementing real applications?
• Lack of precise semantics
• Security
• Fault-tolerance
• Transactions (what does this mean in this
  context)
• Lack of an established query language model
• Overhead
• Due to decoupling
• Due to statelessness
• Hard to program/debug
• No persistent data (integration with databases)

13
Collaborations

• How can our community collaborate better within
  and outside itself?
• What are the pre-requisites for successful
  collaborations?
• At what level and scale can collaborations be
  carried forward
• Benchmarks, applications, systems?
• How can joint, collaborative research efforts be
  boot-strapped and lead?
• What sources of joint funding exist?

14
Collaborations cont.d

• How can we expand the community to include
  researchers from related fields?
• How do we relate to efforts in
• Databases, streams and continuous query
  processing
• Event processing complex event processing
• ECA rules and active databases
• How do we want to position ourselves as a
  community?

15
Management Issues

• How can we manage systems scalably?
• How can we deploy, configure and manage systems
  scalabely?
• How do we ensure that the end-to-end QoS issues
  are met?
• What about multidimensional QoS issues (e.g.,
  real-time fault tolerance security)?
• How can we manage QoS in heterogeneous
  environments?
• What are common manageable units and options in
  systems?
• Can a common standardized system be specialized
  (i.e., customized, optimized) by tools and made
  suitable for different use cases
• Are there common management patterns across
  different systems?
• Can we build management tools for systems that
  operate in a heterogeneous environment and still
  adhere to QoS?
• Are there use cases for reconfiguration and
  redeployment in systems?

16
Interoperability Standards

• How should standardization efforts proceed?
• What parts of the system should be subjected to
  standardization?
• Data formats, APIs, programming models?
• What should not be standardized?
• What concepts are most in flux, application
  dependent, etc.
• What standards exist today in this space?
• What are their limitations?
• What standards should exist?

17
Discussion in Break-out Sessions

• Strike break-out sessions and members
• Tasks for break-out sessions
• Designate one or preferably more scribes
• Discuss by roughly aiming to answer the questions
  asked
• Prepare a 10 minute summary on results of
  discussion
• Add final feedback to Wiki (scribes !!)

18
Break-out Session Rooms

• To be determined