• Groups 1 and 2:
  1. Consider two distributed snapshot algorithms such as Samadi's and Mattern's algorithms. Discuss their pros and cons.
  2. Solve textbook Q14.14
• Groups 3 and 4:
  1. Compare Timewarp and SPEEDS in terms of performance, process creation/termination, dynamic memory allocation, and I/O handling.
  2. Solve textbook Q14.15

• Groups 1 and 2:
  1. Compare Ivy and Dash in terms of considtency models, shared-data granularity, HW/SW implementation, false sharing, and implementation. What types of applications can Ivy and Dash benefit respectively?
  2. Solve slide p24's non-turn-in exercise 1.
• Groups 3 and 4:
  1. Compare distributed shared memory and message passing (such as MPI) in terms of programmability and performance. Discuss about their pros and cons using two types of applications: computer graphics such as 3D ray tracing and spatial simulation such as molecular dynamics.
  2. Solve slide p24's non-turn-in exercise 2.

• Groups 1 and 2:
  Compare NFS and AFS in terms of file-accessing models, file-sharing semantics, modification propagation (write through or delayed write), server-side/client-side caching, and client/server-initiated validation
• Groups 3 and 4:
  Suppose that a user wants to use NSF or AFS for grid-computing middleware where s/he runs a massively parallel application with a large number of remote computers. Each process running at a different computer needs to share data files local to the user's computer. Discuss which of NSF and AFS will work out better for grid computing.

• Groups 1 and 2:
  1. Could the gossip architecture be used for a distributed computer game as described below? The players move figures around a common scene. The state of the game is replicated at the players' workstations and at a server, which contains services controlling the game overall, such as collision detection. Updates are multicast to all replicas. (Textbook Q18.11)
  2. Suppose that a user wants to use Gossip for grid-computing middleware where s/he runs a massively parallel application with a large number of remote computers. Each process running at a different computer has a postive sequential identifier, periodically takes an execution snapshot, monitors its logical neighbor with a one-larger identifier, and resumes this neighbor at a new computer if it has crashed for some reason. Discuss how Gossip faciliates this fault tolerance in grid computing.
• Groups 3 and 4:
  1. The quorum-based replication protocol can address network partition problems. Why didn't Code use this protocol? Explain the reason.
  2. Suppose that a user wants to use Coda for grid-computing middleware where s/he runs a massively parallel application with a large number of remote (network-detachable) computers. Each process running at a different computer has a postive sequential identifier, periodically takes an execution snapshot, monitors its logical neighbor with a one-larger identifier, and resumes this neighbor at a new computer if it has crashed for some reason. Discuss how Coda faciliates this fault tolerance in grid computing.

• Groups 1 and 2: Summarize how Condor, Legion, NetSolve, and Globus facilitate the following two features:
  1. Discoveries of remote computing resources
  2. Job deployment to remote computers
• Groups 3 and 4: Summarize how Condor, Legion, NetSolve, and Globus facilitate the following two features:
  1. Fault tolerance, (i.e., a job resumption at a new site)
  2. I/O, (i.e., rerouting files and standard I/O data to remote jobs)

NetSolve

1. http://icl.cs.utk.edu/netsolve/
2. Henri Casanova, Jack Dongarra, Chris Johnson, and Michelle Miller, "Section 7.3: Case Study: NetSolve", In Ian Foster and Carl Kesselman, editors, The Grid: Blueprint for a New Computing Infrastracture, Morgan Kaufmann Publishers, July 1998, pages 171-175 (available from the instructor)

Legion

1. http://legion.virginia.edu/
2. Dennies Gannon and Andrew Gimshaw, "Section 9.4: The Legion Grid Architecture", In Ian Foster and Carl Kesselman, editors, The Grid: Blueprint for a New Computing Infrastracture, Morgan Kaufmann Publishers, July 1998, pages 222-227 (available from the instructor)

Condor

1. http://www.cs.wisc.edu/condor
2. Douglas Thain, Todd Tannenbaum, and Miron Livny, "Condor and the Grid", in Fran Berman, Anthony J.G. Hey, Geoffrey Fox, editors, Grid Computing: Making The Global Infrastructure a Reality, John Wiley, 2003. ISBN: 0-470-85319-0

Globus

1. http://www.globus.org/
2. Ian Foster and Carl Kesselman, "Chapter 11: The Globus Toolkit", In Ian Foster and Carl Kesselman, editors, The Grid: Blueprint for a New Computing Infrastracture, Morgan Kaufmann Publishers, July 1998, pages 222-227 (available from the instructor)