An architectural style for solving computationally intensive problems on large networks
by Yuriy Brun, Nenad Medvidovic
Abstract:
Large networks, such as the Internet, pose an ideal medium for solving computationally intensive problems, such as NP-complete problems, yet no well-scaling architecture for computational Internet-sized systems exists. We propose a software architectural style for large networks, based on a formal mathematical study of crystal growth that will exhibit properties of (1) discreetness (nodes on the network cannot learn the algorithm or input of the computation), (2) fault-tolerance (malicious, faulty, and unstable nodes may not break the computation), and (3) scalability (communication among the nodes does not increase with network or problem size).
Citation:
Yuriy Brun and Nenad Medvidovic, An architectural style for solving computationally intensive problems on large networks, in Proceedings of Software Engineering for Adaptive and Self-Managing Systems (SEAMS), 2007.
Bibtex:
@inproceedings{Brun07seams,
  author = {Yuriy Brun and Nenad Medvidovic},
  title = {\href{http://people.cs.umass.edu/brun/pubs/pubs/Brun07seams.pdf}{An
  architectural style for solving computationally intensive problems on large
  networks}},
  booktitle = {Proceedings of Software Engineering for Adaptive and
  Self-Managing Systems (SEAMS)},
  venue = {SEAMS},
  month = {May},
  date = {26--27},
  year = {2007},
  address = {Minneapolis, {MN}, {USA}},
  doi = {10.1109/SEAMS.2007.4},
  accept = {$\frac{18}{26} \approx 69\%$},

  note = {\href{http://dx.doi.org/10.1109/SEAMS.2007.4}{DOI:
  10.1109/SEAMS.2007.4}},

  abstract = {Large networks, such as the Internet, pose an ideal medium for
  solving computationally intensive problems, such as NP-complete problems, yet
  no well-scaling architecture for computational Internet-sized systems exists.
  We propose a software architectural style for large networks, based on a
  formal mathematical study of crystal growth that will exhibit properties of
  (1) discreetness (nodes on the network cannot learn the algorithm or input of
  the computation), (2) fault-tolerance (malicious, faulty, and unstable nodes
  may not break the computation), and (3) scalability (communication among the
  nodes does not increase with network or problem size).},

  fundedBy = {NSF ITR-0312780},
}