TY - GEN
T1 - Dealing with undependable workers in decentralized network supercomputing
AU - Davtyan, Seda
AU - Konwar, Kishori
AU - Russell, Alexander
AU - Shvartsman, Alexander
N1 - Funding Information:
This work is supported in part by the NSF Award 1017232 .
PY - 2013
Y1 - 2013
N2 - Internet supercomputing is an approach to solving partitionable, computation-intensive problems by harnessing the power of a vast number of interconnected computers. This paper presents a new algorithm for the problem of using network supercomputing to perform a large collection of independent tasks, while dealing with undependable processors. The adversary may cause the processors to return bogus results for tasks with certain probabilities, and may cause a subset F of the initial set of processors P to crash. The adversary is constrained in two ways. First, for the set of non-crashed processors P - F , the average probability of a processor returning a bogus result is inferior to 1/2 . Second, the adversary may crash a subset of processors F , provided the size of P - F is bounded from below. We consider two models: the first bounds the size of P - F by a fractional polynomial, the second bounds this size by a poly-logarithm. Both models yield adversaries that are much stronger than previously studied. Our randomized synchronous algorithm is formulated for n processors and t tasks, with n ≤ t, where depending on the number of crashes each live processor is able to terminate dynamically with the knowledge that the problem is solved with high probability. For the adversary constrained by a fractional polynomial, the time complexity of the algorithm is O (t/n ε log n log log n), its work is O (t log n log log n) and message complexity is O (n log n log log n). For the poly-log constrained adversary, the time complexity is O (n), work is O (t poly log n), and message complexity is O (n poly log n). All bounds are shown to hold with high probability.
AB - Internet supercomputing is an approach to solving partitionable, computation-intensive problems by harnessing the power of a vast number of interconnected computers. This paper presents a new algorithm for the problem of using network supercomputing to perform a large collection of independent tasks, while dealing with undependable processors. The adversary may cause the processors to return bogus results for tasks with certain probabilities, and may cause a subset F of the initial set of processors P to crash. The adversary is constrained in two ways. First, for the set of non-crashed processors P - F , the average probability of a processor returning a bogus result is inferior to 1/2 . Second, the adversary may crash a subset of processors F , provided the size of P - F is bounded from below. We consider two models: the first bounds the size of P - F by a fractional polynomial, the second bounds this size by a poly-logarithm. Both models yield adversaries that are much stronger than previously studied. Our randomized synchronous algorithm is formulated for n processors and t tasks, with n ≤ t, where depending on the number of crashes each live processor is able to terminate dynamically with the knowledge that the problem is solved with high probability. For the adversary constrained by a fractional polynomial, the time complexity of the algorithm is O (t/n ε log n log log n), its work is O (t log n log log n) and message complexity is O (n log n log log n). For the poly-log constrained adversary, the time complexity is O (n), work is O (t poly log n), and message complexity is O (n poly log n). All bounds are shown to hold with high probability.
UR - http://www.scopus.com/inward/record.url?scp=84892745768&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84892745768&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-35668-1_3
DO - 10.1007/978-3-642-35668-1_3
M3 - Conference contribution
AN - SCOPUS:84892745768
SN - 9783642356674
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 27
EP - 41
BT - Distributed Computing and Networking - 14th International Conference, ICDCN 2013, Proceedings
T2 - 14th International Conference on Distributed Computing and Networking, ICDCN 2013
Y2 - 3 January 2013 through 6 January 2013
ER -