U. Stern and D.L. Dill. Parallelizing the Mur' veri er. In Proc. Computer Aided Veri cation, volume 1254 of LNCS, pages 256-267. Springer, 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....in automatic veri cation tools. As a matter of fact, in application of such tools to practical veri cation problems the computational power available (memory and time) is the main limiting factor. Recently, some attempts to use multiprocessors and networks of workstations have been undertaken. In [UD97] the authors described a parallel version of the veri er Mur . In their approach, the table of all reached states is partitioned over the nodes of the parallel machine, which allows the table to be larger than on a single node. The explicit state enumeration is then performed in parallel. For the ....

U.Stern and D. L. Dill. Parallelizing the mur' verier. In O. Grumberg, editor, Proceedings of Computer Aided Verication (CAV '97), volume 1254 of LNCS, pages 256-267, Berlin, Germany, 1997. Springer.

....t into the main memory of a single machine. Several Spin specic partition functions are experimented, the most advantageous one being a function that takes into account only a fraction of the state vector. Another distributed state enumeration algorithm has been implemented in the Mur verier [SD97] The speedups obtained are close to linear and the hash function used for state space partition provides a good load balancing. However, experimental data reported concerns relatively small state spaces (approximatively 1.5 M states) on a 32 node UltraSparc Myrinet network of workstations. ....

....have shown that in this case there is almost no termination detection overhead, two waves being always suOEcient. This distributed termination detection scheme seems to use less messages than the centralized termination detection schemes used in the parallel versions of Spin [LS99] and Mur [SD97] which in all cases require several broadcast message exchanges between a coordinator machine and all other machines. 3.2 Merging of partitioned LTSs into monolithic LTSs After constructing a collection of N Bcg les representing a partitioned Lts by using the Distributor algorithm, the next ....

[Article contains additional citation context not shown here]

U. Stern and D. Dill. Parallelizing the Mur' Verier. In Computer Aided Verication, volume 1254, pages 256267. Lecture Notes in Computer Science, Springer-Verlag,

....states. Examples of the memory saving approach are, e.g. in [6, 38, 14, 15, 27, 28, 10] In an auxiliary storage approach, one tries to exploit disk storage as well as distributed processors (network storage) to enlarge the available memory (and CPU) Examples of this approach are, e.g. in [25, 26, 21, 30, 22]. In this paper we explore the possibility of trading space with time in order to reduce the amount of memory needed to complete a given veri cation task. 2 More speci cally, our goal here is to devise a Breadth First State Space Exploration (BFS) algorithm with graceful degradation. In ....

U. Stern and D. Dill. Parallelizing the mur' verier. In Proc. 9th Int. Conference on: Computer Aided Veri- cation, volume 1254, pages 256-267, Haifa, Israel, 1997. LNCS, Springer.

....problem [9] Several sequential methods (partial order reductions, on the y search) to overcome this barrier have been proposed and successfully implemented in automatic veri cation tools. Recently, some attempts to use multiprocessors and networks of workstations have been undertaken. In [23] the authors describe a parallel version of the veri er Mur . The table of all reached states is partitioned over the nodes of the parallel machine and the explicit state enumeration is performed in parallel. A similar approach to distributed reachability analysis has been taken in [18] A ....

U. Stern and D.L. Dill. Parallelizing the Mur' verier. In Proc. CAV

....over its range and hence implements what is called random load balancing. Since states are equally likely to be mapped to any node, all nodes will receive approximately the same number of states and hence the load will be equally distributed. This approach is very similar to the one taken by [SD97] The di erence is that Uppaal uses symbolic states, each covering (in nitely) many concrete states. In order to achieve optimal performance, the lookup performed on the Passed list is actually an inclusion check. An unexplored symbolic state taken from the Waiting list is compared with all the ....

U. Stern and D. L. Dill. Parallelizing the Mur' verier. In Orna Grumberg, editor, Proc. of the 9th Int. Conf. on Computer Aided Verication, volume 1254 of Lecture Notes in Computer Science, pages 256-267. Springer{ Verlag, June 1997. Haifa, Isreal, June 22-25. p. 6

....problem [9] Several sequential methods (partial order reductions, on the y search) to overcome this barrier have been proposed and successfully implemented in automatic veri cation tools. Recently, some attempts to use multiprocessors and networks of workstations have been undertaken. In [23] the authors describe a parallel version of the veri er Mur . The table of all reached states is partitioned over the nodes of the parallel machine and the explicit state enumeration is performed in parallel. A similar approach to distributed reachability analysis has been taken in [18] A ....

U. Stern and D.L. Dill. Parallelizing the Mur' verier. In Proc. CAV

....over its range and hence implements what is called random load balancing. Since states are equally likely to be mapped to any node, all nodes will receive approximately the same number of states and hence the load will be equally distributed. This approach is very similar to the one taken by [SD97] The di erence is that Uppaal uses symbolic states, each covering (in nitely) many concrete states. In order to achieve optimal performance, the lookup performed on the Passed list is an inclusion check. An unexplored symbolic state taken from the Waiting list is compared with all the explored ....

Ulrich Stern and David L. Dill. Parallelizing the Mur' Verier. In Orna Grumberg, editor, Proc. of the 9th Int. Conf. on Computer Aided Verication, volume 1254 of Lecture Notes in Computer Science, pages 256-267. Springer{Verlag, June 1997. Haifa, Isreal, June 22-25.

....TOSCA Corresponding Author: Enrico Tronci. Tel: 39 0862 433129. Fax: 39 0862 433180. 2 In an auxiliary storage approach one tries to exploit disk storage as well as distributed processors (network storage) to enlarge the available memory (and CPU) Examples of this approach are, e.g. in [16, 17, 13, 21, 14]. In this paper we study the possibility of exploiting statistical properties of protocol transition graphs to improve state exploration algorithms. This is quite similar to what is usually done when optimizing a CPU on the basis of program pro ling [12] Our algorithm allows us to reduce the ....

U. Stern and D. Dill. Parallelizing the mur' verier. In Proc. 9th Int. Conference on Computer Aided Verication, volume 1254, pages 256-267, Haifa, Israel, 1997. LNCS, Springer.

....over its range and hence implements what is called random load balancing. Since states are equally likely to be mapped to any node, all nodes will receive approximately the same number of states and hence the load will be equally distributed. This approach is very similar to the one taken by [SD97] The di erence is that Uppaal uses symbolic states, each covering (in nitely) many concrete states. In order to achieve optimal performance, the lookup performed on the Passed list is actually an inclusion check. An unexplored symbolic state taken from the Waiting list is compared with all the ....

U. Stern and D. L. Dill. Parallelizing the Mur' verier. In Orna Grumberg, editor, Proc. of the 9th Int. Conf. on Computer Aided Verication, volume 1254 of Lecture Notes in Computer Science, pages 256-267. Springer{Verlag, June 1997. Haifa, Isreal, June 22-25.

....further computers within a network as a substitute for external storage. The algorithms described in [30, 6] divide the underlying problem into several tasks. However, they are designed in the way that only a single computer can be employed to sequentially handle one task at a time. Stern and Dill [34] show how to carry out a parallel reachability analysis. The distribution of the underlying structure is similar to the one presented here. But their algorithm is not appropriate for model checking temporal logic formulas. 18] presents a parallel reachability analysis algorithm for BDDs. They ....

....known a priori but computed while constructing the graph. 8 Hence, the partition of the whole game graph into blocks cannot be determined in advance. We propose the following way to construct and store the graph which is inspired by the work pool presentation of [25] and is similarly applied in [34]. First, we employ adjacency lists instead of matrices. Note that we need also links to the predecessors as well as to the successors of a node for the labelling algorithm. Let f be a function mapping the states of the game graph to a processor of our network. Usually, one takes a function in the ....

U. Stern and D. L. Dill. Parallelizing the Mur' verier. In O. Grumberg, editor, Computer Aided Verication, 9th International Conference, volume 1254 of Lecture Notes in Computer Science, pages 256-267. Springer-Verlag, June 1997. Haifa, Israel, June 22-25.

....further computers within a network as a substitute for external storage. The algorithms described in [19, 2] divide the underlying problem into several tasks. However, they are designed in the way that only a single computer can be employed to sequentially handle one task at a time. Stern and Dill [23] show how to carry out a parallel reachability analysis. The distribution of the underlying struc ture is similar to the one presented here. But their algorithm is not suitable for model checking temporal logic formulae. 9] presents a parallel reachability analysis algorithm for BDDs. They argue ....

....is unknown a priori but computed while constructing the graph. 3 Hence, the partition of the game graph into blocks cannot be determined in advance. We propose the following way to construct and store the graph which is inspired by the work pool presentation of [15] and is similarly applied in [23]. Let f be a function mapping the states of the game graph to a processor of our network. Usually, one takes a function in the spirit of a hash function assigning to every state an integer and subsequently its value modulo the number of processors. Then, f determines the location of every state ....

U. Stern and D. L. Dill. Parallelizing the Mur' verier. In Proc. 9th Int. Conf. Computer Aided Verication, volume 1254 of LNCS, Springer, 1997.

.... have been developed that address this problem: symbolic methods such as BDDs, methods that exploit symmetry, partial order reduction techniques, etc [4] One obvious approach that has been applied successfully by a number of researchers is to parallelize (or distribute) the state space search [1, 15]. Distributed reachability analysis and state space generation has also been investigated in the related eld of performance analysis in the context of stochastic Petri nets [3, 8] see the second paper for further references) Since the stateof the art in model checking and performance analysis ....

....Rather, the goal should be to view parallelization as an orthogonal feature, which can always be easily added when the appropriate hardware is available. Research supported by Esprit Project 26270, Veri cation of Hybrid Systems (VHS) To some extend this goal has been achieved in the work of [3, 15, 8], all with very similar solutions. Stern and Dill [15] for example, present a simple but elegant approach to parallelize the Mur tool [5] using the message passing paradigm. In parallel Mur , the state table, which stores all reached protocol states, is partitioned over the nodes of the parallel ....

[Article contains additional citation context not shown here]

U. Stern and D. L. Dill. Parallelizing the Mur' verier. In Orna Grumberg, editor, Computer Aided Verication, 9th International Conference, volume 1254 of LNCS, pages 256-67. Springer-Verlag, June 1997. Haifa, Isreal, June 22-25.

No context found.

U. Stern and D.L. Dill. Parallelizing the Mur' veri er. In Proc. Computer Aided Veri cation, volume 1254 of LNCS, pages 256-267. Springer, 1997.

No context found.

U. Stern and D.L. Dill. Parallelizing the Mur' veri er. In Proc. CAV, volume 1254 of LNCS, pages 256-267. Springer, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC