The Experts below are selected from a list of 12522 Experts worldwide ranked by ideXlab platform
Wlodzimierz Bielecki - One of the best experts on this subject based on the ideXlab platform.
-
Parallel tiled Nussinov RNA folding loop nest generated using both Dependence Graph transitive closure and loop skewing
BMC Bioinformatics, 2017Co-Authors: Marek Palkowski, Wlodzimierz BieleckiAbstract:Background RNA secondary structure prediction is a compute intensive task that lies at the core of several search algorithms in bioinformatics. Fortunately, the RNA folding approaches, such as the Nussinov base pair maximization, involve mathematical operations over affine control loops whose iteration space can be represented by the polyhedral model. Polyhedral compilation techniques have proven to be a powerful tool for optimization of dense array codes. However, classical affine loop nest transformations used with these techniques do not optimize effectively codes of dynamic programming of RNA structure predictions. Results The purpose of this paper is to present a novel approach allowing for generation of a parallel tiled Nussinov RNA loop nest exposing significantly higher performance than that of known related code. This effect is achieved due to improving code locality and calculation parallelization. In order to improve code locality, we apply our previously published technique of automatic loop nest tiling to all the three loops of the Nussinov loop nest. This approach first forms original rectangular 3D tiles and then corrects them to establish their validity by means of applying the transitive closure of a Dependence Graph. To produce parallel code, we apply the loop skewing technique to a tiled Nussinov loop nest. Conclusions The technique is implemented as a part of the publicly available polyhedral source-to-source TRACO compiler. Generated code was run on modern Intel multi-core processors and coprocessors. We present the speed-up factor of generated Nussinov RNA parallel code and demonstrate that it is considerably faster than related codes in which only the two outer loops of the Nussinov loop nest are tiled.
-
parallel tiled nussinov rna folding loop nest generated using both Dependence Graph transitive closure and loop skewing
BMC Bioinformatics, 2017Co-Authors: Marek Palkowski, Wlodzimierz BieleckiAbstract:RNA secondary structure prediction is a compute intensive task that lies at the core of several search algorithms in bioinformatics. Fortunately, the RNA folding approaches, such as the Nussinov base pair maximization, involve mathematical operations over affine control loops whose iteration space can be represented by the polyhedral model. Polyhedral compilation techniques have proven to be a powerful tool for optimization of dense array codes. However, classical affine loop nest transformations used with these techniques do not optimize effectively codes of dynamic programming of RNA structure predictions. The purpose of this paper is to present a novel approach allowing for generation of a parallel tiled Nussinov RNA loop nest exposing significantly higher performance than that of known related code. This effect is achieved due to improving code locality and calculation parallelization. In order to improve code locality, we apply our previously published technique of automatic loop nest tiling to all the three loops of the Nussinov loop nest. This approach first forms original rectangular 3D tiles and then corrects them to establish their validity by means of applying the transitive closure of a Dependence Graph. To produce parallel code, we apply the loop skewing technique to a tiled Nussinov loop nest. The technique is implemented as a part of the publicly available polyhedral source-to-source TRACO compiler. Generated code was run on modern Intel multi-core processors and coprocessors. We present the speed-up factor of generated Nussinov RNA parallel code and demonstrate that it is considerably faster than related codes in which only the two outer loops of the Nussinov loop nest are tiled.
-
ACS - Tile Merging Technique to Generate Valid Tiled Code by Means of the Transitive Closure of a Dependence Graph
Hard and Soft Computing for Artificial Intelligence Multimedia and Security, 2016Co-Authors: Wlodzimierz Bielecki, Piotr SkotnickiAbstract:The paper presents a novel approach for generation of parallel tiled code of arbitrarily nested loops whose original inter-tile Dependence Graphs contain cycles. We demonstrate that the problem of cyclic Dependences can be reduced to the problem of finding strongly connected components of an inter-tile Dependence Graph, and then solved by merging tiles within each component. Our technique is derived via a combination of the Polyhedral Model and Iteration Space Slicing frameworks. The effectiveness and efficiency of generated code is verified by means of well-known linear algebra kernels from the PolyBench benchmark suite.
-
perfectly nested loop tiling transformations based on the transitive closure of the program Dependence Graph
Applied Categorical Structures, 2015Co-Authors: Wlodzimierz Bielecki, Marek PalkowskiAbstract:A novel approach to producing tiled code for perfectly nested loops is presented. It is based on the transitive closure of the program Dependence Graph. The approach is derived via a combination of the Polyhedral and Iteration Space Slicing frameworks that allows us to enlarge the effectiveness of the tiling transformation. The results of the evaluation of the effectiveness of a presented algorithm and the efficiency of tiled codes produced by means of the algorithm are discussed.
Marek Palkowski - One of the best experts on this subject based on the ideXlab platform.
-
parallel tiled nussinov rna folding loop nest generated using both Dependence Graph transitive closure and loop skewing
BMC Bioinformatics, 2017Co-Authors: Marek Palkowski, Wlodzimierz BieleckiAbstract:RNA secondary structure prediction is a compute intensive task that lies at the core of several search algorithms in bioinformatics. Fortunately, the RNA folding approaches, such as the Nussinov base pair maximization, involve mathematical operations over affine control loops whose iteration space can be represented by the polyhedral model. Polyhedral compilation techniques have proven to be a powerful tool for optimization of dense array codes. However, classical affine loop nest transformations used with these techniques do not optimize effectively codes of dynamic programming of RNA structure predictions. The purpose of this paper is to present a novel approach allowing for generation of a parallel tiled Nussinov RNA loop nest exposing significantly higher performance than that of known related code. This effect is achieved due to improving code locality and calculation parallelization. In order to improve code locality, we apply our previously published technique of automatic loop nest tiling to all the three loops of the Nussinov loop nest. This approach first forms original rectangular 3D tiles and then corrects them to establish their validity by means of applying the transitive closure of a Dependence Graph. To produce parallel code, we apply the loop skewing technique to a tiled Nussinov loop nest. The technique is implemented as a part of the publicly available polyhedral source-to-source TRACO compiler. Generated code was run on modern Intel multi-core processors and coprocessors. We present the speed-up factor of generated Nussinov RNA parallel code and demonstrate that it is considerably faster than related codes in which only the two outer loops of the Nussinov loop nest are tiled.
-
Parallel tiled Nussinov RNA folding loop nest generated using both Dependence Graph transitive closure and loop skewing
BMC Bioinformatics, 2017Co-Authors: Marek Palkowski, Wlodzimierz BieleckiAbstract:Background RNA secondary structure prediction is a compute intensive task that lies at the core of several search algorithms in bioinformatics. Fortunately, the RNA folding approaches, such as the Nussinov base pair maximization, involve mathematical operations over affine control loops whose iteration space can be represented by the polyhedral model. Polyhedral compilation techniques have proven to be a powerful tool for optimization of dense array codes. However, classical affine loop nest transformations used with these techniques do not optimize effectively codes of dynamic programming of RNA structure predictions. Results The purpose of this paper is to present a novel approach allowing for generation of a parallel tiled Nussinov RNA loop nest exposing significantly higher performance than that of known related code. This effect is achieved due to improving code locality and calculation parallelization. In order to improve code locality, we apply our previously published technique of automatic loop nest tiling to all the three loops of the Nussinov loop nest. This approach first forms original rectangular 3D tiles and then corrects them to establish their validity by means of applying the transitive closure of a Dependence Graph. To produce parallel code, we apply the loop skewing technique to a tiled Nussinov loop nest. Conclusions The technique is implemented as a part of the publicly available polyhedral source-to-source TRACO compiler. Generated code was run on modern Intel multi-core processors and coprocessors. We present the speed-up factor of generated Nussinov RNA parallel code and demonstrate that it is considerably faster than related codes in which only the two outer loops of the Nussinov loop nest are tiled.
-
perfectly nested loop tiling transformations based on the transitive closure of the program Dependence Graph
Applied Categorical Structures, 2015Co-Authors: Wlodzimierz Bielecki, Marek PalkowskiAbstract:A novel approach to producing tiled code for perfectly nested loops is presented. It is based on the transitive closure of the program Dependence Graph. The approach is derived via a combination of the Polyhedral and Iteration Space Slicing frameworks that allows us to enlarge the effectiveness of the tiling transformation. The results of the evaluation of the effectiveness of a presented algorithm and the efficiency of tiled codes produced by means of the algorithm are discussed.
Daniel Khankin - One of the best experts on this subject based on the ideXlab platform.
-
Dependence Graph and master switch for seamless dependent routes replacement in sdn extended abstract
Network Computing and Applications, 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.
-
NCA - Dependence Graph and master switch for seamless dependent routes replacement in SDN (extended abstract)
2017 IEEE 16th International Symposium on Network Computing and Applications (NCA), 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.
Yefim Dinitz - One of the best experts on this subject based on the ideXlab platform.
-
Dependence Graph and master switch for seamless dependent routes replacement in sdn extended abstract
Network Computing and Applications, 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.
-
NCA - Dependence Graph and master switch for seamless dependent routes replacement in SDN (extended abstract)
2017 IEEE 16th International Symposium on Network Computing and Applications (NCA), 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.
Shlomi Dolev - One of the best experts on this subject based on the ideXlab platform.
-
Dependence Graph and master switch for seamless dependent routes replacement in sdn extended abstract
Network Computing and Applications, 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.
-
NCA - Dependence Graph and master switch for seamless dependent routes replacement in SDN (extended abstract)
2017 IEEE 16th International Symposium on Network Computing and Applications (NCA), 2017Co-Authors: Yefim Dinitz, Shlomi Dolev, Daniel KhankinAbstract:We study the problem of seamlessly updating several routes in a network, in the context of Software-Defined Networking (SDN). A set of routes pairs (C i , N i ) is given, where each new N i should replace the existing C i . We look for a way of gradual updating, so that routing cycles are never created during the replacement process. In that, we follow the recent paper of Delaet et al., which considered the case of updating a single route. In addition, we require avoiding congestion on links. We provide an example of several routes replacement, where the strategy suggested by Delaet et al. fails: it arrives at a deadlock, while a legal way of replacement exists. We suggest a Dependence Graph model for solving the problem. The Dependence Graph nodes are: a) the sub-routes resulting from sub-dividing all N i and C i by the routers common to N i and C i , and b) the potentially congested links. We define which new sub-routes are legal for replacement. Further, we describe the changes in routing and in the Dependence Graph resulting from launching a legal new subroute. Summarizing, we reduce the route replacement problem to finding an (optimal) sequence of launchings of currently legal new sub-routes, using the dynamic Dependence Graph. Moreover, we suggest a novel meta-approach for resolving deadlocks, by utilizing the optical wires that connect the SDN controller to the routers.