The Experts below are selected from a list of 8169 Experts worldwide ranked by ideXlab platform
Zechao Shang - One of the best experts on this subject based on the ideXlab platform.
-
catch the wind graph Workload Balancing on cloud
International Conference on Data Engineering, 2013Co-Authors: Zechao ShangAbstract:Graph partitioning is a key issue in graph database processing systems for achieving high efficiency on Cloud. However, the balanced graph partitioning itself is difficult because it is known to be NP-complete. In addition a static graph partitioning cannot keep all graph algorithms efficient for a long time in parallel on Cloud because the Workload Balancing in different iterations for different graph algorithms are all possible different. In this paper, we investigate graph behaviors by exploring the working window (we call it wind) changes, where a working window is a set of active vertices that a graph algorithm really needs to access in parallel computing. We investigated nine classic graph algorithms using real datasets, and propose simple yet effective policies that can achieve both high graph Workload Balancing and efficient partition on Cloud.
-
ICDE - Catch the Wind: Graph Workload Balancing on cloud
2013 IEEE 29th International Conference on Data Engineering (ICDE), 2013Co-Authors: Zechao ShangAbstract:Graph partitioning is a key issue in graph database processing systems for achieving high efficiency on Cloud. However, the balanced graph partitioning itself is difficult because it is known to be NP-complete. In addition a static graph partitioning cannot keep all graph algorithms efficient for a long time in parallel on Cloud because the Workload Balancing in different iterations for different graph algorithms are all possible different. In this paper, we investigate graph behaviors by exploring the working window (we call it wind) changes, where a working window is a set of active vertices that a graph algorithm really needs to access in parallel computing. We investigated nine classic graph algorithms using real datasets, and propose simple yet effective policies that can achieve both high graph Workload Balancing and efficient partition on Cloud.
Ewald Speckenmeyer - One of the best experts on this subject based on the ideXlab platform.
-
a fast parallel sat solver efficient Workload Balancing
Annals of Mathematics and Artificial Intelligence, 1996Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present a fast parallel SAT-solver on a message based MIMD machine. The input formula is dynamically divided into disjoint subformulas. Small subformulas are solved by a fast sequential SAT-solver running on every processor, which is based on the Davis-Putnam procedure with a special heuristic for variable selection. The algorithm uses optimized data structures to modify Boolean formulas. Additionally efficient Workload Balancing algorithms are used, to achieve a uniform distribution of Workload among the processors. We consider the communication network topologiesd-dimensional processor grid and linear processor array. Tests with up to 256 processors have shown very good efficiency-values (>0.95).
-
A fast parallel SAT-solver — efficient Workload Balancing
Annals of Mathematics and Artificial Intelligence, 1996Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present a fast parallel SAT-solver on a message based MIMD machine. The input formula is dynamically divided into disjoint subformulas. Small subformulas are solved by a fast sequential SAT-solver running on every processor, which is based on the Davis-Putnam procedure with a special heuristic for variable selection. The algorithm uses optimized data structures to modify Boolean formulas. Additionally efficient Workload Balancing algorithms are used, to achieve a uniform distribution of Workload among the processors. We consider the communication network topologiesd-dimensional processor grid and linear processor array. Tests with up to 256 processors have shown very good efficiency-values (>0.95).
-
Efficient Workload-Balancing on Grids, Hypercubes and Trees
Operations Research ’93, 1994Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present several algorithms for achieving efficient Workload Balancing (WLB) on message based MIMD machines. Given a set of n processors P = {p1,...,p n } and a communication network N ⊂ P × P. At a fixed point in time every processor p ∈ P has a Workload (WL) λ(p) ∈ ℝ+, which is an estimation for the time needed to solve the problems placed on p. In the following we assume, that WL is dividible in infinitely small pieces, which can be exchanged between processors. The basic step move(p, q, l) with p, q ∈ P, (p,q) ∈ N, −λ(q)≤ l ≤ λ(p) means that Workload l is moved from p to q if l > 0 or Workload −l is moved from q to p if l < 0. WL λ(p) changes into λ(p) − l and λ(q) changes into λ(q) + l.
Remco Germs - One of the best experts on this subject based on the ideXlab platform.
-
Placement of Effective Work-In-Progress Limits in Route-Specific Unit-Based Pull Systems.
International Journal of Production Research, 2011Co-Authors: Nick Ziengs, Jan Riezebos, Remco GermsAbstract:Unit-based pull systems control the throughput time of orders in a production system by limiting the number of orders on the shop floor. In production systems where orders can follow different routings on the shop floor, route-specific pull systems that control the progress of orders on the shop floor by placing limits on the number of orders in (parts of) a routing, have shown to be effective in controlling throughput times. This is because route-specific pull systems are able to create a balanced distribution of the amount of work on the shop floor, which leads to shorter and more reliable throughput times. The placement of limits on work-in-progress in a route-specific pull system determines to a large extend the Workload Balancing capability of such a system. This paper shows how the placement of work-in-progress limits affects the Workload Balancing capability and thereby the throughput time performance of a route-specific unit-based pull system, namely POLCA.
-
Workload Balancing capability of pull systems in MTO production
International Journal of Production Research, 2009Co-Authors: Remco Germs, Jan RiezebosAbstract:Pull systems focusing on throughput time control and applicable in situations with high variety and customisation are scarce. This paper compares three unit-based pull systems that can cope with such situations: POLCA, CONWIP and m-CONWIP. These systems control the shop floor throughput time of orders by limiting the number of orders on the shop floor. However, their effectiveness in terms of reducing total throughput time is questioned. Theory states that an improvement in the average total throughput time will be due to the Workload Balancing capability of a pull system, but that many pull systems lack this capability. This paper shows that this Workload Balancing capability exists for POLCA and m-CONWIP, but not for CONWIP. The magnitude of the effect differs strongly, depending on the configuration of the system, the order arrival pattern and the variability of the processing time of the orders.
Qionghai Dai - One of the best experts on this subject based on the ideXlab platform.
-
Depth Assisted Adaptive Workload Balancing for Parallel View Synthesis
IEEE Transactions on Multimedia, 2018Co-Authors: Xin Jin, Zhanqi Liu, Li Qian, Qionghai DaiAbstract:Depth image-based rendering has been adopted by MPEG as the recommended view synthesis technique for free viewpoint TV applications. In this paper, a Workload Balancing algorithm is proposed for parallel view synthesis on multicore platforms. First, view synthesis Workload is defined as the function of the number of hole-pixels in the warped images. Then, a novel depth assisted prediction method is proposed to predict the number of hole-pixels in the current frame by exploiting the depth differences between the neighboring frames, which reflects the movement of objects in video content. Feeding the predicted Workload to the proposed cost function, each input frame is partitioned adaptively to balance the synthesis Workload among the cores. The proposed Workload prediction method outperforms the existing approaches both in terms of frame average prediction error and standard deviation in prediction error. Applying the proposed Workload Balancing method, the parallel view synthesis system provides higher acceleration ratio and better synchronization performance among the cores compared with other parallel processing systems without sacrificing the subjective and objective quality. It is also robust to different platforms, which shows high potential in being applied to mobile oriented applications.
-
ICASSP - A Workload balanced parallel view synthesis for FTV
2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2015Co-Authors: Zhanqi Liu, Xin Jin, Li Chenyang, Qionghai DaiAbstract:In this paper, a parallel system together with an adaptive Workload Balancing algorithm is proposed for view synthesis on multi-core platforms. Based on system level data parallelism, an adaptive Workload Balancing method is proposed for depth image based rendering by evaluating the number of non-hole pixels after warping. Experimental results demonstrated that with the proposed Workload Balancing algorithm, the Workload difference among the cores is reduced by 90.65% on average for 2-core systems and by 79.57% on average for 4-core systems, respectively. Compared with the parallel system without the proposed Balancing algorithm, synthesis speed is further improved by 7.5% for 2-core systems and 8.9% for 4-core systems at maximum, respectively, without degradation in the subjective and objective quality.
Max Böhm - One of the best experts on this subject based on the ideXlab platform.
-
a fast parallel sat solver efficient Workload Balancing
Annals of Mathematics and Artificial Intelligence, 1996Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present a fast parallel SAT-solver on a message based MIMD machine. The input formula is dynamically divided into disjoint subformulas. Small subformulas are solved by a fast sequential SAT-solver running on every processor, which is based on the Davis-Putnam procedure with a special heuristic for variable selection. The algorithm uses optimized data structures to modify Boolean formulas. Additionally efficient Workload Balancing algorithms are used, to achieve a uniform distribution of Workload among the processors. We consider the communication network topologiesd-dimensional processor grid and linear processor array. Tests with up to 256 processors have shown very good efficiency-values (>0.95).
-
A fast parallel SAT-solver — efficient Workload Balancing
Annals of Mathematics and Artificial Intelligence, 1996Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present a fast parallel SAT-solver on a message based MIMD machine. The input formula is dynamically divided into disjoint subformulas. Small subformulas are solved by a fast sequential SAT-solver running on every processor, which is based on the Davis-Putnam procedure with a special heuristic for variable selection. The algorithm uses optimized data structures to modify Boolean formulas. Additionally efficient Workload Balancing algorithms are used, to achieve a uniform distribution of Workload among the processors. We consider the communication network topologiesd-dimensional processor grid and linear processor array. Tests with up to 256 processors have shown very good efficiency-values (>0.95).
-
Efficient Workload-Balancing on Grids, Hypercubes and Trees
Operations Research ’93, 1994Co-Authors: Max Böhm, Ewald SpeckenmeyerAbstract:We present several algorithms for achieving efficient Workload Balancing (WLB) on message based MIMD machines. Given a set of n processors P = {p1,...,p n } and a communication network N ⊂ P × P. At a fixed point in time every processor p ∈ P has a Workload (WL) λ(p) ∈ ℝ+, which is an estimation for the time needed to solve the problems placed on p. In the following we assume, that WL is dividible in infinitely small pieces, which can be exchanged between processors. The basic step move(p, q, l) with p, q ∈ P, (p,q) ∈ N, −λ(q)≤ l ≤ λ(p) means that Workload l is moved from p to q if l > 0 or Workload −l is moved from q to p if l < 0. WL λ(p) changes into λ(p) − l and λ(q) changes into λ(q) + l.
