Multiple memory module architecture offers higher performance by providing potentially doubled memory bandwidth. Two key problems in gaining high performance in this kind of architecture are variable partitioning and scheduling. However there’s little research work that has been done on these problems. In this paper, we present a new graph model for tackling the variable partitioning problem, n...

Data-graph computations are a parallel-programming model popularized by programming systems such as Pregel, GraphLab, PowerGraph, and GraphChi. A fundamental issue in parallelizing data-graph computations is the avoidance of races between computation occuring on overlapping regions of the graph. Common solutions such as locking protocols and bulk-synchronous execution often sacrifice performanc...

The DAG-based task graph model has been found effective in scheduling for performance prediction and optimization of parallel applications. However the scheduling complexity and solution normally depend on the problem size. In this paper, we propose a symbolic scheduling scheme for a parameterized task graph which models coarse-grain DAG parallelism independent of the problem size. The algorith...

We consider point to point link scheduling in Spatial Time Division Multiple Access (STDMA) wireless networks under the physical interference model. We propose a novel link scheduling algorithm based on a line graph representation of the network, by embedding the interferences between pairs of nodes into the edge weights of the line graph. Our algorithm achieves lower schedule length and lower ...

17 SCHEDULING AND LOAD BALANCING This paper presented a static scheduling heuristic called best-imaginary-level(BIL) scheduling for heterogeneous processors. The input graph is an acyclic precedence graph, where a node has diierent execution times on diierent processors. The static level of a node, or BIL, incorporates the eeect of interprocessor communication (IPC) overhead and processor heter...

Distributed systems such as Grid- and Cloud Computing provision web services to their users in all of the world. One of the most important concerns which service providers encounter is to handle total cost of ownership (TCO). The large part of TCO is related to power consumption due to inefficient resource management. Task scheduling module as a key component can has drastic impact on both user...

We propose a method for deriving parallel, scheduling optimized protocol implementations from sequential protocol speciications. We start with an SDL spec-iication, identify a common path for optimization and perform a data dependency analysis. The resulting common path graph is parallelized as far as permitted by the data dependency graph. The degree of paral-lelism is extended even further by...

This paper considers the scheduling of parallel realtime tasks with arbitrary-deadlines. Each job of a parallel task is described as a directed acyclic graph (DAG). In contrast to prior work in this area, where decomposition-based scheduling algorithms are proposed based on the DAG-structure and inter-task interference is analyzed as self-suspending behavior, this paper generalizes the federate...