| D. Grunwald and S. Vajracharya. The DUDE runtime system: An object-oriented macro-dataflow approach to integrated task and object parallelism. Technical Report CU-CS-779-95, Dept. of Computer Science, University of Colorado, 1994. |
....type of input data to expect for the application. On the other hand, runtime systems like the NWS often lack direct knowledge of the structure of the application itself. However, it could be beneficial if our new language could save data flow information for the runtime system [Grimshaw et al. 93, Grunwald Vajracharya 94] Consider, for example, a runtime system attempting to dynamically load balance a performance critical application. When it discovers a load imbalance, it attempts to migrate a task from an overloaded processor to an underutilized processor. But which task Preferably one whose movement will not ....
D. Grunwald and S. Vajracharya. The DUDE runtime system: An object-oriented macro-dataflow approach to integrated task and object parallelism. Technical Report CU-CS-779-95, Dept. of Computer Science, University of Colorado, 1994.
....[39] implements nested data parallelism, a model which supports hierarchical parallelism and data structures through vectors of vectors. NESL is an applicative language, and provides no constructs to control data decomposition or granularity of parallelism. Several task oriented parallel languages [40, 41, 42, 43] support fork join parallelism suitable for divide and conquer. Crandall et. al [44] report experiences with dual level parallel programs on an SMP cluster and motivate further research into dual tier programming models and environments. Proteus [45] is a custom built hierarchical SMP cluster ....
D.Grunwald and S.Vajracharya, The DUDE Runtime System: An Object-Oriented MacroDataflow Approach to Integrated Task and Object Parallelism, Tech. Report CU-CS-779-95, Dept. Computer Science, Univ. of Colorado, 1994.
....[39] implements nested data parallelism, a model which supports hierarchical parallelism and data structures through vectors of vectors. NESL is an applicative language, and provides no constructs to control data decomposition or granularity of parallelism. Several task oriented parallel languages [40, 41, 42, 43] support fork join parallelism suitable for divide and conquer. Crandall et. al [44] report experiences with dual level parallel programs on an SMP cluster and motivate further research into dual tier programming models and environments. Proteus [45] is a custom built hierarchical SMP cluster ....
D. Grunwald and S. Vajracharya, "The DUDE runtime system: An object-oriented macrodataflow approach to integrated task and object parallelism," Tech. Rep. CU-CS-779-95, Dept. of Computer Science, University of Colorado, 1994.
....or granularity of parallelism. Other systems with hierarchical data structures include systems with support for grid hierarchies in adaptive mesh refinement [92, 64, 109, 116] Many hierarchical programs arise from divide and conquer algorithms. Several task oriented parallel languages [70, 27, 66, 75] support fork join parallelism suitable for divideand conquer. 2.4.3 Single Tier Parallel Languages We now review prominent parallel programming languages that apply to block structured applications, but do not explicitly address a multi tier or hierarchical programming style. Data parallel ....
D. Grunwald and S. Vajracharya. The DUDE runtime system: An objectoriented macro-dataflow approach to integrated task and object parallelism. Technical Report CU-CS-779-95, Dept. of Computer Science, University of Colorado, 1994.
No context found.
D. Grunwald and S. Vajracharya. The DUDE runtime system: An object-oriented macro-dataflow approach to integrated task and object parallelism. Technical Report CU-CS-779-95, Dept. of Computer Science, University of Colorado, 1994.
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