Results 1 - 10
of
10
On Partitioning Dynamic Adaptive Grid Hierarchies
- Proceedings of the 29th Annual Hawaii International Conference on System Sciences
, 1996
"... This paper presents a computationally efficient runtime partitioning and load-balancing scheme for the Distributed Adaptive Grid Hierarchies that underlie adaptive mesh-refinement methods. The partitioning scheme yields an efficient parallel computational structure that maintains locality to reduce ..."
Abstract
-
Cited by 78 (23 self)
- Add to MetaCart
(Show Context)
This paper presents a computationally efficient runtime partitioning and load-balancing scheme for the Distributed Adaptive Grid Hierarchies that underlie adaptive mesh-refinement methods. The partitioning scheme yields an efficient parallel computational structure that maintains locality to reduce communications. Further, it enables dynamic re-partitioning and loadbalancing of the adaptive grid hierarchy to be performed cost-effectively. The run-time partitioning support presented has been implemented within the framework of a data-management infrastructure supporting dynamic distributed data-structures for parallel adaptive numerical techniques. This infrastructure is the foundational layer of a computational toolkit for the Binary Black-Hole NSF Grand Challenge project. 1 Introduction Dynamically adaptive methods for the solution of partial differential equations that employ locally optimal approximations can yield highly advantageous ratios for cost/accuracy when compared to metho...
Systems Engineering For High Performance Computing Software: The Hdda/dagh Infrastructure For Implementation Of Parallel Structured Adaptive Mesh Refinement
- In Structured Adaptive Mesh Refinement Grid Methods, IMA Volumes in Mathematics and its Applications
, 1997
"... . This paper defines, describes and illustrates a systems engineering process for development of software systems implementing high performance computing applications. The example which drives the creation of this process is development of a flexible and extendible program development infrastructure ..."
Abstract
-
Cited by 50 (6 self)
- Add to MetaCart
(Show Context)
. This paper defines, describes and illustrates a systems engineering process for development of software systems implementing high performance computing applications. The example which drives the creation of this process is development of a flexible and extendible program development infrastructure for parallel structured adaptive meshes, the HDDA/DAGH package. The fundamental systems engineering principles used (hierarchical abstractions based on separation of concerns) are well-known but are not commonly applied in the context of high performance computing software. Application of these principles will be seen to enable implementation of an infrastructure which combines breadth of applicability and portability with high performance. Key words. Software systems engineering, Structured adaptive mesh-refinement, High performance software development, Distributed dynamic data-structures. 1. Overview. This paper describes the systems engineering process which was followed in the develop...
Distributed Dynamic Data-Structures for Parallel Adaptive Mesh-Refinement
- PROCEEDINGS OF THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING
, 1995
"... This paper presents the design and implementation of dynamic distributed data-structures to support parallel adaptive (multigrid) finite difference codes based on hierarchical adaptive mesh-refinement (AMR) techniques for the solution of partial differential equations. The abstraction provided by th ..."
Abstract
-
Cited by 38 (11 self)
- Add to MetaCart
(Show Context)
This paper presents the design and implementation of dynamic distributed data-structures to support parallel adaptive (multigrid) finite difference codes based on hierarchical adaptive mesh-refinement (AMR) techniques for the solution of partial differential equations. The abstraction provided by the datastructures is a dynamic hierarchical grid where operations on the grid are independent of its distribution across processors in a parallel execution environment, and of the number of levels in the grid hierarchy. The distributed dynamic data-structures have been implemented aspart of a computational toolkit for the Binary Black-Hole NSF Grand Challenge project.
An Architecture for High Performance File System I/O
"... Abstract — This paper presents an architecture of current filesystem implementations as well as our new filesystem SpadFS and operating system Spad with rewritten VFS layer targeted at high performance I/O applications. The paper presents microbenchmarks and real-world benchmarks of different filesy ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — This paper presents an architecture of current filesystem implementations as well as our new filesystem SpadFS and operating system Spad with rewritten VFS layer targeted at high performance I/O applications. The paper presents microbenchmarks and real-world benchmarks of different filesystems on the same kernel as well as benchmarks of the same filesystem on different kernels — enabling the reader to make conclusion how much is the performance of various tasks affected by operating system and how much by physical layout of data on disk. The paper describes our novel features — most notably continuous allocation of directories and cross-file readahead — and shows their impact on performance.
SPE 37977 A New Generation EOS Compositional Reservoir Simulator: Part II – Framework and Multiprocessing
"... This paper was prepared for presentation at the 1997 SPE ..."
A new generation EOS compositional . . .
, 1997
"... This paper describes the design and implementation of a Problem Solving Environment (PSE) for developing parallel reservoir simulators that use multiple fault blocks, multiple physical models and dynamic locally adaptive meshrefinements. The objective of the PSE is to reduce the complexity of buildi ..."
Abstract
- Add to MetaCart
This paper describes the design and implementation of a Problem Solving Environment (PSE) for developing parallel reservoir simulators that use multiple fault blocks, multiple physical models and dynamic locally adaptive meshrefinements. The objective of the PSE is to reduce the complexity of building flexible and efficient parallel reservoir simulators through the use of a high-level programming interface for problem specification and model composition, object -oriented programming abstractions that implement application objects, and distributed dynamic data-management that efficiently supports adaptation and parallelism.
An Architecture for High Performance File System I/O
"... Abstract — This paper presents an architecture of current filesystem implementations as well as our new filesystem SpadFS and operating system Spad with rewritten VFS layer targeted at high performance I/O applications. The paper presents microbenchmarks and real-world benchmarks of different filesy ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — This paper presents an architecture of current filesystem implementations as well as our new filesystem SpadFS and operating system Spad with rewritten VFS layer targeted at high performance I/O applications. The paper presents microbenchmarks and real-world benchmarks of different filesystems on the same kernel as well as benchmarks of the same filesystem on different kernels — enabling the reader to make conclusion how much is the performance of various tasks affected by operating system and how much by physical layout of data on disk. The paper describes our novel features — most notably continuous allocation of directories and cross-file readahead — and shows their impact on performance.
SAFIUS -- A Secure . . .
, 2004
"... With storage data requirements growing at around 40 % every year, the entities that control the data are different from entities that consume the data. The drastic growth in storage requirements and the ubiquitous accessibility of data requires a re-look at the traditional storage architectures. O ..."
Abstract
- Add to MetaCart
With storage data requirements growing at around 40 % every year, the entities that control the data are different from entities that consume the data. The drastic growth in storage requirements and the ubiquitous accessibility of data requires a re-look at the traditional storage architectures. One possible model to tackle the storage data growth is storage service outsourcing and storage maintenance outsourcing, forcing the storage to be treated as untrusted entity. This thesis proposes SAFIUS, an architecture for a secure, accountable distributed filesystem that can reside over untrusted storage. The design leverages trust on a lockserver to provide secure access of data residing on untrusted storage, to a large number
System Sensitive Runtime Management of Dynamic Applications
, 2001
"... OF THE THESIS System Sensitive Runtime Management of Dynamic Applications by Shweta Sinha Thesis Director: Professor Manish Parashar In this thesis we investigate the design and evaluation of a system sensitive partitioning and load balancing framework for distributed adaptive grid hierarchies ..."
Abstract
- Add to MetaCart
OF THE THESIS System Sensitive Runtime Management of Dynamic Applications by Shweta Sinha Thesis Director: Professor Manish Parashar In this thesis we investigate the design and evaluation of a system sensitive partitioning and load balancing framework for distributed adaptive grid hierarchies that underlie parallel adaptive mesh-refinement (AMR) techniques for the solution of partialdi #erential equations. The framework uses system capabilities and current system state to select the appropriate distribution parameters (e.g., partitioning granularity, load per processor) to maximize overall application performance. System state is dynamically monitored, and is used to compute the relative abilities of the computing nodes to do work. These capacities are then used for partitioning and load balancing.
