Particle-based models are widespread in the field of computer graphics and are mostly used in soft-body dynamics, for simulating surfaces such as cloth, fluids and biologic tissue. As model resolution and scenario com...
详细信息
ISBN:
(纸本)9781479984480
Particle-based models are widespread in the field of computer graphics and are mostly used in soft-body dynamics, for simulating surfaces such as cloth, fluids and biologic tissue. As model resolution and scenario complexity increases, the computation required for these particular applications becomes overwhelming for a single processing unit, especially when interactivity is required, thus parallelization must be employed in order to provide a fast, flexible and scalable simulation environment. High-performance computing architectures such as graphics clusters may provide the parallelcomputing and rendering power required, but the distributed and remote nature of the computation and rendering process introduce specific challenges that must be tackled. We propose a parallel, distributed, modular system architecture for a particle-based simulator on GPU clusters, encapsulating powerful parallel and distributed processing, distributed rendering and remote interaction techniques, for flexible, fast simulation of large models and complex scenarios. For validating and evaluating the proposed architecture, we perform a visual comparison of two largely used numeric integration methods, namely the explicit Velocity Verlet and implicit Euler integration techniques.
We describe a general purpose master-worker parallel computation Python module called PyMW. PyMW is intended to support rapid development, testing and deployment of large scale master-worker style computations on a de...
详细信息
ISBN:
(纸本)9781424437511
We describe a general purpose master-worker parallel computation Python module called PyMW. PyMW is intended to support rapid development, testing and deployment of large scale master-worker style computations on a desktop grid or volunteer computing environment. This module targets nonexpert computer users by hiding complicated task submission and result retrieval procedures behind a simple interface. PyMW also provides a unified interface to multiple computing environments with easy extension to support additional environments. In this paper, we describe the internal structure and external interface to the PyMW module and its support for the Condor computing environment and the Berkeley Open Infrastructure for Network computing (BOINC) platform. We demonstrate the effectiveness and scalability of PyMW by performing master-worker style computations on a desktop grid using Condor and a BOINC volunteer computing project.
Advances in communication for parallel programming have yielded one-sided messaging systems. The MPI bindings for Ruby have been augmented to include the remote memory access functions of MPI-2.
ISBN:
(纸本)0780321754
Advances in communication for parallel programming have yielded one-sided messaging systems. The MPI bindings for Ruby have been augmented to include the remote memory access functions of MPI-2.
This paper reviews the problem of correctness in the context of scaleable distributed systems. It takes the concept of a strongly eventually consistent (SEC) data structure based on the work done on conflict free repl...
详细信息
ISBN:
(纸本)9781728189468
This paper reviews the problem of correctness in the context of scaleable distributed systems. It takes the concept of a strongly eventually consistent (SEC) data structure based on the work done on conflict free replicated data types and examines the theoretical basis of a practical application to generalised database management systems. It reasons that the condition required for a database management system to implement a monotonic semilattice and remain generic is not possible. It further proposes a framework which extends the SEC condition to capture enough information to make such a system practical to implement. Finally, it evaluates such an implementation using a modified storage engine for Apache Ignite demonstrating not just its practicality but its scaleability in Ignite and other database management systems like it. The key contributions of this work are Semantic and intent defined conflict free replicated data types, an enriched CRDT which enables generalised database systems to meet the SEC criteria for a monotonic semilattice object and Semantic Intent Configuration Language, SICL, a language and framework for distributedcomputing.
SRAM (static random access memory)-based pipelined algorithmic solutions have become competitive alternatives to TCAMs (ternary content addressable memories) for high-throughput IP lookup. Multiple pipelines can be ut...
详细信息
SRAM (static random access memory)-based pipelined algorithmic solutions have become competitive alternatives to TCAMs (ternary content addressable memories) for high-throughput IP lookup. Multiple pipelines can be utilized in parallel to improve the throughput further. However, several challenges must be addressed to make such solutions feasible. First, the memory distribution over different pipelines, as well as across different stages of each pipeline, must be balanced. Second, the traffic among these pipelines should be balanced. Third, the intra-flow packet order (i.e. the sequence) must be preserved. In this paper, we propose a parallel SRAM-based multi-pipeline architecture for IP lookup. A two-level mapping scheme is developed to balance the memory requirement among the pipelines as well as across the stages in each pipeline. To balance the traffic, we propose an early caching scheme to exploit the data locality inherent in the architecture. Our technique uses neither a large reorder buffer nor complex reorder logic. Instead, a flow-aware queuing scheme exploiting the flow information is used to maintain the intra-flow sequence. Extensive simulation using real-life traffic traces shows that the proposed architecture with 8 pipelines can achieve a throughput of up to 10 billion packets per second, i.e. 3.2 Tbps for minimum size (40 bytes) packets, while preserving intra-flow packet order. (c) 2009 Elsevier Inc. All rights reserved.
Detecting malware during execution using machine learning models presents some hard-to-solve problems relating to data set construction seldom discussed in the literature. We identify, name these problems and show our...
详细信息
ISBN:
(纸本)9781665432818
Detecting malware during execution using machine learning models presents some hard-to-solve problems relating to data set construction seldom discussed in the literature. We identify, name these problems and show our solutions to them in the form of Curator, a specialized distributed system for detonating potentially malicious programs, extracting behavior information, and correctly labeling said behavior to construct an accurate, consistent, and reliable data set. We demonstrate Curator's need by using generated data sets to train machine learning models based on Naive Bayes, Logistic Regression, and Random Forests. Our work is currently focused on the Windows operating system.
In HPC supercomputers, the CPU, memory, network and storage play a critical role in application performance. Established benchmarks measure the theoretical peak performance of these components, but storage benchmarks ...
详细信息
ISBN:
(纸本)9798350369205;9798350369199
In HPC supercomputers, the CPU, memory, network and storage play a critical role in application performance. Established benchmarks measure the theoretical peak performance of these components, but storage benchmarks often focus solely on I/O and lack realism. To address this, we present numio, a benchmark that simulates and evaluates overlapping compute, communication and I/O phases. Furthermore, we present a novel ensemble-driven system benchmarking strategy. This approach involves running multiple benchmarks in parallel to analyse their interactions and assess the system's ability to handle the workload. Using a real-world example, we demonstrate how this approach reveals performance issues in complex HPC systems that remain hidden when using traditional methods using isolated benchmarks on empty systems.
The upcoming 5G telecommunication system is expected to provide high data rates and ultra-low latency to meet the challenging demands of future applications. Multi-access Edge computing (MEC) is a central piece of the...
详细信息
ISBN:
(纸本)9781538655559
The upcoming 5G telecommunication system is expected to provide high data rates and ultra-low latency to meet the challenging demands of future applications. Multi-access Edge computing (MEC) is a central piece of the solution by providing a cloud computing platform at the edge of the radio access network. In this paper, we argue that edge computing should be invisible to clients and should not require modifications in client applications. A prototype shows how requests to cloud services can be transparently redirected to the closest edge computing host by leveraging Software-Defined Networking, while still allowing to use the cloud as a fallback. In addition, we discuss how to scale our approach for large networks.
The Vertically Integrated Projects (VIP) Program at Georgia Tech provides a multidisciplinary research experience aimed at engaging undergraduate and graduate research students in large-scale computing research projec...
详细信息
ISBN:
(纸本)9798350311990
The Vertically Integrated Projects (VIP) Program at Georgia Tech provides a multidisciplinary research experience aimed at engaging undergraduate and graduate research students in large-scale computing research projects. Since 2019, the Future computing with the Rogues Gallery VIP course has engaged over 75 students in research on topics related to novel architectures and "post-Moore" computing platforms built around quantum, neuromorphic, near-memory, and reconfigurable computing. One of the key takeaways from this course for the course designers has been on the correlation between these novel computing platforms and traditional skills, techniques, and tools that are used in the HPC and parallelcomputing arenas. We discuss these parallels as well as the impacts of this course on general student success and research outcomes.
暂无评论