相关论文: Building A High Performance Parallel File System U…

We present an efficient distributed memory parallel algorithm for computing connected components in undirected graphs based on Shiloach-Vishkin's PRAM approach. We discuss multiple optimization techniques that reduce communication volume as…

At the scale of Uber's monorepos, traditional Git workflows become a fundamental bottleneck. Cloning multi-gigabyte repositories, maintaining local checkouts, periodically syncing from upstream, and executing repetitive fetch or push…

Distributed File Systems (DFS) are essential for managing vast datasets across multiple servers, offering benefits in scalability, fault tolerance, and data accessibility. This paper presents a comprehensive evaluation of three prominent…

Large-scale training systems typically use synchronous training, requiring all GPUs to be healthy simultaneously. In our experience training on O(100K) GPUs, synchronous training results in a low efficiency due to frequent failures and long…

Graph processing at scale presents many challenges, including the irregular structure of graphs, the latency-bound nature of graph algorithms, and the overhead associated with distributed execution. While existing frameworks such as Spark…

We describe a scalable parallelization of Geant4 using commodity hardware in a collaborative effort between the College of Computer Science and the Department of Physics at Northeastern University. The system consists of a Beowulf cluster…

This paper explores the changes required of TCP to efficiently support cluster file systems such as Hadoop Distributed File System (HDFS) where the storage nodes are connected through a software defined networking (SDN). Traditional chain…

The data production for the CDF experiment is conducted on a large Linux PC farm designed to meet the needs of data collection at a maximum rate of 40 MByte/sec. We present two data production models that exploits advances in computing and…

Training foundation models, such as GPT-3 and PaLM, can be extremely expensive, often involving tens of thousands of GPUs running continuously for months. These models are typically trained in specialized clusters featuring fast,…

As particle physics experiments push their limits on both the energy and the intensity frontiers, the amount and complexity of the produced data are also expected to increase accordingly. With such large data volumes, next-generation…

Heterogeneous graph neural networks (HGNNs) have emerged as powerful algorithms for processing heterogeneous graphs (HetGs), widely used in many critical fields. To capture both structural and semantic information in HetGs, HGNNs first…

Data-intensive, graph-based computations are pervasive in several scientific applications, and are known to to be quite challenging to implement on distributed memory systems. In this work, we explore the design space of parallel algorithms…

With the ever-growing popularity of Graph Neural Networks (GNNs), efficient GNN inference is gaining tremendous attention. Field-Programming Gate Arrays (FPGAs) are a promising execution platform due to their fine-grained parallelism,…

Parallel programming models can encourage performance portability by moving the responsibility for work assignment and data distribution from the programmer to a runtime system. However, analyzing the resulting implicit memory allocations,…

DotGrid platform is a Grid infrastructure integrated with a set of open and standard protocols recently implemented on the top of Microsoft .NET in Windows and MONO .NET in UNIX/Linux. DotGrid infrastructure along with its proposed…

Genetic Algorithms (GAs) are a powerful technique to address hard optimisation problems. However, scalability issues might prevent them from being applied to real-world problems. Exploiting parallel GAs in the cloud might be an affordable…

ROOT is an object-oriented C++ framework conceived in the high-energy physics (HEP) community, designed for storing and analyzing petabytes of data in an efficient way. Any instance of a C++ class can be stored into a ROOT file in a…

Storage systems are essential building blocks for cloud computing infrastructures. Although high performance storage servers are the ultimate solution for cloud storage, the implementation of inexpensive storage system remains an open…

Dataset storage, exchange, and access play a critical role in scientific applications. For such purposes netCDF serves as a portable and efficient file format and programming interface, which is popular in numerous scientific application…

We present the outline of a research project aimed at designing and constructing a hybrid computing system that can be easily scaled up to petaflops speeds. As a first step, we envision building a prototype which will consist of three main…