分布式、并行与集群计算
Worldwide, Geo-distributed Data Centers (GDCs) provide computing and storage services for massive workflow applications, resulting in high electricity costs that vary depending on geographical locations and time. How to reduce electricity…
Parallel programs require software support to coordinate access to shared data. For this purpose, modern programming languages provide strongly-consistent shared objects. To account for their many usages, these objects offer a large API.…
In this paper, we present Raptr--a Byzantine fault-tolerant state machine replication (BFT SMR) protocol that combines strong robustness with high throughput, while attaining near-optimal theoretical latency. Raptr delivers exceptionally…
The increasing volume and complexity of IoT systems demand a transition from the cloud-centric model to a decentralized IoT architecture in the so-called Computing Continuum, with no or minimal reliance on central servers. This paradigm…
This article introduces a novel middleware that utilizes cost-effective, low-power computing devices like Raspberry Pi to analyze data from wireless sensor networks (WSNs). It is designed for indoor settings like historical buildings and…
Contention resolution addresses the challenge of coordinating access by multiple processes to a shared resource such as memory, disk storage, or a communication channel. Originally spurred by challenges in database systems and bus networks,…
One of the most important tasks in scientific publishing is the articles' evaluation via the editorial board and the reviewers' community. Additionally, in scientific publishing great concern exists regarding the peer-review process and how…
This paper presents the design, implementation, and operation of a novel distributed fault-tolerant middleware. It uses interconnected WSNs that implement the Map-Reduce paradigm, consisting of several low-cost and low-power mini-computers…
Serverless computing abstracts away server management, enabling automatic scaling, efficient resource utilization, and cost-effective pricing models. However, despite these advantages, it faces the significant challenge of cold-start…
Training large Deep Neural Network (DNN) models at scale often encounters straggler issues, mostly in communications due to network congestion, RNIC/switch defects, or topological asymmetry. Under advanced pipeline parallelism, even minor…
Unstructured mesh tallies are a bottleneck in Monte Carlo neutral particle transport simulations of fusion reactors. This paper introduces the PUMI-Tally library that takes advantage of mesh adjacency information to accelerate these tallies…
Designing reconfiguration schemes for consensus protocols is challenging because subtle corner cases during reconfiguration could invalidate the correctness of the protocol. Thus, most systems that embed consensus protocols conservatively…
Autonomous reconfiguration of agent-based systems is a key challenge in the study of programmable matter, distributed robotics, and molecular self-assembly. While substantial prior work has focused on size-preserving transformations, much…
Deploying a Hierarchical Federated Learning (HFL) pipeline across the computing continuum (CC) requires careful organization of participants into a hierarchical structure with intermediate aggregation nodes between FL clients and the global…
Large-scale distributed model training requires simultaneous training on up to thousands of machines. Faulty machine detection is critical when an unexpected fault occurs in a machine. From our experience, a training task can encounter two…
This study was aimed at simultaneously achieving sufficient accuracy and high performance for general matrix multiplications. Recent architectures, such as NVIDIA GPUs, feature high-performance units designed for low-precision matrix…
The occurrence of bubbles in pipeline parallelism is an inherent limitation that can account for more than 40% of the large language model (LLM) training time and is one of the main reasons for the underutilization of GPU resources in LLM…
Existing LLM serving strategies can be categorized based on whether prefill and decode phases are disaggregated: non-disaggregated (NoDG) or fully disaggregated (FuDG). However, the NoDG strategy leads to strong prefill-decode interference…
Drone fleets with onboard cameras coupled with computer vision and DNN inferencing models can support diverse applications. One such novel domain is for one or more buddy drones to assist Visually Impaired People (VIPs) lead an active…
As fusion energy devices advance, plasma simulations are crucial for reactor design. Our work extends BIT1 hybrid parallelization by integrating MPI with OpenMP and OpenACC, focusing on asynchronous multi-GPU programming. Results show…