Analog memristive devices have the potential to merge computing and memory, support local learning, reach high densities, enable 3D stacking, and low energy consumption for neuromorphic computing applications. Yet, in...
详细信息
The paper proposes a classification of the main groups of measurement information compression methods. We carried out an analysis of existing gaps in the range of algorithms and approaches used in practice. The paper ...
详细信息
Ensuring reproducibility in high-performance computing (HPC) applications is a significant challenge, particularly when nondeterministic execution can lead to untrustworthy results. Traditional methods that compare fi...
详细信息
ISBN:
(纸本)9798400706233
Ensuring reproducibility in high-performance computing (HPC) applications is a significant challenge, particularly when nondeterministic execution can lead to untrustworthy results. Traditional methods that compare final results from multiple runs often fail because they provide sources of discrepancies only a posteriori and require substantial resources, making them impractical and unfeasible. This paper introduces an innovative method to address this issue by using scalable capture and comparing intermediate multi-run results. By capitalizing on intermediate checkpoints and hash-based techniques with user-defined error bounds, our method identifies divergences early in the execution paths. We employ Merkle trees for checkpoint data to reduce the I/O overhead associated with loading historical data. Our evaluations on the nondeterministic HACC cosmology simulation show that our method effectively captures differences above a predefined error bound and significantly reduces I/O overhead. Our solution provides a robust and scalable method for improving reproducibility, ensuring that scientific applications on HPC systems yield trustworthy and reliable results.
As research and deployment of AI grows, the computational burden to support and sustain its progress inevitably does too. To train or fine-tune state-of-the-art models in NLP, computer vision, etc., some form of AI ha...
详细信息
ISBN:
(纸本)9798400703874
As research and deployment of AI grows, the computational burden to support and sustain its progress inevitably does too. To train or fine-tune state-of-the-art models in NLP, computer vision, etc., some form of AI hardware acceleration is virtually a requirement. Recent large language models require considerable resources to train and deploy, resulting in significant energy usage, potential carbon emissions, and massive demand for GPUs and other hardware accelerators. However, this surge carries large implications for energy sustainability at the HPC/datacenter level. In this paper, we study the effects of power-capping GPUs at a research supercomputing center on GPU temperature and power draw;we show significant decreases in both temperature and power draw, reducing power consumption and potentially improving hardware life-span, with minimal impact on job performance. To our knowledge, our work is the first to conduct and make available a detailed analysis of the effects of GPU power-capping at the supercomputing scale. We hope our work will inspire HPCs/datacenters to further explore, evaluate, and communicate the impact of power-capping AI hardware accelerators for more sustainable AI.
5G and future 6G networks support diverse combinations of access technologies, architectures, and radio frequencies, with each combination termed as a "band" henceforth. Through comprehensive measurements in...
详细信息
ISBN:
(纸本)9798400704970
5G and future 6G networks support diverse combinations of access technologies, architectures, and radio frequencies, with each combination termed as a "band" henceforth. Through comprehensive measurements in 12 cities across 5 countries, we experimentally show that operator-configured default bands are often highly suboptimal, particularly under mobility. We then propose smart band switching, where a UE's band can be dynamically changed to improve the network performance and boost the application QoE. We discuss challenges, opportunities, and design choices for building a practical smart band switching system. We further develop preliminary UE-side band-switching logic on commodity smartphones, and evaluate it on commercial 5G networks.
NASA Glenn Research Center's quantum metrology approach is to combine measurements and models. measurement results and models are subsequently integrated with NASA's aerospace competency needs to provide an un...
详细信息
ISBN:
(纸本)9781510670839;9781510670822
NASA Glenn Research Center's quantum metrology approach is to combine measurements and models. measurement results and models are subsequently integrated with NASA's aerospace competency needs to provide an understanding of how spaceflight components work together in quantum network architectures. Trade studies and device measurements are performed within NASA's Quantum Metrology Laboratory (NQML) whereas dynamic quantum network modeling occurs via the NASA Quantum Communications analysis Suite (NQCAS) simulation tool. In illustrating the synthesis of the network model and metrology for quantum network development, we have focused on the evaluation of a degenerate Spontaneous Parametric Down Conversion (SPDC) source. Here we present an overview of Hong-Ou-Mandel and Joint Spectral Intensity measurements of the degenerate SPDC source. Results of these experiments are input into NQCAS to evaluate source suitability for entanglement swapping. This demonstrates the technology development approach of coupling of quantum measurement and free space quantum network models.
The iCloud Private Relay (PR) is a new feature introduced by Apple in June 2021 that aims to enhance online privacy by protecting a subset of web traffic from both local eavesdroppers and websites that use IP-based tr...
详细信息
ISBN:
(纸本)9798400700989
The iCloud Private Relay (PR) is a new feature introduced by Apple in June 2021 that aims to enhance online privacy by protecting a subset of web traffic from both local eavesdroppers and websites that use IP-based tracking. The service is integrated into Apple's latest operating systems and uses a two-hop architecture where a user's web traffic is relayed through two proxies run by disjoint entities. PR's multi-hop architecture resembles traditional anonymity systems such as Tor and mix networks. Such systems, however, are known to be susceptible to a vulnerability known as traffic analysis: an intercepting adversary (e.g., a malicious router) can attempt to compromise the privacy promises of such systems by analyzing characteristics (e.g., packet timings and sizes) of their network traffic. In particular, previous works have widely studied the susceptibility of Tor to website fingerprinting and flow correlation, two major forms of traffic analysis. In this work, we are the first to investigate the threat of traffic analysis against the recently introduced PR. First, we explore PR's current architecture to establish a comprehensive threat model of traffic analysis attacks against PR. Second, we quantify the potential likelihood of these attacks against PR by evaluating the risks imposed by real-world AS-level adversaries through empirical measurement of Internet routes. Our evaluations show that some autonomous systems are in a particularly strong position to perform traffic analysis on a large fraction of PR traffic. Finally, having demonstrated the potential for these attacks to occur, we evaluate the performance of several flow correlation and website fingerprinting attacks over PR traffic. Our evaluations show that PR is highly vulnerable to state-of-the-art website fingerprinting and flow correlation attacks, with both attacks achieving high success rates. We hope that our study will shed light on the significance of traffic analysis to the current PR deployme
The rapid growth of IoT devices generates valuable data. However, current centralized IoT device data trading solutions pose privacy and security risks. Emerging blockchain technology aims to address these issues by d...
The proceedings contain 122 papers. The topics discuss include: centralization problem for opinion convergence in decentralized networks;annotators’ perspectives: exploring the influence of identity on interpreting m...
ISBN:
(纸本)9798400704093
The proceedings contain 122 papers. The topics discuss include: centralization problem for opinion convergence in decentralized networks;annotators’ perspectives: exploring the influence of identity on interpreting misogynoir;maximizing influence with graph neural networks;structure and dynamics of a charitable donor co-attendance network;how popularity shapes user interactions in tech-related online communities;together apart: decoding support dynamics in online COVID-19 communities;classifying severe weather events by utilizing social sensor data and social network analysis;the art of active listening;analyzing bias in recommender systems: a comprehensive evaluation of YouTube’s recommendation algorithm;exploring inequity in park usage amidst the COVID-19 pandemic;and understanding online attitudes with pre-trained language models.
The proceedings contain 27 papers. The topics discussed include: a latency-levelling load balancing algorithm for fog and edge computing;an offloading algorithm for maximizing inference accuracy on edge device in an e...
ISBN:
(纸本)9781450394796
The proceedings contain 27 papers. The topics discussed include: a latency-levelling load balancing algorithm for fog and edge computing;an offloading algorithm for maximizing inference accuracy on edge device in an edge intelligence system;multi-step prediction of worker resource usage at the extreme edge;energy-constrained D2D assisted federated learning in edge computing;site-specific ray generation for accurate estimation of signal power;performance analysis of general P4 forwarding devices with controller feedback;uncovering 5G performance on public transit systems with an app-based measurement study;5G new radio sidelink link-level simulator and performance analysis;a spatial model for using the age of information in cooperative driving applications;and a novel mixed method of machine learning based models in vehicular traffic flow prediction.
暂无评论