SciSpot: Scientific Computing On Temporally Constrained Cloud Preemptible VMs
Not provided.
Computer Science↗
SEARCH · Engineering Papers
Results for “servers”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
DGaaS: GPU as a Service on Distributed Computing System
In the rapidly evolving landscape of scientific computing, Graphics Processing Units (GPUs) have become indispensable for their unparalleled ability to handle parallel tasks in complex calculations, simulations, and data analysis. Their utility is further magnified in machine learning and AI applications, where they significantly accelerate model training and predictive analytics. Within this context, the Triton Inference Server emerges as a pivotal open-source tool, specializing in AI inferencing and optimizing GPU utilization across various platforms and frameworks. This paper presents an in-depth study on distributed High Throughput Computing (HTC), specifically focusing on the HTCondor framework and its resource provisioning tools, GlideinWMS and HEPCloud. These systems enable large-scale scientific experiments like CMS and DUNE to efficiently access and utilize vast computational resources. The paper explores the core architectural components of GlideinWMS, including jobs, user pools, and worker nodes, and discusses their integration with GPUs and the Triton server. The primary aim of this research is to develop a solution that optimizes GPU utilization by leveraging Glideins and containers. This approach allows computational jobs, particularly those involving AI models, to use GPUs only when essential, thereby facilitating efficient sharing of limited GPU resources. To validate this architecture, the study conducted three key tests involving custom scripts, container-based servers, and Triton server deployments. However, the study faces challenges, notably in locating the Triton server and ensuring secure remote access. To address these issues, future work will focus on developing a proxy mechanism and enhancing security protocols. In conclusion, this study offers a comprehensive roadmap for effective and efficient GPU utilization in distributed High Throughput Computing. It aims to contribute significantly to the scientific community by solving pressing problems and implementing robust solutions in collaboration with the GlideinWMS and HEPCloud teams. The research sets the stage for a more efficient, scalable, and cost-effective paradigm in scientific computing.
97 MATHEMATICS AND COMPUTING↗
Publication and Impact of Preprints Included in the First 100 Editions of the CDC COVID-19 Science Update: Content Analysis
Preprints are publicly available manuscripts posted to various servers that have not been peer reviewed. Although preprints have existed since 1961, they have gained increased popularity during the COVID-19 pandemic due to the need for immediate, relevant information. The aim of this study is to evaluate the publication rate and impact of preprints included in the Centers for Disease Control and Prevention (CDC) COVID-19 Science Update and assess the performance of the COVID-19 Science Update team in selecting impactful preprints. All preprints in the first 100 editions (April 1, 2020, to July 30, 2021) of the Science Update were included in the study. Preprints that were not published were categorized as “unpublished preprints.” Preprints that were subsequently published exist in 2 versions (in a peer-reviewed journal and on the original preprint server), which were analyzed separately and referred to as “peer-reviewed preprint” and “original preprint,” respectively. Time to publish was the time interval between the date on which a preprint was first posted and the date on which it was first available as a peer-reviewed article. Impact was quantified by Altmetric Attention Score and citation count for all available manuscripts on August 6, 2021. Preprints were analyzed by publication status, publication rate, preprint server, and time to publication. Of the 275 preprints included in the CDC COVID-19 Science Update during the study period, most came from three servers: medRxiv (n=201, 73.1%), bioRxiv (n=41, 14.9%), and SSRN (n=25, 9.1%), with 8 (2.9%) coming from other sources. Additionally, 152 (55.3%) were eventually published. The median time to publish was 2.3 (IQR 1.4-3.7). When preprints posted in the last 2.3 months were excluded (to account for the time to publish), the publication rate was 67.8%. Moreover, 76 journals published at least one preprint from the CDC COVID-19 Science Update, and 18 journals published at least three. The median Altmetric Attention Score for unpublished preprints (n=123, 44.7%) was 146 (IQR 22-552) with a median citation count of 2 (IQR 0-8); for original preprints (n=152, 55.2%), these values were 212 (IQR 22-1164) and 14 (IQR 2-40), respectively; for peer-review preprints, these values were 265 (IQR 29-1896) and 19 (IQR 3-101), respectively. Prior studies of COVID-19 preprints found publication rates between 5.4% and 21.1%. Preprints included in the CDC COVID-19 Science Update were published at a higher rate than overall COVID-19 preprints, and those that were ultimately published were published within months and received higher attention scores than unpublished preprints. These findings indicate that the Science Update process for selecting preprints had a high fidelity in terms of their likelihood to be published and their impact. The incorporation of high-quality preprints into the CDC COVID-19 Science Update improves this activity’s capacity to inform meaningful public health decision-making.
59 BASIC BIOLOGICAL SCIENCES↗
Prediction of protein assemblies, the next frontier: The CASP14-CAPRI experiment
Here we present the results for CAPRI Round 50, the fourth joint CASP-CAPRI protein assembly prediction challenge. The Round comprised a total of twelve targets, including six dimers, three trimers, and three higher-order oligomers. Four of these were easy targets, for which good structural templates were available either for the full assembly, or for the main interfaces (of the higher-order oligomers). Eight were difficult targets for which only distantly related templates were found for the individual subunits. Twenty-five CAPRI groups including eight automatic servers submitted ~1250 models per target. Twenty groups including six servers participated in the CAPRI scoring challenge submitted ~190 models per target. The accuracy of the predicted models was evaluated using the classical CAPRI criteria. The prediction performance was measured by a weighted scoring scheme that takes into account the number of models of acceptable quality or higher submitted by each group as part of their five top-ranking models. Compared to the previous CASP-CAPRI challenge, top performing groups submitted such models for a larger fraction (70–75%) of the targets in this Round, but fewer of these models were of high accuracy. Scorer groups achieved stronger performance with more groups submitting correct models for 70–80% of the targets or achieving high accuracy predictions. Servers performed less well in general, except for the MDOCKPP and LZERD servers, who performed on par with human groups. In addition to these results, major advances in methodology are discussed, providing an informative overview of where the prediction of protein assemblies currently stands.
protein complexes↗
Striped Data Analysis Framework
A columnar data representation is known to be an efficient way for data storage, specifically in cases when the analysis is often done based only on a small fragment of the available data structures. A data representation like Apache Parquet is a step forward from a columnar representation, which splits data horizontally to allow for easy parallelization of data analysis. Based on the general idea of columnar data storage, working on the [LDRD Project], we have developed a striped data representation, which, we believe, is better suited to the needs of High Energy Physics data analysis. A traditional columnar approach allows for efficient data analysis of complex structures. While keeping all the benefits of columnar data representations, the striped mechanism goes further by enabling easy parallelization of computations without requiring special hardware. We will present an implementation and some performance characteristics of such a data representation mechanism using a distributed no-SQL database or a local file system, unified under the same API and data representation model. The representation is efficient and at the same time simple so that it allows for a common data model and APIs for wide range of underlying storage mechanisms such as distributed no-SQL databases and local file systems. Striped storage adopts Numpy arrays as its basic data representation format, which makes it easy and efficient to use in Python applications. The Striped Data Server is a web service, which allows to hide the server implementation details from the end user, easily exposes data to WAN users, and allows to utilize well known and developed data caching solutions to further increase data access efficiency. We are considering the Striped Data Server as the core of an enterprise scale data analysis platform for High Energy Physics and similar areas of data processing. We have been testing this architecture with a 2TB dataset from a CMS dark matter search and plan to expand it to multiple 100 TB or even PB scale. We will present the striped format, Striped Data Server architecture and performance test results.
Gutsche, Oliver↗
Energy–Performance Trade-offs in Privacy-Preserving Federated Learning on SmartNIC-Enabled HPC Systems
Federated learning (FL) is increasingly deployed on accelerator-rich high-performance computing (HPC) systems, yet the system-level energy cost of privacy-aware FL remains poorly understood, particularly across heterogeneous networking and server-placement options. We present a measurement-driven study of energy–performance trade-offs for FL on GH200-class nodes across three deployment configurations: CPU-Ethernet, CPU-InfiniBand (RDMA-capable), and a DPU-hosted FL server over InfiniBand using a BlueField-3 SmartNIC/DPU. Using NVIDIA FLARE (NVFLARE), we align node-level power telemetry with per-round timing extracted from NVFLARE logs to quantify time-to-solution (TTS), energy-to-solution (ETS), energy-delay product (EDP), and synchronization behavior for three transformer models (ALBERT, DistilBERT, BERT), trained with and without differential privacy (DP). We find that interconnect choice is the dominant driver of runtime and energy: host-managed InfiniBand consistently reduces communication overhead versus Ethernet, yielding lower TTS/ETS/EDP. In contrast, in our NVFLARE deployment, placing the FL server on the DPU does not consistently match CPU-InfiniBand performance and can be slower—especially for larger models—highlighting that server placement alone is not sufficient to guarantee end-to-end gains. Finally, under our fixed-round protocol, DP increases per-round cost and runtime variance; ETS increases largely in proportion to TTS because average node power remains relatively stable across configurations.
Kotevska, Olivera [ORNL] (ORCID:0000000316772243)↗
Network Anomaly Detection Using Federated Learning
The internet is turning out to be an integral part of every-one's lives as more and more devices are being connected to serve societal needs. Our work is motivated by two ma-jor observations. Firstly, one drawback of connecting to the network is the threat of network attacks that can compromise users' private information, leading to data loss and adversely affecting productivity. There are several traditional security mechanisms to defend against these attacks, such as firewalls, virtual private networks (VPNs), demilitarized zones (DMZs), and vulnerability scanners. One way to prevent these attacks is early detection and prevention. However, these kinds of architecture do not scale very well because of their centralized nature. Secondly, we observe from heuristics and data set distributions that the majority of the requests made to a server are innocuous. Therefore, almost all server request data sets are highly imbalanced, weighted highly towards the harmless requests.
Marfo, William↗
Differential Power Processing for Ultra-Efficient Data Storage
Here this paper presents the hardware, software, and power codesign of an ultra-efficient data storage server with differential power processing (DPP). DPP can reduce the power conversion stress, improve the efficiency, and enhance the functionality of modular power electronics systems. The power inputs of a large number of hard disk drives (HDDs) were connected in series and supported by a multiport ac-coupled differential power processing (MAC-DPP) converter through a multiwinding transformer. Methods for controlling the multi-input multi-output power flow in the multiwinding transformer while avoiding core saturation were investigated. A ten-port MAC-DPP prototype with 700-W/in 3 power density was built to support a 450-W HDD storage system with ten series-stacked voltage domains. The prototype was tested on a 50-HDD server testbench, and the overall system loss is below 1 W (99.77% system efficiency). The server was able to maintain high-speed reading and writing operation of all 50 HDDs against the worst hot-swapping scenarios. A variety of hardware/software configurations and many cloud storage techniques were tested on the fully functioning server. Experimental results show that the energy efficiency of large-scale information systems (CPU/GPU clusters, memory banks, HDD arrays, etc.) can be greatly improved by software, hardware, and power codesign.
42 ENGINEERING↗
CAFE AU LAIT: Compute-Aware Federated Augmented Low-Rank AI Training
Federated finetuning is crucial for unlocking the knowledge embedded in pretrained Large Language Models (LLMs) when data are geographically distributed across clients. Unlike finetuning with data from a single institution, federated finetuning allows collaboration across multiple institutions, enabling the utilization of diverse and decentralized datasets while preserving data privacy. Given the high computing costs of LLM training and the emphasis on energy efficiency in Federated Learning (FL), Low-Rank Adaptation (LoRA) has emerged as a widely adopted algorithm due to its significantly reduced number of trainable parameters. However, this assumes that all data silos have the necessary computing resources to compute local updates of LLMs. Nevertheless, in practice, the computing resources across clients are highly heterogeneous: while some may have access to hundreds of GPUs, others might have limited or no GPU access. Recently, federated finetuning using synthetic data has been proposed, allowing clients to participate in a collaborative training run without training LLMs locally. However, our experimental results reveal a performance gap between models trained using synthetic data and those trained using local updates. Motivated by the observed heterogeneity in computing resources and the performance gap, we propose a novel two-stage algorithm that leverages the storage and computing capabilities of a strong server. In the first stage, under the coordination of the strong server, clients with limited computing resources collaborate to generate synthetic data, which is transferred to and stored on the strong server. In the second stage, the strong server uses this synthetic data on behalf of the resource-constrained clients to perform federated LoRA finetuning alongside clients with sufficient computing resources. This approach ensures that all clients can participate in the finetuning process. Experimental results demonstrate that incorporating local updates from even a small fraction of clients improves performance compared to using synthetic data for all clients. Furthermore, we incorporate the Gaussian mechanism in both stages to guarantee client-level differential privacy.
Wang, Jiayi [ORNL]↗
A Fast VANET-Assisted Scheme for Event Data Recorders
An event data recorder (EDR) is a device installed in a vehicle to record information. Similar to a black box in an airplane, an EDR is used in the study of automobile accidents. Many schemes have been proposed that use vehicle network technology to help record EDR data, including schemes involving storing data on roadside units or nearby vehicles and schemes leveraging blockchain technology. However, these schemes do not take into account the vehicle company’s server; with the increased use of autonomous vehicles, the data related to these vehicles are always uploaded to the vehicle company’s server. In this scenario, we classify the situation into different cases, according to whether or not it is an emergency and whether the vehicle and the server are connected. For these cases, we propose a scheme whereby a vehicle uploads the EDR data to a cloud server and sends the evidence of storage to the nearby vehicle through a vehicular ad hoc network. Our scheme offers a fast response due to the use of symmetric cryptography algorithms while also considering security requirements.
Liu, Wei↗
OpenStudio®-MCP [SWR-26-035]
OpenStudio®-MCP is a Model Context Protocol (MCP) server that lets AI assistants perform building energy modeling through natural language. Rather than requiring users to learn the OpenStudio® SDK, EnergyPlus® scripting, or Ruby/Python automation, the server translates conversational requests into sequences of tool calls that create models, design HVAC systems, run simulations, and extract results — all within a single chat session. The server's 124 tools are organized into a skills architecture where each skill encapsulates a domain of building energy modeling (envelope, HVAC, loads, weather, simulation, results) behind typed, LLM-friendly interfaces. High-leverage operations like applying ASHRAE 90.1 baseline systems or generating standards-compliant typical buildings are exposed as single tool calls that internally wire dozens of OpenStudio® objects. Bundled measures from ComStock™ and Openstudio® -common-measures-gem are wrapped with dedicated tools and typed arguments rather than exposed through a generic measure interface, so AI models get consistent, error-resistant recipes without needing to discover measure arguments at runtime. A key design decision is structured results extraction: six SQL-based tools return surgical ~300–1,000 token responses (end-use breakdowns, envelope summaries, HVAC sizing, timeseries data) instead of requiring the AI to parse ~100K-token raw HTML reports, making iterative design exploration practical within context window limits. The codebase is designed as a reference implementation — explicit, well-commented, and modular — so that other simulation engines (EnergyPlus® standalone, TRNSYS, DOE-2) can use it as a template for building their own MCP servers.
Ball, Brian [National Laboratory of the Rockies (N↗
Extreme Efficiency 240 Vac to Load Data Center Power Delivery Topologies and Control
The research explored innovative new circuit topologies and power converter architectures for data center power delivery, from the 240 V ac voltage at the input of a server rack, down to the less than 1 V dc power supply of the latest microprocessors. The research pursued three parallel tracks to increase the power conversion efficiency and to reduce the overall size of the servers: 1) An innovative new ac-dc power converter topology which enables improved efficiency through use of a novel direct step-down topology, well suited for the popular 48V dc power distribution. 2) A new series-stacked power delivery approach for servers comprising a large number of similar loads, such as hard drives in storage servers. The approach greatly reduces the power conversion losses, and achieves efficiency well over 99%. 3) A new family of hybrid switched-capacitor power converters that enable improved efficiency and reduced size through the use of dense ceramic capacitors and new circuit topologies that leverage GaN and low-voltage Si transistor technology. Together, the accomplished research achieved the goal of reducing the power conversion losses in datacenters, thereby reducing greenhouse gas emissions, operating costs, and overall electricity consumption.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Deployment of inference as a service at the US CMS Tier-2 data centers
Coprocessors, especially GPUs, will be a vital ingredient of data production workflows at the HL-LHC. At CMS, the GPU-as-a-service approach for production workflows is implemented by the SONIC project (Services for Optimized Network Inference on Coprocessors). SONIC provides a mechanism for outsourcing computationally demanding algorithms, such as neural network inference, to remote servers, where requests from multiple clients are intelligently distributed across multiple GPUs by a load-balancing service. This talk highlights the recent progress in deploying SONIC at selected U.S. CMS Tier-2 data centers. Using realistic CMS Run3 data processing workflows, such as those containing transformer-based algorithms, we demonstrate how SONIC is integrated into the production-like environment to enable accelerated inference offloading. We will present developments from both the client and server sides, including production job and data center configurations for NVIDIA and AMD GPUs. We will also present performance scaling benchmarks and discuss the challenges of operating SONIC in CMS production, such as server discovery, GPU saturation, fallback server logic, etc.
Holzman, Burt↗
A Prototype Software to Demonstrate a Data Catalog for Hanford Environmental Datasets
Ensuring that data on long-term environmental remediation at the Hanford Site is high-quality, traceable, and easily accessible is an ongoing challenge, complicated by decades of data collection, multiple contractors maintaining data sources, and the wide range of data types. A centralized data catalog, known as the Hanford Environmental Information and Data Index (HEIDI), has been under development as part of the Hanford Environmental Data Management (HEDM) program to address these challenges. HEIDI fulfills a critical need to bring together a wide range of data types and sizes from multiple authoritative data sources, while documenting the data pedigree and quality information (i.e., traceable to the data source/originator). This document describes additional development and maturation of the HEIDI prototype. Key accomplishments included deploying the catalog software, Esri Geoportal Server, on a server accessible to Hanford Local Area Network users, conducting cybersecurity evaluations, investigating integrated authentication solutions, and conducting functional testing of the catalog prototype. The server-based deployment enabled targeted feedback, leading to enhancements including improved accessibility features and an expanded metadata schema. Specifications for the server-based deployment of the prototype catalog and the HEIDI metadata schema are provided in this document to support subsequent HEIDI deployment by the U.S. Department of Energy Richland Operations Office.
54 ENVIRONMENTAL SCIENCES↗
Optical authentication of images
Systems and methods performed for generating authentication information for an image using optical computing are provided. When a user takes a photo of an object, an optical authentication system receives light reflected and/or emitted from the object. The system also receives a random key from an authentication server. The system converts the received light to plenoptic data and uploads it to the authentication server. In addition, the system generates an optical hash of the received light using the random key, converts the generated optical hash to a digital optical hash, and uploads the digital optical hash to the authentication server. When the authentication server receives the upload, it verifies whether the time of the upload is within a certain threshold time from the sending of the random key and whether the digital optical hash was generated from the same light as the plenoptic data.
97 MATHEMATICS AND COMPUTING↗
Optical authentication of images
Systems and methods performed for generating authentication information for an image using optical computing are provided. When a user takes a photo of an object, an optical authentication system receives light reflected and/or emitted from the object. The system also receives a random key from an authentication server. The system converts the received light to plenoptic data and uploads it to the authentication server. In addition, the system generates an optical hash of the received light using the random key, converts the generated optical hash to a digital optical hash, and uploads the digital optical hash to the authentication server. When the authentication server receives the upload, it verifies whether the time of the upload is within a certain threshold time from the sending of the random key and whether the digital optical hash was generated from the same light as the plenoptic data.
Murialdo, Maxwell R.↗
CMS Token Transition
Within the LHC community, a momentous transition has been occurring in authorization. For nearly 20 years, services within the Worldwide LHC Computing Grid (WLCG) have authorized based on mapping an identity, derived from an X.509 credential, or a group/role, derived from a VOMS extension issued by the experiment. A fundamental shift is occurring to capabilities: the credential, a bearer token, asserts the authorizations of the bearer, not the identity. By the HL-LHC era, the CMS experiment plans for the transition to tokens, based on the WLCG Common JSON Web Token profile, to be complete. Services in the technology architecture include the INDIGO Identity and Access Management server to issue tokens; a HashiCorp Vault server to store and refresh access tokens for users and jobs; a managed token bastion server to push credentials to the HTCondor CredMon service; and HTCondor to maintain valid tokens in long-running batch jobs. We will describe the transition plans of the experiment, current status, configuration of the central authorization server, lessons learned in commissioning token-based access with sites, and operational experience using tokens for both job submissions and file transfers.
43 PARTICLE ACCELERATORS↗
Machine Learning Decoding of Full Duplex Signals
A full duplex signal is when two endpoints (server one and server two) transmit on a single conductor pair simultaneously and with the same frequency. This results in the waveforms created from each server to be merged with one another when observed at any point along the transmission line making physical analysis of the wave unobtainable. This project was orchestrated to find the means to separate the merged signal into two separate signals which represent the signals originally sent from each server without an active tap.
97 - MATHEMATICS AND COMPUTING↗