Background: Anomaly detection is essential for detecting unusual behaviors in dynamic networks that may represent emerging security threats. Traditional models focus on First-Order Network representations that overloo...
详细信息
Robustly estimating energy consumption in High-Performance Computing (HPC) is essential for assessing the energy footprint of modern workloads, particularly in fields such as Artificial Intelligence (AI) research, dev...
详细信息
ISBN:
(纸本)9783031800832;9783031800849
Robustly estimating energy consumption in High-Performance Computing (HPC) is essential for assessing the energy footprint of modern workloads, particularly in fields such as Artificial Intelligence (AI) research, development, and deployment. The extensive use of supercomputers for AI training has heightened concerns about energy consumption and carbon emissions. Existing energy estimation tools often assume exclusive use of computing nodes, a premise that becomes problematic with the advent of supercomputers integrating microservices, as seen in initiatives like Acceleration as a Service (XaaS) and cloud computing. This work investigates the impact of executed instructions on overall power consumption, providing insights into the comprehensive behaviour of HPC systems. We introduce two novel mathematical models to estimate a process's energy consumption based on the total node energy, process usage, and a normalised vector of the probability distribution of instruction types for CPU and GPU processes. Our approach enables energy accounting for specific processes without the need for isolation. Our models demonstrate high accuracy, predicting CPU power consumption with a mere 1.9% error. For GPU predictions, the models achieve a central relative error of 9.7%, showing a clear tendency to fit the test data accurately. These results pave the way for new tools to measure and account for energy consumption in shared supercomputing environments.
The proceedings contain 19 papers. The special focus in this conference is on integrated Formal Methods. The topics include: Implementing, Specifying, and Verifying the QOI Format in Dafny: A Case Study...
ISBN:
(纸本)9783031765537
The proceedings contain 19 papers. The special focus in this conference is on integrated Formal Methods. The topics include: Implementing, Specifying, and Verifying the QOI Format in Dafny: A Case Study;veriCode: Correct Translation of Abstract Specifications to C Code;proving Termination via Measure Transfer in Equivalence Checking;PLACIDUS: Engineering Product Lines of Rigorous Assurance Cases;stateful Functional modeling with Refinement (a Lean4 Framework);modeling Register Pairs in CompCert;monitoring Extended Hypernode Logic;towards Quantitative Analysis of Simulink Models Using Stochastic Hybrid Automata;monitoring Real-Time systems Under Parametric Delay;veyMont: Choreography-Based Generation of Correct Concurrent Programs with Shared Memory;correct and Complete Symbolic Execution for Free;solvent: Liquidity Verification of Smart Contracts;StEVe: A Rational Verification Tool for Stackelberg Security Games;PyQBF: A Python Framework for Solving Quantified Boolean Formulas;Improving SAT Solver Performance Through MLP-Predicted Genetic Algorithm Parameters;active Learning of Runtime Monitors Under Uncertainty;specify What? Enhancing Neural Specification Synthesis by Symbolic Methods.
In creating integrated automated production control systems, there is an open question of simulation fidelity distribution between MES- and APC-systems according to the requirement for speed and decision-making horizo...
详细信息
ISBN:
(纸本)9783031734168;9783031734175
In creating integrated automated production control systems, there is an open question of simulation fidelity distribution between MES- and APC-systems according to the requirement for speed and decision-making horizon at each level. Operational planning forms a set of discrete operations organized by time and indicators. The period of the operations includes the transient and steady-state mode of the manufacturing process, but does not affect the values of transient mode indicators. Transient mode is the "noise" for operational production planning. The period of scheduled operations should be determined for petroleum refineries so that the level of the "noise" does not exceed the required production planning accuracy. The paper presents a method of calculating the transient mode parameters and the analysis results for refinery process time series, enabling us to determine the minimum operation period for the required operational production planning accuracy in the MES-system. Presents response time, average control action period and obtained estimates of transient speed on the example unit for each group of real refinery processes.
This study investigates the use of Artificial Adaptive (AA) systems for predicting energy consumption in Smart Buildings, focusing on optimizing energy usage with sensors, actuators, and intelligent systems. Using the...
详细信息
The proceedings contain 35 papers. The topics discussed include: an analysis of the standard unit of processing on the Sperry UNIVAC series 1100 computer system;the critical need for computermeasurement in the 1980&#...
The proceedings contain 35 papers. The topics discussed include: an analysis of the standard unit of processing on the Sperry UNIVAC series 1100 computer system;the critical need for computermeasurement in the 1980's;capacity management: a few suggestions for survival;monitoring data communications systems with emphasis on open system interconnections (OSI);parameterization of CPU loading in packet computer networks;issues in design and measurement of local area networks;the design of a telecommunications network at the travelers insurance companies;a multi-class queueing network model of an interactive system;alternative parameterizations of product-form queueing networks;the use of analytic modelling in system selection;consider the performance of large software systems before implementation;and controlling the cost of computer performance evaluation.
The proceedings contain 38 papers. The topics discussed include: assessing manufacturing agility: an updated state-of-the-art and operational insights;a review of quality improvement framework for industry 4.0;uncover...
The proceedings contain 38 papers. The topics discussed include: assessing manufacturing agility: an updated state-of-the-art and operational insights;a review of quality improvement framework for industry 4.0;uncovering the potential and pitfalls of process mining in manufacturing;long-pulse fiber laser micro-hole drilling of nickel superalloy;image processing with deep-learning and transfer learning for cutting tool degradation monitoring;methodologies for connecting an external signal-processing unit for adaptive control in machining systems;research on internal surface finishing of multi-cross channels;static stiffness analysis of an electronically preloaded rack and pinion feed drive system;and an advanced method to assess mechanical filtering effect of tactile coordinate-measuring system in dimensional measurement of additive manufacturing parts.
Brain tumors are one of the most serious and potentially fatal diseases in the world, to have any hope of being successfully treated it is imperative that they be discovered early. Conventional diagnostic methods such...
详细信息
An ongoing problem in robotics is the calculation of body motion given motion in the visual field, also known as ego-motion estimation. This is a problem which has been solved in the visual system of most animals, inc...
详细信息
In the context of Industry 4.0, which is characterised by networked systems, Internet of Things devices, and cyber-physical networks, quantum cryptography is essential to strengthening communication channels. In contr...
详细信息
暂无评论