Search

Power management (PM) is cumbersome for today’s computing systems. Attainable performance is bounded by the architecture’s computing efficiency and capped in temperature, current, and power. PM is composed of multiple interacting layers. High-level controllers (HLCs) involve application-level policies, operating system agents (OSPMs), and PM governors and interfaces. The application of high-level control decisions is currently delegated to an on-chip power management unit executing tailored PM firmware routines. The complexity of this structure arises from the scale of the interaction, which pervades the whole system architecture. This paper aims to characterize the cost of the communication backbone between high-level OSPM agents and the on-chip power management unit (PMU) in high performance computing (HPC) processors. For this purpose, we target the System Control and Management Interface (SCMI), which is an open standard proposed by Arm. We enhance a fully open-source, end-to-end FPGA-based HW/SW framework to simulate the interaction between a HLC, a HPC system, and a PMU. This includes the application-level PM policies, the drivers of the operating system-directed configuration and power management (OSPM) governor, and the hardware and firmware of the PMU, allowing us to evaluate the impact of the communication backbone on the overall control scheme. With this framework, we first conduct an in-depth latency study of the communication interface across the whole PM hardware (HW) and software (SW) stack. Finally, we studied the impact of latency in terms of the quality of the end-to-end control, showing that the SCMI protocol can sustain reactive power management policies.

Abstract Supercomputers are the most powerful computing machines available to society. They play a central role in economic, industrial, and societal development. While they are used by scientists, engineers, decision-makers, and data-analyst to computationally solve complex problems, supercomputers and their hosting datacenters are themselves complex power-hungry systems. Improving their efficiency, availability, and resiliency is vital and the subject of many research and engineering efforts. Still, a major roadblock hinders researchers: dearth of reliable data describing the behavior of production supercomputers. In this paper, we present the result of a ten-year-long project to design a monitoring framework (EXAMON) deployed at the Italian supercomputers at CINECA datacenter. We disclose the first holistic dataset of a tier-0 Top10 supercomputer. It includes the management, workload, facility, and infrastructure data of the Marconi100 supercomputer for two and half years of operation. The dataset (published via Zenodo) is the largest ever made public, with a size of 49.9TB before compression. We also provide open-source software modules to simplify access to the data and provide direct usage examples.

In this study the optimal sizing of a hybrid battery/hydrogen Energy Storage System “ESS” is assessed via a model-based parametric analysis in the context of a real hybrid renewable microgrid located in Huelva, Spain, supplying a realtime monitored residential load (3.5 kW; 5.6 MWh/year) in island mode. Four storage configurations (battery-only, H2-only, hybrid battery priority and hybrid H2 priority) are assessed under different Energy Management Strategies, analysing system performance parameters such as Loss of Load “LL” (kWh;%), Over Production “OP” (kWh;%), round-trip storage efficiency ηESS (%) and total storage cost (€) depending on the ESS sizing characteristics. A parallel approach to the storage optimal sizing via both multi-dimensional sensitivity analysis and PSO is carried out, in order to address both sub-optimal and optimal regions, respectively. Results show that a hybridised ESS capacity is beneficial from an energy security and efficiency point of view but can represent a substantial additional total cost (between 100 and 300 k€) to the hybrid energy system, especially for the H2 ESS which presents higher costs. Reaching 100% supply from renewables is challenging and introducing a LL threshold induces a substantial relaxation of the sizing and cost requirements. Increase in battery capacity is more beneficial for the LL abatement while increasing H2 capacity is more useful to absorb large quantities of excess energy. The optimal design via PSO technique is complemented to the parametric study.

To cope with the increasing complexity of digital systems programming, deep learning techniques have recently been proposed to enhance software deployment by analysing source code for different purposes, ranging from performance and energy improvement to debugging and security assessment. As embedded platforms for cyber-physical systems are characterised by increasing heterogeneity and parallelism, one of the most challenging and specific problems is efficiently allocating computational kernels to available hardware resources. In this field, deep learning applied to source code can be a key enabler to face this complexity. However, due to the rapid development of such techniques, it is not easy to understand which of those are suitable and most promising for this class of systems. For this purpose, we discuss recent developments in deep learning for source code analysis, and focus on techniques for kernel mapping on heterogeneous platforms, highlighting recent results, challenges and opportunities for their applications to cyber-physical systems.

In this work, we present an innovative approach for damage detection of infrastructures on-edge devices, exploiting a brain-inspired algorithm. The proposed solution exploits recurrent spiking neural networks (LSNNs), which are emerging for their theoretical energy efficiency and compactness, to recognise damage conditions by processing data from low-cost accelerometers (MEMS) directly on the sensor node. We focus on designing an efficient coding of MEMS data to optimise SNN execution on a low-power microcontroller. We characterised and profiled LSNN performance and energy consumption on a hardware prototype sensor node equipped with an STM32 embedded microcontroller and a digital MEMS accelerometer. We used a hardware-in-the-loop environment with virtual sensors generating data on an SPI interface connected to the physical microcontroller to evaluate the system with a data stream from a real viaduct. We exploited this environment also to study the impact of different on-sensor encoding techniques, mimicking a bio-inspired sensor able to generate events instead of accelerations. Obtained results show that the proposed optimised embedded LSNN (eLSNN), when using a spike-based input encoding technique, achieves 54% lower execution time with respect to a naive LSNN algorithm implementation present in the state-of-the-art. The optimised eLSNN requires around 47 kCycles, which is comparable with the data transfer cost from the SPI interface. However, the spike-based encoding technique requires considerably larger input vectors to get the same classification accuracy, resulting in a longer pre-processing and sensor access time. Overall the event-based encoding techniques leads to a longer execution time (1.49×) but similar energy consumption. Moving this coding on the sensor can remove this limitation leading to an overall more energy-efficient monitoring system.

Objectives: We described Klebsiella pneumoniae producing carbapenemase (CPKP) spread from 01/01/2015 to 13/09/16 in a tertiary level hospital. Methods: The first positive surveillance rectal swab (SRS) or clinical sample (CS) collected in the medical department (MD), surgical department (SD) and intensive care department (ICD) were included in the study. A validated in-house Real-Time PCR method was used to detect carbapenemases; multilocus sequence typing (MLST) was used for further characterization of the strains. Results: 21535 patients were included: 213 CPKP strains from surveillance rectal swab (SRS) and 98 from clinical samples (CS) were collected. The percentage of CPKP detected in SRS with respect to CS increased in the medical MD from 2015 to 2016 (p = 0.01) and in ICD from 2012 to 2015 (p = 0.0001), while it decreased in SD from 2014 to 2016 (p = 0.003); 68.5% of the positive SRS had a previous negative SRS; CPKP was more frequently identified in CS than in SRS in MD. Twelve strains harboured more than one carbapenemase gene. Many other species harbouring a carbapenemase gene were collected. Conclusions: MDs need more inclusive surveillance criteria. The late detection of positive SRS underlined the risk of colonization during hospitalization.

Background and aims. The aim of this study is to define the epidemiology and the antimicrobial resistance profile of Ureaplasma spp. and Mycoplasma hominis isolated from urogenital specimens of patients attending the Microbiology and Virology Unit of Padua between January 2014 and December 2015. Materials and methods. The analysis was carried out on a total of 10861 samples. Species identification and antimicrobial susceptibility tests were performed using bioMérieux Mycoplasma IST 2. Results. 2668 samples (24.6%) from 2043 patients were positive: 2466 samples positive for Ureaplasma spp. (1897 patients) and 8 samples positive for M. hominis (8 patients), while in 194 samples (138 patients) was detected a coinfection. As for antimicrobial susceptibility of Ureaplasma spp. we found a resistance (R+I) rate of 90.6% for ciprofloxacin, of 74,1% for ofloxacin, of 52.8% for azithromycin and of 47.0% for erythromycin.Conclusions. Our report shows a high prevalence of Ureaplasma spp. in the study population. Surveillance of antibiotic resistance is critical for an appropriate therapeutic approach, which must always be contextualized with patient’s symptomatology.

The aim of this study is to identify and monitor the presence of Aeromonas spp. strains in stool cultures. We analyzed 5564 stool cultures from September 2012 to August 2013. Sixty-three patients were positive for Aeromonas spp. The most frequent symptoms were: diarrhea (46.0%) and abdominal pain (12.7%). Pediatric subjects were 28. Samples’ microscopic examination showed leukocytes in 38.1% of cases. It is still controversial whether Aeromonas are responsible for human gastroenteritis, but their presence in faecies of symptomatic patients supports their etiologic role. We propose search for toxins by polymerase chain reaction to identify strains that require an antibiotic therapy.

Introduction: In the management of nosocomial infections, the emergence of carbapenem-resistance in Enterobacteriaceae strains is a matter of increasing concern. The aim of this study was to detect and actively survey the presence of these strains by using phenotypic and genotypic assays. Methods: The screening of the strains with reduced carbapenem susceptibility was carried out using the automated Vitek®2 System (bioMérieux). Subsequently, we performed both different confirmation phoenotypic assays (Modified Hodge Test, Kirby-Bauer test, Etest® MBL MP/MPI, Etest® AmpC, Mueller-Hinton/cloxacillin test) and molecular identification tests ( PCR assays). Results: All the screened carbapenemase-producing strains (on the bases of the phenotypic tests results) were confirmed to harbour β-lactamase genes (by PCR analysis): 460 strains were positive for the presence of the blaKPC and 6 for the presence of the blaVIM gene. Conclusions: KPC-positive Klebsiella pneumoniae strains caused in our hospital,in the period 2008-2012, an outbreak of infection that was subsequently restricted with effective containment measures. It is mandatory that all carbapenemase-producing species will be closely monitored in hospital setting, to contain the possible rapid spread of strains showing more expansion properties and virulence.