BlastFunction: A Full-stack Framework Bringing FPGA Hardware Acceleration to Cloud-native Applications

“Cloud-native” is the umbrella adjective describing the standard approach for developing applications that exploit cloud infrastructures’ scalability and elasticity at their best. As the application complexity and user-bases grow, designing for performance becomes a first-class engineering concern. As an answer to these needs, heterogeneous computing platforms gained widespread attention as powerful tools to continue meeting SLAs for compute-intensive cloud-native workloads. We propose BlastFunction, an FPGA-as-a-Service full-stack framework to ease FPGAs’ adoption for cloud-native workloads, integrating with the vast spectrum of fundamental cloud models. At the IaaS level, BlastFunction time-shares FPGA-based accelerators to provide multi-tenant access to accelerated resources without any code rewriting. At the PaaS level, BlastFunction accelerates functionalities leveraging the serverless model and scales functions proactively, depending on the workload’s performance. Further lowering the FPGAs’ adoption barrier, an accelerators’ registry hosts accelerated functions ready to be used within cloud-native applications, bringing the simplicity of a SaaS-like approach to the developers. After an extensive experimental campaign against state-of-the-art cloud scenarios, we show how BlastFunction leads to higher performance metrics (utilization and throughput) against native execution, with minimal latency and overhead differences. Moreover, the scaling scheme we propose outperforms the main serverless autoscaling algorithms in workload performance and scaling operation amount.

Retrofittable Solutions Capability for Gas Turbine Compressors

The increasing introduction of renewable energy capacity has changed the perspective on the operation of conventional power plants, introducing the necessity of reaching extreme off-design conditions. There is a strong interest in the development and optimization of technologies that can be retrofitted to an existing power plant to enhance flexibility as well as increase performance and lower emissions. Under the framework of the European project TURBO-REFLEX, a typical F-class gas turbine compressor designed and manufactured by Ansaldo Energia has been studied. Numerical analyses were performed using the TRAF code, which is a state-of-the-art 3D CFD RANS/URANS flow solver. In order to assess the feasibility of lower minimum environmental load operation, by utilizing a reduction in the compressor outlet mass-flow rate, with a safe stability margin, two different solutions have been analyzed: blow-off extractions and extra-closure of Variable Inlet Guide Vanes. The numerical steady-state results are compared and discussed in relation to an experimental campaign, which was performed by Ansaldo Energia. The purpose is to identify the feasibility of the technologies and implementation opportunity in the existing thermal power plant fleet.

Characterization and experimental comparison of commercial PEMFC stacks for marine applications

PEM Fuel Cells are considered among the most promising technologies for hydrogen utilization in both stationary and automotive applications. The number of FC installations on board ships – alone or in hybrid configuration with batteries – is increasing significantly, although international regulations that drive their installation are still missing. In this scenario, the project TecBia aims to identify a dedicated test protocol and the best commercial PEMFC technology for marine applications, assessing the integration of a 140 kW PEMFC system on the Zero Emission Ultimate Ship (ZEUS) vessel. The system design and technology provider has been chosen after a technical comparison based on a dedicated experimental campaign. The experimental campaign had two goals: (i) analyse the performance of the different PEMFC systems to define the best characteristics for maritime applications; (ii) verify the compliance with naval requirements with reference to current and future standards. The present study shows the resulting test protocol for FC Systems (FCS) for maritime applications, defined starting from the existing international regulations on FCS installations and on naval environment requirements; the results of its application on the commercial system chosen for the installation on ZEUS are reported.

Post-Test Numerical Analysis of a Helium-Cooled Breeding Blanket First Wall under LOFA Conditions with the MELCOR Fusion Code

The validation of numerical tools employed in the analysis of incidental transients in a fusion reactor is a topic of main concern. KIT is taking part in this task providing both experimental data and by performing numerical analysis in support of the main codes used for the safety analyses of the Helium Cooled Pebble Bed (HCPB) blanket concept. In recent years, an experimental campaign has been performed in the KIT-HELOKA facility to investigate the behavior of a First Wall Mock-Up (FWMU) under Loss Of Flow Accident (LOFA) conditions. The aim of the experimental campaign was twofold: to check the expected DEMO thermal-hydraulics conditions during normal and off-normal conditions and to provide robust data for code validation. The present work is part of these validation efforts, and it deals with the analysis of the LOFA experimental campaign with the system code MELCOR 1.8.6 for fusion. A best-estimate methodology has been used in support of this analysis to ease the distinction between user’s assumptions and code limitations. The numerical analyses are here described together with their goals, achievements, and lesson learnt.

The influences of the number of concrete dowels to shear resistance based on push out tests

To reduce the depth of floor-beam structures and to save the cost of headed-shear studs, many types of shallow composite beam have been developed during the last few years. Among them, the shallow-hollow steel beam consists of web openings, infilled with in-situ concrete (named concrete dowel) has been increasingly focused recently. In this new kind of structure, this concrete dowel plays an important role as the principal shear connector. This article presents an investigation on the shear transferring mechanism and failure behavior of the trapezoid shape concrete dowel. An experimental campaign of static push-out tests has been conducted with variability in the number of web openings (WOs). The results indicate that the mechanical behavior of concrete dowel could be divided into crushing, compression, and tension zones and exhibits brittle behavior. The longitudinal shear resistance and specimen's stiffness are strongly affected by the number of considered WOs

Confinement of Masonry Columns with Natural Lime-Based Mortar Composite: An Experimental Investigation

The paper presents the results of an experimental campaign on the confinement of masonry square columns with fiber-reinforced lime mortar (FRLM) composites made of a natural lime-based matrix. The experimental results show the effectiveness of such a composite for increasing both strength and ductility performances of strengthened columns. Predictive formulas from the literature and from the Italian guidelines CNR-DT 215/2018 do not perfectly fit the experimental outcomes and do not confirm the strength increase of the confined columns. The reason can be attributed to the very low mechanical properties of the natural matrix used to form such a composite. Therefore, considering that the use of a natural and sustainable matrix fully compatible with the masonry substrate is a fundamental requirement for strengthening masonry columns of buildings belonging to architectural heritage, an additional future effort should be made by researchers involved in this field. In particular, for a reliable prediction of the strength of masonry columns confined with composites made of natural matrices, wider experimental campaigns are necessary to refine available formulas with respect to different substrates and component materials.

INSIGHTS INTO THE FLOW WITHIN THE WELL DOCK OF A MOTHERSHIP DURING FEEDER VESSEL DOCKING MANOEUVRES

An experimental campaign has been undertaken to explore the flow around a feeder vessel as it manoeuvres in and out of the well dock of a mothership. The parent hulls for this study are drawn from the floating harbour transhipper concept created by Sea Transport Corporation. Laser measurement techniques have been employed to analyse the flow field within the well dock while the feeder vessel both enters and departs. For the Master of the feeder vessel to safely perform these manoeuvres, the complex flows resulting from the highly confined nature of the well dock concept need to be understood and potentially mitigated. It is shown that the inclusion of vents in the well dock can significantly influence the flow and that their effectiveness is determined by the size of the vents. This study further progresses the authors’ recent work on the same novel concept where the confined water effect of the well dock and inclusion of vents is quantified for both the seakeeping behaviour and the docking/departure performance. It is concluded that the use of vents is very beneficial when a feeder vessel docks or departs the well dock, however a compromise on the vent size must be reached in order to reduce adverse effects on feeder vessel motions when docked and exposed to a seaway. It is likely that the optimum solution, that covers all operational parameters, only requires the inclusion of relatively small vents.

Effect of Different Environments’ Conditioning on the Debonding Phenomenon in Fiber-Reinforced Cementitious Matrix-Concrete Joints

This paper presents the results of an experimental study conducted to understand the bond capacity through single-lap, direct-shear tests of fiber-reinforced cementitious matrix (FRCM)-concrete joints under an alkaline and hot water environment. The experimental campaign was focused on a FRCM system equipped with two different types of fibers, (PBO) and Carbon. After the conditioning, the specimens conditioned were subjected to visual inspection, and the experimental results were compared with the unconditioned specimens. Moreover, in this present work, the number of layers and the conditioning time were varied.

Enhanced confinement in diverted negative-triangularity L-mode plasmas in TCV

The favourable confinement properties of negative-triangularity (NT) tokamak configurations were discovered in the TCV tokamak in the late 1990’s and were documented over the two following decades, through investigations of predominantly electron-heated plasmas in limited topologies. The most recent experimental campaign in TCV has marked a leap forward, characterized by the development of a variety of diverted NT shapes that are robustly stable with basic Ohmic heating. The application of auxiliary heating, directed now at both electrons and ions (using electron-cyclotron resonance heating as well as neutral-beam injection), has enabled the achievement of record performances for L-mode plasmas, with normalized β values reaching 2.8 transiently (as well as 2 in steady state, but reverting to a limited configuration) and with comparable ion and electron temperatures. The systematic confinement enhancement with NT is confirmed in these experiments. The L-mode existence space is broader than at positive triangularity, with only sporadic transitions to Hmode observed up to 1.4-MW heating power regardless of the magnetic-field-gradient direction relative to the X-point. These experiments are planned to be continued with even higher power following a heating-source upgrade.