SearcharxivSearch

arXiv subjects

Ramon Canal

Publications and source records attributed to Ramon Canal.

At least 19 recordsLinked to original sources

Security and RAS in the Computing Continuum

Security and RAS are two non-functional requirements under focus for current systems developed for the computing continuum. Due to the increased number of interconnected computer systems across the continuum, security becomes especially pervasive at all levels, from the smallest edge device to the high-performance cloud at the other end. Similarly, RAS (Reliability, Availability, and Serviceability) ensures the robustness of a system towards hardware defects. Namely, making them reliable, with high availability and design for easy service. In this paper and as a result of the Vitamin-V EU project, the authors detail the comprehensive approach to malware and hardware attack detection; as well as, the RAS features envisioned for future systems across the computing continuum.

cs.CR

Hypervisor Extension for a RISC-V Processor

This paper describes our experience implementing a Hypervisor extension for a 64-bit RISC-V processor. We describe the design process and the main required parts with a brief explanation of each one.

cs.AR

Vitamin-V: Expanding Open-Source RISC-V Cloud Environments

Among the key contributions of Vitamin-V (2023-2025 Horizon Europe project), we develop a complete RISC-V open-source software stack for cloud services with comparable performance to the cloud-dominant x86 counterpart. In this paper, we detail the software suites and applications ported plus the three cloud setups under evaluation.

cs.DC

SafeLS: Toward Building a Lockstep NOEL-V Core

Safety-critical systems such as those in automotive, avionics and space, require appropriate safety measures to avoid silent data corruption upon random hardware errors such as those caused by radiation and other types of electromagnetic interference. Those safety measures must be able to prevent faults from causing the so-called common cause failures (CCFs), which occur when a fault produces identical errors in redundant elements so that comparison fails to detect the errors and a failure arises. The usual solution to avoid CCFs in CPU cores is using lockstep cores, so that two cores execute the same flow of instructions, but with some time staggering so that their state is never identical and faults can only lead to different errors, which are then detectable by means of comparison. This paper extends Gaisler's RISC-V NOEL-V core with lockstep; and presents future prospects for its use and distribution.

cs.AR

NEUROPULS: NEUROmorphic energy-efficient secure accelerators based on Phase change materials aUgmented siLicon photonicS

This special session paper introduces the Horizon Europe NEUROPULS project, which targets the development of secure and energy-efficient RISC-V interfaced neuromorphic accelerators using augmented silicon photonics technology. Our approach aims to develop an augmented silicon photonics platform, an FPGA-powered RISC-V-connected computing platform, and a complete simulation platform to demonstrate the neuromorphic accelerator capabilities. In particular, their main advantages and limitations will be addressed concerning the underpinning technology for each platform. Then, we will discuss three targeted use cases for edge-computing applications: Global National Satellite System (GNSS) anti-jamming, autonomous driving, and anomaly detection in edge devices. Finally, we will address the reliability and security aspects of the stand-alone accelerator implementation and the project use cases.

cs.AR

Validation, Verification, and Testing (VVT) of future RISC-V powered cloud infrastructures: the Vitamin-V Horizon Europe Project perspective

Vitamin-V is a project funded under the Horizon Europe program for the period 2023-2025. The project aims to create a complete open-source software stack for RISC-V that can be used for cloud services. This software stack is intended to have the same level of performance as the x86 architecture, which is currently dominant in the cloud computing industry. In addition, the project aims to create a powerful virtual execution environment that can be used for software development, validation, verification, and testing. The virtual environment will consider the relevant RISC-V ISA extensions required for cloud deployment. Commercial cloud systems use hardware features currently unavailable in RISC-V virtual environments, including virtualization, cryptography, and vectorization. To address this, Vitamin-V will support these features in three virtual environments: QEMU, gem5, and cloud-FPGA prototype platforms. The project will focus on providing support for EPI-based RISC-V designs for both the main CPUs and cloud-important accelerators, such as memory compression. The project will add the compiler (LLVM-based) and toolchain support for the ISA extensions. Moreover, Vitamin-V will develop novel approaches for validating, verifying, and testing software trustworthiness. This paper focuses on the plans and visions that the Vitamin-V project has to support validation, verification, and testing for cloud applications, particularly emphasizing the hardware support that will be provided.

cs.DC

An Automotive Case Study on the Limits of Approximation for Object Detection

The accuracy of camera-based object detection (CBOD) built upon deep learning is often evaluated against the real objects in frames only. However, such simplistic evaluation ignores the fact that many unimportant objects are small, distant, or background, and hence, their misdetections have less impact than those for closer, larger, and foreground objects in domains such as autonomous driving. Moreover, sporadic misdetections are irrelevant since confidence on detections is typically averaged across consecutive frames, and detection devices (e.g. cameras, LiDARs) are often redundant, thus providing fault tolerance. This paper exploits such intrinsic fault tolerance of the CBOD process, and assesses in an automotive case study to what extent CBOD can tolerate approximation coming from multiple sources such as lower precision arithmetic, approximate arithmetic units, and even random faults due to, for instance, low voltage operation. We show that the accuracy impact of those sources of approximation is within 1% of the baseline even when considering the three approximate domains simultaneously, and hence, multiple sources of approximation can be exploited to build highly efficient accelerators for CBOD in cars.

cs.AR

A possible surviving companion of the SN Ia in the Galactic SNR G272.2-3.2

We use the {\it Gaia} EDR3 to explore the Galactic supernova remnant SNR G272.2-3.2, produced by the explosion of a Type Ia supernova (SNIa), about 7,500 years ago, to search for a surviving companion. From the abundances in the SNR ejecta, G272.2-3.2 is a normal SN Ia. The {\it Gaia} parallaxes allow to select the stars located within the estimated distance range of the SNR, and the {\it Gaia} proper motions to study their kinematics. From the {\it Gaia} EDR3 photometry, we construct the HR diagram of the selected sample, which we compare with the theoretical predictions for the evolution of possible star companions of SNIa. We can discard several proposed types of companions by combining kinematics and photometry. We can also discard hypervelocity stars. We focus our study on the kinematically most peculiar star, {\it Gaia} EDR3 5323900215411075328 (hereafter MV-G272), a 8.9 $\sigma$ outlier in proper motion. It is of M1-M2 stellar type. Its trajectory on the sky locates it at the center of the SNR, 6,000--8,000 years ago, a unique characteristic among the the sample. Spectra allow a stellar parameters determination and a chemical abundance analysis. In conclusion, we have a candidate to be the surviving companion of the SN Ia that resulted in SNR G272.2-3.2. It is supported by its kinematical characteristics and its trajectory within the SNR. This opens the possibility of a single-degenerate scenario for a SN Ia with an M-type dwarf companion.

astro-ph.SR

The RECIPE Approach to Challenges in Deeply Heterogeneous High Performance Systems

RECIPE (REliable power and time-ConstraInts-aware Predictive management of heterogeneous Exascale systems) is a recently started project funded within the H2020 FETHPC programme, which is expressly targeted at exploring new High-Performance Computing (HPC) technologies. RECIPE aims at introducing a hierarchical runtime resource management infrastructure to optimize energy efficiency and minimize the occurrence of thermal hotspots, while enforcing the time constraints imposed by the applications and ensuring reliability for both time-critical and throughput-oriented computation that run on deeply heterogeneous accelerator-based systems. This paper presents a detailed overview of RECIPE, identifying the fundamental challenges as well as the key innovations addressed by the project. In particular, the need for predictive reliability approaches to maximize hardware lifetime and guarantee application performance is identified as the key concern for RECIPE, and is addressed via hierarchical resource management of the heterogeneous architectural components of the system, driven by estimates of the application latency and hardware reliability obtained respectively through timing analysis and modelling thermal properties, mean-time-to-failure of subsystems. We show the impact of prediction accuracy on the overheads imposed by the checkpointing policy, as well as a possible application to a weather forecasting use case.

cs.DC

Tycho's supernova: the view from {\it Gaia}

SN 1572 (Tycho Brahe's supernova) clearly belongs to the Ia (thermonuclear) type. It was produced by the explosion of a white dwarf in a binary system. Its remnant has been the first of this type to be explored in search of a possible surviving companion, the mass donor that brought the white dwarf to the point of explosion. A high peculiar motion with respect to the stars at the same location in the Galaxy, mainly due to the orbital velocity at the time of the explosion, is a basic criterion for the detection of such companions. Radial velocities from the spectra of the stars close to the geometrical center of Tycho's supernova remnant, plus proper motions of the same stars, obtained by astrometry with the {\it Hubble Space Telescope}, have been used so far. In addition, a detailed chemical analysis of the atmospheres of a sample of candidate stars had been made. However, the distances to the stars, remained uncertain. Now, the Second {\it Gaia} Data Release (DR2) provides unprecedent accurate distances and new proper motions for the stars can be compared with those made from the {\it HST}. We consider the Galactic orbits that the candidate stars to SN companion would have in the future. We do this to explore kinematic peculiarity. We also locate a representative sample of candidate stars in the Toomre diagram. Using the new data, we reevaluate here the status of the candidates suggested thus far, as well as the larger sample of the stars seen in the central region of the remnant.

astro-ph.SR

Platform-Agnostic Steal-Time Measurement in a Guest Operating System

Steal time is a key performance metric for applications executed in a virtualized environment. Steal time measures the amount of time the processor is preempted by code outside the virtualized environment. This, in turn, allows to compute accurately the execution time of an application inside a virtual machine (i.e. it eliminates the time the virtual machine is suspended). Unfortunately, this metric is only available in particular scenarios in which the host and the guest OS are tightly coupled. Typical examples are the Xen hypervisor and Linux-based guest OSes. In contrast, in scenarios where the steal time is not available inside the virtualized environment, performance measurements are, most often, incorrect. In this paper, we introduce a novel and platform agnostic approach to calculate this steal time within the virtualized environment and without the cooperation of the host OS. The theoretical execution time of a deterministic microbenchmark is compared to its execution time in a virtualized environment. When factoring in the virtual machine load, this solution -as simple as it is- can compute the steal time. The preliminary results show that we are able to compute the load of the physical processor within the virtual machine with high accuracy.

cs.OS

No surviving companion in Kepler's supernova

We have surveyed Kepler's supernova remnant in search of the companion star of the explosion. We have gone as deep as 2.6 solar luminosities in the stars within 20% of the radius of the remnant. We use FLAMES at the VLT-UT2 telescope to obtain high resolution spectra of the stellar candidates selected from HST images. The resulting set of stellar parameters suggests that these stars come from a rather ordinary mixture of field stars (mostly giants). A few of the stars seem to have low [Fe/H] (< -1) and they are consistent with being metal-poor giants. The radial velocities and rotational velocities vrot sin i are very well determined. There are no fast rotating stars as vrot sin i < 20 km/s. The radial velocities from the spectra and the proper motions determined from HST images are compatible with those expected from the Besançon model of the Galaxy. The strong limits placed on luminosity suggest that this supernova could have arisen either from the core-degenerate scenario or from the double-degenerate scenario.

astro-ph.SR

The origin of the cosmic gamma-ray background in the MeV range

There has been much debate about the origin of the diffuse $γ$--ray background in the MeV range. At lower energies, AGNs and Seyfert galaxies can explain the background, but not above $\simeq$0.3 MeV. Beyond $\sim$10 MeV blazars appear to account for the flux observed. That leaves an unexplained gap for which different candidates have been proposed, including annihilations of WIMPS. One candidate are Type Ia supernovae (SNe Ia). Early studies concluded that they were able to account for the $γ$--ray background in the gap, while later work attributed a significantly lower contribution to them. All those estimates were based on SN Ia explosion models which did not reflect the full 3D hydrodynamics of SNe Ia explosions. In addition, new measurements obtained since 2010 have provided new, direct estimates of high-z SNe Ia rates beyond $z\sim$2. We take into account these new advances to see the predicted contribution to the gamma--ray background. We use here a wide variety of explosion models and a plethora of new measurements of SNe Ia rates. SNe Ia still fall short of the observed background. Only for a fit, which would imply $\sim$150\% systematic error in detecting SNe Ia events, do the theoretical predictions approach the observed fluxes. This fit is, however, at odds at the highest redshifts with recent SN Ia rates estimates. Other astrophysical sources such as FSRQs do match the observed flux levels in the MeV regime, while SNe Ia make up to 30--50\% of the observed flux.

astro-ph.HE

No surviving evolved companions to the progenitor of supernova SN 1006

Type Ia supernovae are thought to occur as a white dwarf made of carbon and oxygen accretes sufficient mass to trigger a thermonuclear explosion$^{1}$. The accretion could occur slowly from an unevolved (main-sequence) or evolved (subgiant or giant) star$^{2,3}$, that being dubbed the single-degenerate channel, or rapidly as it breaks up a smaller orbiting white dwarf (the double- degenerate channel)$^{3,4}$. Obviously, a companion will survive the explosion only in the single-degenerate channel$^{5}$. Both channels might contribute to the production of type Ia supernovae$^{6,7}$ but their relative proportions still remain a fundamental puzzle in astronomy. Previous searches for remnant companions have revealed one possible case for SN 1572$^{8,9}$, though that has been criticized$^{10}$. More recently, observations have restricted surviving companions to be small, main-sequence stars$^{11,12,13}$, ruling out giant companions, though still allowing the single-degenerate channel. Here we report the result of a search for surviving companions to the progenitor of SN 1006$^{14}$. None of the stars within 4' of the apparent site of the explosion is associated with the supernova remnant, so we can firmly exclude all giant and subgiant companions to the progenitor. Combined with the previous results, less than 20 per cent of type Iae occur through the single degenerate channel.

astro-ph.GA

Evidence for Type Ia Supernova Diversity from Ultraviolet Observations with the Hubble Space Telescope

We present ultraviolet (UV) spectroscopy and photometry of four Type Ia supernovae (SNe 2004dt, 2004ef, 2005M, and 2005cf) obtained with the UV prism of the Advanced Camera for Surveys on the Hubble Space Telescope. This dataset provides unique spectral time series down to 2000 Angstrom. Significant diversity is seen in the near maximum-light spectra (~ 2000--3500 Angstrom) for this small sample. The corresponding photometric data, together with archival data from Swift Ultraviolet/Optical Telescope observations, provide further evidence of increased dispersion in the UV emission with respect to the optical. The peak luminosities measured in uvw1/F250W are found to correlate with the B-band light-curve shape parameter dm15(B), but with much larger scatter relative to the correlation in the broad-band B band (e.g., ~0.4 mag versus ~0.2 mag for those with 0.8 < dm15 < 1.7 mag). SN 2004dt is found as an outlier of this correlation (at > 3 sigma), being brighter than normal SNe Ia such as SN 2005cf by ~0.9 mag and ~2.0 mag in the uvw1/F250W and uvm2/F220W filters, respectively. We show that different progenitor metallicity or line-expansion velocities alone cannot explain such a large discrepancy. Viewing-angle effects, such as due to an asymmetric explosion, may have a significant influence on the flux emitted in the UV region. Detailed modeling is needed to disentangle and quantify the above effects.

astro-ph.HE

The binary progenitor of Tycho Brahe's 1572 supernova

The brightness of type Ia supernovae, and their homogeneity as a class, makes them powerful tools in cosmology, yet little is known about the progenitor systems of these explosions. They are thought to arise when a white dwarf accretes matter from a companion star, is compressed and undergoes a thermonuclear explosion Unless the companion star is another white dwarf (in which case it should be destroyed by the mass-transfer process itself), it should survive and show distinguishing properties. Tycho's supernova is one of the only two type Ia supernovae observed in our Galaxy, and so provides an opportunity to address observationally the identification of the surviving companion. Here we report a survey of the central region of its remnant, around the position of the explosion, which excludes red giants as the mass donor of the exploding white dwarf. We found a type G0--G2 star, similar to our Sun in surface temperature and luminosity (but lower surface gravity), moving at more than three times the mean velocity of the stars at that distance, which appears to be the surviving companion of the supernova.

astro-ph

Galactic Cosmic Rays from Superbubbles and the Abundances of Lithium, Beryllium, and Boron

In this article we study the galactic evolution of the LiBeB elements within the framework of a detailed model of the chemical evolution of the Galaxy that includes galactic cosmic ray nucleosynthesis by particles accelerated in superbubbles. The chemical composition of the superbubble consists of varying proportions of ISM and freshly supernova synthesized material. The observational trends of 6 LiBeB evolution are nicely reproduced by models in which GCR come from a mixture of 25% of supernova material with 75% of ISM, except for 6 Li, for which maybe an extra source is required at low metallicities. To account for 7 Li evolution several additional sources have been considered (neutrino-induced nucleosynthesis, nova outbursts, C-stars). The model fulfills the energetic requirements for GCR acceleration.

astro-ph

Reclassification of gamma-ray bursts

We have applied two different automatic classifier algorithms to the BATSE Current GRB Catalog data and we obtain three different classes of GRBs. Our results confirm the existence of a third, intermediate class of GRBs, with mean duration \sim 25-50 s, as deduced from a cluster analysis and from a neural network algorithm. Our analyses imply longer durations than those found by Mukherjee et al. (1998) and Horvath (1998), whose intermediate class had durations \sim 2-10 s. From the neural network analysis no difference in hardness between the two longest classes is found, and from both methods we find that the intermediate-duration class constitutes the most homogeneous sample of GRBs in its space distribution while the longest-duration class constitutes the most inhomogeneous one with \sim 0.1, being thus the deepest population of GRBs with z_max \sim 10. The trend previously found in long bursts, of spatial inhomogeneity increasing with hardness, only holds for this new longest-duration class.

astro-ph