Search results

Monitoring a microservice system may bring a lot of benefits to development teams such as early detection of run-time errors and various performance anomalies. In this study, we explore deep learning (DL) solutions for detection of anomalous system's behavior based on collected monitoring data that consists of applications' and systems' performance metrics. The study is conducted in a collaboratio

We propose a depth cache that keeps the depth data in compressed format, when possible. Compared to previous work, this requires a more flexible cache implementation, where a tile may occupy a variable number of cache lines depending on whether it can be compressed or not. The advantage of this is that the effective cache size increases proportionally to the compression ratio. We show that the dep

We present a novel architecture for flexible control of shading rates in a GPU pipeline, and demonstrate substantially reduced shading costs for various applications. We decouple shading and visibility by restricting and quantizing shading rates to a finite set of screen-aligned grids, leading to simpler and fewer changes to the GPU pipeline compared to alternative approaches. Our architecture int

Dataflow is a natural way of modelling streaming applications, such as multimedia, networking and other signal processing applications. In order to cope with the computational and parallelism demands of such streaming applications, multiprocessor systems are replacing uniprocessor systems. Mapping and scheduling these applications on multiprocessor systems are crucial elements for efficient implem

We study the joint design of optimal linear MIMO encoders and decoders for filtering and transmission of a vector- valued signal over parallel Gaussian channels subject to a real- time constraint. The objective is to minimize the sum of the estimation error variances at the receiving end. The design problem is nonconvex, but it is shown that a global optimum can be found by solving a related two-s

This paper presents a dataflow design methodology and an associated co-exploration environment, focusing on the optimization of buffer sizes. The approach is applicable to dynamic dataflow designs and its performance is presented and validated by experimental results on the porting of an MPEG-4 Simple Profile decoder to the STM STHORM manycore platform. For the purpose of this work, the decoder ha

Model-based analysis of in-cylinder pressure sensor signals has been a key component for internal combustion engine research, diagnostics and controller development during the past decades. This analysis is often based on simple thermodynamic models of the in-cylinder processes. In order for the analysis to give accurate results, the models need to be sufficiently calibrated. This paper investigat

Whatever their size and the ecosystem they live in, all organisms may disperse at some stage of their life cycle. Dispersal dynamics are to a varying extent dependent on organismal size, life history, ecological niche, survival capacities and phylogeny. Moves towards a synthesis in dispersal ecology have focused primarily on vertebrates and higher plants, yet recent studies suggest that the disper

We present an algorithm, called Bonsai, for rapidly building bounding volume hierarchies for ray tracing. Our method starts by computing midpoints of the triangle bounding boxes and then performs a rough hierarchical top-down split using the midpoints, creating triangle groups with tight bounding boxes. For each triangle group, a mini tree is built using an improved sweep SAH method. Once all mini

Custom architectures are often adopted as more efficient alternatives to general purpose processors in terms of performance and power. However, the design of such architectures requires experts both in hardware and the application domain. In this paper we propose a method for speeding up the design space exploration. Our method, based on Pareto points, identifies sets of solutions in terms of scal

Using ultralow-power wake-up receivers (WRxs) can reduce idle listening energy cost in wireless sensor networks with low traffic intensity. This has led to many WRx analog front-end (AFE) designs presented in literature, with a large variety of trade-offs between the sensitivity, the data rate, and the power consumption. Energy consumed during wake-up in a network depends on many parameters and wi

A survey of faster than Nyquist results over the last 15 years, with and extension of earlier binary modulation/convolutional coding results to four-level modulation.

Domain-continuous visibility determination algorithms have proved to be very efficient at reducing noise otherwise prevalent in stochastic sampling. Even though they come with an increased overhead in terms of geometrical tests and visibility information management, their analytical nature provides such a rich integral that the pay-off is often worth it. This paper presents a time-continuous, prim

Partially Premixed Combustion (PPC) is a promising combustion concept with high thermodynamic efficiency and low emission level, and also with minimal modification of standard engine hardware. To use PPC in a production oriented engine, the optimal intake charge conditions for PPC should be included in the analysis. The experiments in this paper investigated and confirmed that the optimal intake c

The ignition delay is an important quantity in low temperature combustion concepts where a prolonged ignition delay gives an enhanced fuel-air mixing process, favorable for decreased particulate and NOx emission levels. This article investigates three different low-order physics-based correlation models and their ability to predict the ignition delay for the purpose of model-based controller desig

The computation power we expect from the various smart devices we use keeps increasing. Not only do we want faster devices but also less power hungry and energy efficient devices, both for the environment and our personal convenience (remember that "mobile phone" attached to a power plug at all times?).One way of addressing this demand is to build custom processor architectures that focus on a spe

Partially premixed combustion (PPC) is a low temperature, direct-injection combustion concept that has shown to give promising emission levels and efficiencies over a wide operating range. In this concept, high EGR ratios, high octane-number fuels and early injection timings are used to slow down the auto-ignition reactions and to enhance the fuel and are mixing before the start of combustion. A d

Abstract in UndeterminedDataflow descriptions have been used ina wide range of Digital Signal Processing (DSP)applications, such as multi-media processing, andwireless communications. Among various forms ofdataflow modeling, Synchronous Dataflow (SDF) isgeared towards static scheduling of computationalmodules, which improves system performance andpredictability. However, many DSP applicationsdo no