Abstract: A 3D woven composite, designed for automotive applications, is studied using Abaqus Finite Element (FE) software suite. Python scripts were developed to build FE models of the woven composite in Complete Abaqus Environment (CAE). They can read TexGen or WiseTex files and automatically generate consistent meshes of the fabric and the matrix. A user menu is provided to help define parameters for the FE models, such as type and size of the elements in fabric and matrix as well as the type of matrix-fabric interaction. Node-to-node constraints were imposed to guarantee periodicity of the deformed shapes at the boundaries of the representative volume element of the composite. Tensile loads in three axes and biaxial loads in x-y directions have been applied at different Fibre Volume Fractions (FVFs). A simple damage model was implemented via an Abaqus user material (UMAT) subroutine. Existing tools for homogenization were also used, including voxel mesh generation from TexGen as well as Abaqus Micromechanics plugin. Linear relations between homogenised elastic properties and the FVFs are given. The FE models of composite exhibited balanced behaviour with respect to warp and weft directions in terms of both stiffness and strength.
Abstract: Dynamic Voltage and Frequency Scaling (DVFS)
multicore platforms are promising execution platforms that enable
high computational performance, less energy consumption and
flexibility in scheduling the system processes. However, the
resulting interleaving and memory interference together with per-core
frequency tuning make real-time guarantees hard to be delivered.
Besides, energy consumption represents a strong constraint for the
deployment of such systems on energy-limited settings. Identifying
the system configurations that would achieve a high performance and
consume less energy while guaranteeing the system schedulability is
a complex task in the design of modern embedded systems. This work
studies the trade-off between energy consumption, cores utilization
and memory bottleneck and their impact on the schedulability of
DVFS multicore time-critical systems with a hierarchy of shared
memories. We build a model-based framework using Parametrized
Timed Automata of UPPAAL to analyze the mutual impact of
performance, energy consumption and schedulability of DVFS
multicore systems, and demonstrate the trade-off on an actual case
study.
Abstract: Power management techniques are necessary to save power in the microprocessor. By changing the frequency and/or operating voltage of processor, DVFS can control power consumption. In this paper, we perform a case study to find optimal power state transition for DVFS. We propose the equation to find the optimal ratio between executions of states while taking into account the deadline of processing time and the power state transition delay overhead. The experiment is performed on the Cortex-M4 processor, and average 6.5% power saving is observed when DVFS is applied under the deadline condition.
Abstract: While the feature sizes of recent Complementary Metal
Oxid Semiconductor (CMOS) devices decrease the influence of static
power prevails their energy consumption. Thus, power savings that
benefit from Dynamic Frequency and Voltage Scaling (DVFS) are
diminishing and temporal shutdown of cores or other microchip
components become more worthwhile. A consequence of powering off unused parts of a chip is that the
relative difference between idle and fully loaded power consumption
is increased. That means, future chips and whole server systems gain
more power saving potential through power-aware load balancing,
whereas in former times this power saving approach had only
limited effect, and thus, was not widely adopted. While powering
off complete servers was used to save energy, it will be superfluous
in many cases when cores can be powered down. An important
advantage that comes with that is a largely reduced time to respond
to increased computational demand. We include the above developments in a server power model
and quantify the advantage. Our conclusion is that strategies from
datacenters when to power off server systems might be used in the
future on core level, while load balancing mechanisms previously
used at core level might be used in the future at server level.
Abstract: An adaptive Chinese hand-talking system is presented
in this paper. By analyzing the 3 data collecting strategies for new
users, the adaptation framework including supervised and unsupervised
adaptation methods is proposed. For supervised adaptation,
affinity propagation (AP) is used to extract exemplar subsets, and enhanced
maximum a posteriori / vector field smoothing (eMAP/VFS)
is proposed to pool the adaptation data among different models. For
unsupervised adaptation, polynomial segment models (PSMs) are
used to help hidden Markov models (HMMs) to accurately label
the unlabeled data, then the "labeled" data together with signerindependent
models are inputted to MAP algorithm to generate
signer-adapted models. Experimental results show that the proposed
framework can execute both supervised adaptation with small amount
of labeled data and unsupervised adaptation with large amount
of unlabeled data to tailor the original models, and both achieve
improvements on the performance of recognition rate.
Abstract: I/O workload is a critical and important factor to
analyze I/O pattern and file system performance. However tracing I/O
operations on the fly distributed parallel file system is non-trivial due
to collection overhead and a large volume of data. In this paper, we
design and implement a parallel file system logging method for high
performance computing using shared memory-based multi-layer
scheme. It minimizes the overhead with reduced logging operation
response time and provides efficient post-processing scheme through
shared memory. Separated logging server can collect sequential logs
from multiple clients in a cluster through packet communication.
Implementation and evaluation result shows low overhead and high
scalability of this architecture for high performance parallel logging
analysis.
Abstract: This paper presents an architecture of current filesystem
implementations as well as our new filesystem SpadFS and operating
system Spad with rewritten VFS layer targeted at high performance
I/O applications. The paper presents microbenchmarks and real-world
benchmarks of different filesystems on the same kernel as well as
benchmarks of the same filesystem on different kernels – enabling
the reader to make conclusion how much is the performance of
various tasks affected by operating system and how much by physical
layout of data on disk. The paper describes our novel features–most
notably continuous allocation of directories and cross-file readahead
– and shows their impact on performance.