AES访问驱动Cache计时攻击

首先给出了访问驱动Cache计时攻击的模型,提出了该模型下直接分析、排除分析两种通用的AES加密泄漏Cache信息分析方法;然后建立了AES加密Cache信息泄露模型,并在此基础上对排除分析攻击所需样本量进行了定量分析,给出了攻击中可能遇到问题的解决方案;最后结合OpenSSL v.0.9.8a,v.0.9.8j中两种...     

Ternary-Responsive Field-Effect Transistors and Multilevel Memories Based on Asymmetrically Functionalized Janus Few-Layer WSe2

Hybrids composed of 2D transition metal dichalcogenides with stimuli-responsive molecules are prototypical components for the development of multifunctional field-effect transistors (FETs), whose output currents can be remotely controlled by external inputs. Herein, ternary-responsive FETs based on a few-layer WSe₂ are realized by decorating the two opposite surfaces of the 2D semiconductor with different stimuli-responsive molecules in an asymmetric fashion: the bottom surface is interfaced with a photochromic diarylethene film and the top surface with a ferroelectric poly(vinylidene fluoride–trifluoroethylene) layer. This novel Janus ternary device architecture shows superior functional complexity compared with normal mono-stimuli-responsive FETs. The synergy between the two molecularly induced effects enables the devices to respond orthogonally to an electric field and light irradiation, with an enhanced output current modulation efficiency of 87%. The 9 ferroelectric and 84 photo-generated states ensure 756 current levels in a single device. The over 10 cycles of cyclic endurance and more than 1000 h of retention time confirm the reliability of each state, implementing the demand for high-density non-volatile memories, as well as enriching the diversification in “More than Moore” technologies.     

基于“嵩山”超级计算机系统的大规模管网仿真

供水管网仿真广泛应用于城市供水输配调度,是城市供水管网监测与维护的重要技术手段。由于在面向城市级的大规模管网中产生了海量的计算数据,因此在一般计算平台上无法满足管网仿真计算的算力需求。为提升城市级供水管网仿真的计算效率,提出一种有效的并行化方案。基于“嵩山”超级计算机系统采用中央处理器+数据缓存单元（CPU+DCU）架构,利用其在密集数据计算方面的优势,对“嵩山”超级计算机进行供水管网仿真。参照可移植性异构计算接口（HIP）异构编程模型,在“嵩山”超级计算机上实现供水管网仿真的异构计算,并结合管道数据分割方案,使用消息传递接口开启多进程以实现DCU加速数据通信传递。通过重定义数据类型解决计算过程中结构体传输问题,实现单节点内多DCU的大规模密集计算。在不同计算平台和多种计算策略仿真上的对比结果表明,与传统x86平台相比,该优化方案在小规模数据与大规模数据上的加速比分别达到5.269、10.760,与采用计算统一设备架构异构编程模型的传统GPU异构平台相比,计算性能有明显提高。     

Static Coarse Grain Task Scheduling with Cache Optimization Using OpenMP

Effective use of cache memory is getting more important with increasing gap between the processor speed and memory access speed. Also, use of multigrain parallelism is getting more important to improve effective performance beyond the limitation of loop iteration level parallelism. Considering these factors, this paper proposes a coarse grain task static scheduling scheme considering cache optimization. The proposed scheme schedules coarse grain tasks to threads so that shared data among coarse grain tasks can be passed via cache after task and data decomposition considering cache size at compile time. It is implemented on OSCAR Fortran multigrain parallelizing compiler and evaluated on Sun Ultra80 four-processor SMP workstation using Swim and Tomcatv from the SPEC fp 95. As the results, the proposed scheme gives us 4.56 times speedup for Swim and 2.37 times on 4 processors for Tomcatv respectively against the Sun Forte HPC Ver. 6 update 1 loop parallelizing compiler.     

SONOS非易失性存储器件研究进展

随着半导体存储器件的小型化、微型化,传统多晶硅浮栅存储因为叠层厚度过大,对隧穿氧化层绝缘性要求过高而难以适应未来存储器的发展要求。最近,基于绝缘性能优异的氮化硅的SONOS非易失性存储器件,以其相对于传统多晶硅浮栅存储器更强的电荷存储能力、易于实现小型化和工艺简单等特性而重新受到重视。文章论述了SONOS非易失性存储器件的存储原理和存储性能的影响因素研究进展,并在材料、工艺与结构设计等方面对SONOS存储器件性能改进的研究进展情况进行了分析和讨论。     

主存技术与时态数据库

多年来，时态数据库的研究与实现受到内存资源的制约。硬件技术的进步为数据库管理系统提供了海量内存。本文提出把时态数据库的全部或大部分数据置于主存中，并能在不读写磁盘的情况下连续作多个事务，称之为时态主存数据库（ＴｅｍｐｏｒａｌＭａｉｎＭｅｍｏｒｙＤＢ，以下简称ＴＭＭＤＢ）。该文讨论了ＴＭＭＤＢ的功能、标准、内存布局、与大缓存ＤＢＭＳ的本质差异，以及由“时态”和“主存”技术结合而引起的一系列特殊问题。     

Entrepreneurship in Architectural Robotics: The Simultaneity of Craft,Economics and Design

What do Odico Formwork Robotics, RoboFold, Machineous, ROB Technologies and GREYSHED share in common? They are all architectural robotics startups. Jelle Feringa , Chief Technology Officer at Odico, places the phenomenon of the architectural robotics entrepreneur in a historical and cultural context while highlighting the very practical role startups are poised to play in bridging the gap between academic research and industry, by providing the building industry with much needed new software tools.     

基于Gnutella的P2P搜索改进算法的研究

在对洪泛算法和现有基于索引机制的资源定位策略进行研究基础上,对现有的路由算法进行改进:通过在节点增加缓存的方式来记录查询返回消息以提高重复查询的效率;通过对特定查询消息的生命周期值的修改来保证指引查询时查询消息的可达性.     

AVS解码器在DSP平台上的优化

为在嵌入式DSP平台上提高AVS解码器的运行效率,提出一种基于L1P cache的优化方法。将程序代码划分为4个模块,每个模块容量均小于L1P的容量;并改变传统的针对单个宏块进行顺序处理的模式,使每个程序模块依次处理一个宏块行的数据,从而减少4×M×(N-1)次代码冲刷。实验结果表明,基于L1P Cache的方法提高了处理器读取程序代码的效率,使解码器速度提高20%。     

Preparation of PLZT Ferroelectric Films by RF Sputtering on 200 mmφ Substrate

(Pb, La)(Zr, Ti) O 3 (PLZT) thin films were deposited on 200mm } Pt/Ti/SiO2/Si substrates by RF magnetron Sputtering using multichamber production system. The Pb content in PLZT films deposited at low temperature was measured by X-ray fluorescence spectroscopy (XRF), and ferroelectrics properties were measured. Good uniformities of Pb content and deposition rate were achieved on 200mm } substrate. For ferroelectrics properties, only small deference was observed between the center and the edge of 200mm } substrate.