报告包括电子商务方面的重要观点、政策与环境、网购平台、行业应用、电商物流、电商金融等研究。

报告包括电子商务方面的重要观点、政策与环境、网购平台、行业应用、电商物流、电商金融等研究。

报告包括电子商务方面的重要观点、政策与环境、网购平台、行业应用、电商物流、电商金融等研究。

Data centers are used to house an enterprise IT infrastructure. They constitute the backbone for essential business operations. Demand for cloud-based services and big data analytics are growing among enterprises in Southeast Asia. Several industry verticals in this region are making use of advanced technologies to gain competitive advantage. The complexity associated with business applications has increased because of enormous growth in data volumes. This has triggered an increased need for construction and renovation of data centers in Southeast Asia. Enterprises in this region are also focused on construction of green data center facilities to reduce power consumption and environmental impact. Construction of a data center includes design, architecture, installation of electrical and mechanical systems, and incorporation of security mechanisms. Infrastructure certifications also play a vital role in the building of data center facilities.

Encryption is a technique that is used to protect and secure data during the transfer of plain text information between two users in a cipher format through the internet. The process uses mathematical algorithms that convert information into an undecipherable format that can only be unlocked and accessed with the use of secret encrypted codes. The strength of an encryption code is measured by its length, which is determined by the number of bits and by the type of encryption program. A higher number of bits produces more possible codes or combinations, making a code more secure for a user. Though encryption software is mainly used to encrypt and decrypt data in the form of files on storage devices and protect the files, it can also include features such as file shredding.

Cloud-based video conferencing or VCaaS allows two or more people in different locations to communicate through simultaneous two-way audio and video transmissions over the cloud. A cloud server is used to transfer, access, and store data. Video conferencing is different from a video call as video conferencing serves multiple locations rather than individuals. VCaaS provides organizations with HD video conferencing without any major investment in hardware, infrastructure, or network. In addition, it makes video transmissions easily accessible to anyone at any point of time through devices such as PCs and laptops.

Change and configuration management is an essential part of the product development cycle. The effective management of changes to a product enables an enterprise to get an in-depth understanding of market requirements and assess customers' needs. It also enables enterprises to foster innovative ideas and develop advanced products. Some basic changes are acceptable at the early stage of the design cycle, but as the cycle matures, the changes become more complex and lead to high costs and increase the product development cycle. Change and configuration management software help organizations control the nature of the proposed changes (informal or formal), as well as their analysis, implementation, and release throughout the product cycle.

针对步态识别中步态特征提取高维处理的复杂性,在研究压缩感知理论的基础上,提出将压缩感知理论应用于步态识别中的步态特征提取方面。在充分利用步态图像稀疏性的前提下,利用观测矩阵对步态图像进行投影观测,得到的观测值作为步态特征用于步态识别中,实现了特征提取的降维处理,大大降低了计算的复杂性。在步态图像的重构方面,在压缩采样匹配追踪(CoSaMP)的基础上,提出了基于小波树模型的压缩采样匹配(wavelet model-CoSaMP,WM-CoSaMP)的重构算法,进一步提高了重构精度。通过对比实验,验证了WM-CoSaMP重构算法的优越性,以及压缩感知在步态特征提取方面的优越性。

RFID技术大大提高了库存管理、目标跟踪、供应链管理等诸多领域的工作效率。在这些应用中,经常需要将新的对象加入到系统中,现有对象经常被放错区域。当这些情况发生时,很有必要把这些标签快速、完整地检测出来。当前的检测技术并不能保证收集到所有未知标签,针对这一问题,给出了一种高效协议,解决了如何在不发射标签标志情况下关闭已知标签,通过比较已知标签预期回答与实际测得回答,阅读器实现已知标签和未知标签的检测。据此,提出了一种无冲突时隙配对技术及多散列时隙选择技术,帮助标签选择最佳时隙与阅读器通信,有效解决了未知标签和已知标签检测间的冲突,显著提高了检测效率。模拟实验结果表明,该算法性能优异,与收集系统中所有标签身份的基准算法相比,最优协议的运行时间平均降低了63%,降低幅度最多可达85%。

针对传统的知识推荐算法存在用户冷启动和冷门物品推荐的问题,提出了一种基于三部图网络结构的知识推荐算法。在计算相似度时引入网络结构中的度,综合考虑项目的度和权值及标签的度和权值对推荐算法的影响。实验结果表明,该算法提高了推荐的个性化和多样性,有效地解决了用户冷启动和冷门物品推荐的问题,改善了推荐效果。

With our framework, efficiency programmers write and/or generate customized OpenCL snippets at runtime and the framework automatically fuses, compiles, and executes these operations based on a Python description.We evaluate the framework with case studies of two very different applications: space-time adaptive radar processing and optical flow. For a space-time adaptive radar pro cessing application, our framework’s implementation is competitive with a hand-coded implementation that uses a vendor-optimized library. For optical flow, a computer vision application,the framework achieves frame rates that are between 0.5×and 0.97× hand-coded OpenCL performance.

In this thesis, we begin by analyzing the increasing trend of running large-s cale services on Warehouse Scale Computers to provide user functionality. We explain why current networks are poorly suited for delivering consistent, low-latency p erformance. To understand the b enefits of foregoing generality, we design and implement two approaches DeTail and FastLane.To evaluate these approaches, we employ both simulation and implementation platforms.