Multi-resource scheduling for FPGA systems

In modern cloud data centers, reconfigurable devices (FPGAs) are used as an alternative to Graphics Processing Units to accelerate data-intensive computations (e.g., machine learning, image and signal processing). Currently, FPGAs are configured to execute fixed workloads, repeatedly over long periods of time. This conflicts with the needs, proper to cloud computing, to flexibly allocate different workloads and to offer the use of physical devices to multiple users. This raises the need for novel, efficient FPGA scheduling algorithms that can decide execution orders close to the optimum in a short time. In this context, we propose a novel scheduling heuristic where groups of tasks that execute together are interposed by hardware reconfigurations. Our contribution is based on gathering tasks around a high-latency task that hides the latency of tasks, within the same group, that run in parallel and have shorter latencies. We evaluated our solution on a benchmark of 37500 random workloads, synthesized from realistic designs (i.e., topology, resource occupancy). For this testbench, on average, our heuristic produces optimum makespan solutions in 47.4% of the cases. It produces acceptable solutions for moderately constrained systems (i.e., the deadline falls within 10% of the optimum makespan) in 90.1% of the cases.     

Cokriging-Based Sequential Design Strategies Using Fast Cross-Validation Techniques for Multi-Fidelity Computer Codes

Cokriging-based surrogate models have become popular in recent decades to approximate a computer code output from a few simulations using both coarse and more complex versions of the code. In practical applications, it is common to sequentially add new simulations to obtain more accurate approximations. We propose a method of cokriging-based sequential design, which combines both the error evaluation provided by the cokriging model and the observed errors of a leave-one-out cross-validation procedure. This method is proposed in two versions, the first one selects points one at a time. The second one allows us to parallelize the simulations and to add several design points at a time. The main advantage of the suggested strategies is that at a new design point they choose which code versions should be simulated (i.e., the complex code or one of its fast approximations). A multifidelity application is used to illustrate the efficiency of the proposed approaches. In this example, the accurate code is a two-dimensional finite element model and the less accurate one is a one-dimensional approximation of the system. This article has supplementary material online.     

More than bin packing: Dynamic resource allocation strategies in cloud data centers

Resource allocation strategies in virtualized data centers have received considerable attention recently as they can have substantial impact on the energy efficiency of a data center. This led to new decision and control strategies with significant managerial impact for IT service providers. We focus on dynamic environments where virtual machines need to be allocated and deallocated to servers over time. Simple bin packing heuristics have been analyzed and used to place virtual machines upon arrival. However, these placement heuristics can lead to suboptimal server utilization, because they cannot consider virtual machines, which arrive in the future. We ran extensive lab experiments and simulations with different controllers and different workloads to understand which control strategies achieve high levels of energy efficiency in different workload environments. We found that combinations of placement controllers and periodic reallocations achieve the highest energy efficiency subject to predefined service levels. While the type of placement heuristic had little impact on the average server demand, the type of virtual machine resource demand estimator used for the placement decisions had a significant impact on the overall energy efficiency.     

vMES: Virtualization aware manufacturing execution system

The large scale emergence in the last decade of various cloud solutions, ranging from software-as-a-service (SaaS) based solutions for business process management and implementation to very sophisticated private cloud solutions capable of high performance computing (HPC) and efficient virtualization, constitute the building blocks for engineering the next generation of flexible enterprise systems that can respond with great agility to changes in their environment. These new technologies are adopted at a certain level by manufacturing enterprises in order to advance in a new era of mass customization where flexibility, scalability and agility are the differentiating factors. In this context, this paper introduces the virtualized manufacturing execution system (vMES), an intermediate layer in the manufacturing stack, and discusses the advantages and limitations offered by this approach for manufacturing enterprises. A classification of MES workloads based on the ISA-95 function model is presented, focusing on the virtualization techniques suitable for each workload, considering the algorithms and technologies used and the virtualization overhead. A pilot vMES implementation using a parallel process for smart resource provisioning and automatic scaling is also presented. The pilot implementation using six Adept robots and one IBM CloudBurst 2.1 private cloud and an ISA-95 based MES is described; the virtualization sequence is analyzed in several scenarios of resource workload collocation on physical cloud blades with and without perturbations.     

网络系统资源数据的脆弱性漏洞防御控制仿真

当前漏洞防御控制方法存在着无法准确区分数据类型或脆弱性,需要较高的能量消耗和难以抵御某些未知攻击等不足。针对这些不足,提出网络系统资源数据脆弱性漏洞防御控制方法。根据属性划分网络系统运行数据和历史资源,属性划分后根据漏洞特征相关参数构建强关联规则,结合脆弱性漏洞辨识矩阵和数据关联规则构建漏洞检测模型;计算脆弱性漏洞利用难度,将防御任务优选问题描述为防御成本不高于给定值的条件下,获取造成最小损失的防御措施集的过程。把防御任务优选多目标优化问题转换成单目标优化问题,采用改进的蚁群算法求解优化问题,生成最优防御控制方案。实验结果表明,所提方法可以有效防御已知数据攻击和大部分未知数据攻击,且所需能耗较低。     

Potenziale und Grenzen der Sekundärrohstoffgewinnung – Ergebnisse der r4-Begleitforschung

The socio-economic and ecological impacts of selected projects of the BMBF funding program “r⁴ – Innovative Technologies for Resource Efficiency – Research for the Provision of Raw Materials of Strategic Economic Importance” are presented. Many, but not all, research projects indicate a potential improvement of the supply situation in Germany. In some cases, the provision of secondary raw materials is unprofitable or ecologically detrimental. These cases require a balancing between security of supply and other economic and ecological objectives.     

印制电路板钻铣工序金属回收的探讨

本文探讨了印制电路板生产中钻、铣和磨刷工序产生铜粉回收技术,包括分离装置和原理。     

Information Technology–Supported value Co-Creation and Co-Destruction via social interaction and resource integration in service systems

The paper develops a conceptual framework to study the constructs of information technology (IT)-supported value co-creation and co-destruction through shared processes of social interaction and resource integration as mediated by IT and embedded in interactive value formation practices. In particular, the elements of social interaction and resource integration are identified, and the role of IT is discussed. We conduct a systematic literature review and analyze the data using the service system perspective. Our research contributes by identifying the constructs of value-creating phenomena enabled by IT from the service system perspective and presenting a research agenda for further studies.     

资源循环利用对有色金属工业发展的影响

资源循环利用是把全社会已经使用过的物品、边角料、废弃物作为一种资源,经过技术处理重新服务于人类。而循环经济则是一种新的经济模型。资源循环利用对有色金属工业可持续发展意义重大,值得深入研究。本文对有色金属资源循环的必要性及资源循环型社会建立的条件进行了初步分析,以期引起人们的思考。     

地下汽车库通风排烟系统的节能设计

分析了变频技术对风机调节的影响，指出地下汽车库通风排烟系统采用定风量设计不利于系统节能，该系统宜采用变风量设计，同时介绍了变频技术在地下汽车库通风排烟系统中的应用。