Search

With the growing increase of data processing and Hadoop data center construction requirements, simulations of large scale Hadoop data centers with high precision are becoming a great challenge. In this paper, a new simulator which integrates baseline test and multi-layered network model is introduced and implemented. With baseline test, we can predict precisely the execution time for each MapReduce task. The network model considers the complexity of data center network and it can provide more accurate prediction of data transfer delay. The experimental test is implemented in a large scale data center to evaluate the performance of our simulator. Three experimental environments with 35 nodes, 47 nodes and 80 nodes are configured in the data center, and Terasort, Wordcount and Hive are selected as benchmarks with maximum 100 TB input data. The experiments show that the error comparing between the simulator results and experimental environment results in most cases is less than 10%. The comparison with YARNsim confirms that our simulator is capable to achieve precise simulation for a large scale data center and the baseline test model is suitable to optimize the execution time simulation for any type of Hadoop applications.

Users' online behavior, generated from endpoints of e-commerce website and app, is regarded as big data which can create large business value by mining them to acquire insights of users' preference, inclination and purpose. A good ranking result of product search and product assortment in online category classification can lift user's click rate, increase purchasing conversion rate and improve customer online experience. In this paper, we will introduce an online product based learning to rank model to intelligently learn product ranking. A big data architecture will also be introduced to implement this learning to rank model which analyzes huge amount of users' online behavior.

Many areas in China, e.g., Beijing-Tianjin-Hebei area, have been suffering from severe air pollution problem in recent years. The haze has jeopardized people's health and aroused deep concern from the public. To understand residents concerns better and leverage the power of the crowd, the government encourages organizations and individuals to participate in environmental protection activities. In this paper, we describe a decision supporting system with a set of algorithms implemented to extract insights from fused physical data (Air Quality Index, AQI) and social data (social buzz). The system provides insights for the government to closely improve its public services to curb the air pollution, including understanding the correlation between AQI and residents sentiment, spot and remove illegal sources of air pollution.

Failure Mode and Effects Analysis (FMEA) is a well-known design tool used to analyze component failures and identify the resultant effects. Traditional FMEA model requires engineers' input on estimating technical detection difficulty of part failure. Moreover, failure occurrence can only be computed based on historical maintenance data. However, under the IoT based environment, sensors installed on connected devices can transmit continuous signals of device status, which is greatly helpful to make the device failure analytical work more efficient, more objective and more accurate. In this paper, we propose a new FMEA model based on connected device data. A new failure a difficulty creator is proposed to objectively compute detection difficulty. Failure occurrence level is intelligently generated by a new occurrence creator. A house-roof shaped failure correlation creator is designed to mine the correlation of failures. Finally, a new risk RPN estimation method is used to rank the risk of failure.

House of Quality(HoQ) is an effective way to transform customer requirement into technical solution, aiming at helping RD 2) new customer requirement importance creation method; 3) absolute importance creation method; 4) optimization model for technical mix.

Marketing is all about the process of communicating the value of a product or service to customers in order to stimulate sales, while target marketing maximizes the return on investment into the marketing activities. We present a piece of practice to transform massive data into actionable insights and guidance using statistical analysis and data mining methods, which helps a leading home appliance manufacturer optimize its marketing strategy for increasing customer intimacy, solidifying customer loyalty, and most importantly, boosting the sales volume and profit.

Enterprise resource planning (ERP) systems present a holistic view of the business from single information and IT architecture. But ERP systems are comprehensive, huge and complex systems and warrant careful planning and execution not ensure the successful implementation. A business process is set of interrelated activities which have definable inputs result in output that adds value from a custom perspective. Business Process Management (BPM) is a structured approach to understand, analyze, support, and continuously improve fundamental process. In order to improve the success rate of ERP in project implementation, enterprises often implement BPM before or during the ERP implementation. In this paper, a framework for integration ERP system into BPM system is proposed. The framework closes the gap between ERP system and BPM system, enables full life-cycle of process centric enterprise system implementation.

With fierce competition in traditional manufacturing industry, it is urgent to find an effective way to gain competitive edge for enterprises to win market share. In this paper, we introduce novel ideas on new product design and customized product marketing in order to improve enterprises competitiveness by discovering business insights from multidimensional data, including product/customer profiles, sales data, Voice of Customers (VoC), and so on. A detailed case study on one famous home appliance manufacturer in China is given. Especially, VoC is introduced into new product design and the statistical model - Analysis on Variance (ANOVA) is applied to conduct in-depth data analysis.

In Internet of things, lots of instruments and sensors will connect into Internet and can be controlled through Internet to achieve Smart Planet. One of the key challenges is to integrate instruments and sensors into business process. SOA is an ideal infrastructure for business process management and lots of business process definition languages are now available for process orchestration. Instruments functions are encapsulate to web services and can be organized with other web services equally. But these device-oriented web services are different from common web services because devices can't be controlled by more than one client at the same time. This feature can't be supported in traditional process definition languages. In this paper, a new process definition langue for Internet of Things is provided. It is an XML-based language; the structure of the language includes 3 sections: services, sequence and vars. Services section is used to describe service information. Device-oriented web services and common web services are described in this section. Sequence section is the main body of the process, which includes the specific business process of the IoT application. Sequence is used to enable executing several commands in order.

Script-based automated regression testing is widely used in industry. In this work, we focus on failed tests in a real regression test project. The causes of 197 failed tests produced in automated testing are examined and categorized based on an analysis procedure presented. The result shows that incorrect scripts, oracle mismatches, test tool bugs and misconfigurations involved in testing contribute most to failures instead of product bugs. Detecting and fixing the false positives are laborious and time-consuming. Our empirical study demonstrates that the benefits of test automation are obliterated to some extent in practical settings. To improve the effectiveness of automated regression testing further, some issues should receive more attention.

Cross-organizational services composition is process-oriented and need collaboration of several systems, and these characteristics are different from normal. Traditional services composition modeling methods cannot meet these new requirements because they don't have mechanisms to model the concurrency operations of several systems. Target on this problem, a cross-organizational services composition modeling method based on Pi calculus is proposed. Using Pi calculus, a cross-organizational business process is modeled as a composition of a set concurrent local business processes, the cross-organization controller, the global start activity and global end activity. Local business processes is modeled as a composition of local activities which can receive external start message compared with traditional activities in service composition. This new cross-organizational services composition modeling method can makes the model adapt to dynamic cross-organizational environment. And this model is based on strict formal method, which facilitates the analysis and verification of business process models.

Process model is one of the foundation stones of business process management(BPM). In a lot of industrial practices of BPM, we encounter the requirements to synchronize process models. Although there have already been many research initiatives and mature solutions for two-model synchronization, multi-model synchronization is still a difficult question. Major challenges include how to define multi-model synchronization semantics and how to propagate changes across the multiple models in order to realize incremental synchronization. This paper proposes a solution in which the modification propagation is considered as information flow. After defining synchronization semantics, we work out a propagation scope for each tracked model change. Then an algorithm is designed to generate an propagation plan for all model changes, with the objective to optimize the information flow. The multiple process models get synchronized after all modifications are propagated according to the plan. A case from an industrial practice is used to demonstrate the semantics and algorithm of incremental synchronization of multiple process models.

In the practice of process-centric systems implementations, business process modeling is becoming more crucial. This paper introduces an approach to automatically generate test suites for the process paths to check the consistency, which is very important, between the business process model and the implementation of Enterprise System. The paper also presents a supporting tool to apply this approach in SAP implementation. A large-scale enterprise transformation program is discussed as a case study.

As the rapid development of Internet of things, a large number of devices, instruments and sensors will connect and collaborate to achieve Smart Planet. One of the key challenges is to integrate devices into business process. SOA is an ideal infrastructure for business process management as applications are invoked using standard interfaces and protocols. It's convenient to use device-oriented web services (doWS) to encapsulate devices functions. A doWS may conflict with other doWS because devices can't be controlled by more than one client at the same time. This brings additional complexity to web service composition. In this paper, we propose a comprehensive device collaboration model to doWS choreography. This model includes device layer, doWS layer, compatible resource layer and process layer, where device layer represent devices, doWS layer encapsulates devices functions, compatible resource layer abstracts compatible device operations as resource and avoid of conflict, process layer describes the collaboration process. Analysis shows that, this model can integrate devices with Web Services, and can achieve the conflict detection of doWS. This model is effective in the Internet of things.

In the environment of Internet of things, devices are encapsulated as web services and device operations are encapsulated as web service methods. These web services are stateful because devices can't be controlled by more than one client at the same time. Traditional web service composition methods are lack of considering state of the web services and have great challenge under Internet of things. In this paper, a three-layer model for modeling stateful web services composition based on extended Petri net is proposed. It includes web services layer, ws-resource layer and process layer. A method of a web service is abstracted as a ws-resource. There are two types of ws-resources: exclusive resources can be invoked serially, non-exclusive resources can be invoked simultaneously. In this model, a web service is associated with one or more ws-resources, a web service can be invoked only when all needed ws-resources are available. Finally an example is demonstrated to illustrate how this model works.

Three-dimensional (3D) on-chip memory stacking has been proposed as a promising solution to the “memory wall” challenge with the benefits of low access latency, high data bandwidth, and low power consumption. The stacked memory tiers leverage through-silicon-vias (TSVs) to communicate with logic tiers, and thus dramatically reduce the access latency and improve the data bandwidth without the constraint of I/O pin count. To demonstrate the feasibility of 3D memory stacking, this paper introduces a 3D System-on-Chip (SoC) for H.264 applications that can make use of multiple memory channels offered by 3D integration. Two logic tiers are stacked together with each having an area of 2.5×5.0mm2, with a 3-layer 8-channel 3D DRAM stacked on the top. The design flow for this 3D SoC is also presented. The prototype chip has been fabricated with GlobalFoundries' 130nm low-power process and Tezzaron's 3D TSV technology. The 3D implementation shows that the 3D ICs can alleviate the pressure from I/O pin count and allow parallel memory accesses through multiple channels.

The Internet of Things (IoT) connects physical world to the cyber-world, which may impose predicable and unpredictable changes to enterprise business models and our everyday life. In this paper, we propose a model to describe the impact of IoT on the material, information and capital flowas well as carbon footprint through the supply chain, and further support the model with possible business scenarios. Besides, we raise several issues that may interest scholars and practitioners.

Due to scale effects, the reliable pick-and-place manipulation of micro objects has been still a long-standing challenge in scientific and technological field. This paper presents an active method to manipulate microobjects, which employs inertia force and compound vibration to overcome adhesion forces. The compound vibration comes from electrostatic actuator and piezoelectrically driven microactuator. A micromanipulation system is set up to test and verify the proposed method. The system consists of a microgripper and a piezoelectric ceramics module. Surface and bulk micromachining technology is employed to fabricate the microgripper used in the system from a single crystal silicon wafer. Theoretical analyses were conducted to understand the release principle. Experiments were carried out through the manipulation of 30–80µm polystyrene spheres under an optical microscope. Experimental results show that this method can achieve a repeatability of 100% with releasing accuracy of 4µm, and is substrate independent. Based on this preliminary study, the system prove to be an effective solution for the active pick-and-place manipulation of microobjects.

To effectively manage a supply chain, we should no longer focus on physical flow and informational flow only, but also the financial flow. The financial supply chain management (FSCM) extracts capital tied to the raw material, work-in-process, inventory in transit, and accounts receivable (accounts payable). In this paper we review a set of supply chain financing solutions tailored by the financial institutions, and introduce a few quantitative works from the operations research society. The survey reveals the benefits of actively managing the financial supply chain and highlights the quantitative decision-making process in FSCM. Moreover, we propose several interesting problems in FSCM which are remained to be explored.

In today's volatile business environment, many companies are expanding, merging, contracting, or otherwise redesigning their supply chain networks. Supply chain network rationalization could significantly help to improve supply chain operational efficiency and reduce costs. As a generic technique for the analysis of complex and dynamic systems, simulation could play an important role in this field. This paper presents a real case showing how simulation could be the key enablement for a supply chain network rationalization project in the pharmaceutical industry. An IBM tool — Supply Chain Process Modeler (SCPM) has been applied in this project. Five major simulation scenarios are designed based on the pain points the client has, and numerical results show that simulation has addressed the key issues and provided the client accountable evaluation results for decision-making.

As the endeffectors of micromanipulation systems, microgrippers are crucial point of such systems for their efficiency and their reliability. So the steady and reliable performance of the microgripper is needed. In this work, a hybrid-type electrostatic silicon microgripper integrated vacuum tool is designed and fabricated to realize the steady gripping and reliable placing manipulations. This hybrid-type microgripper keeps the traditional functions of electrostatic comb drive microgripper and also combines the advantages of the vacuum tool. Vacuum tool integrated in this novel microgripper is used to improve its pick and place capability. Bulk micromachining technology is employed to fabricate the microgripper from single crystal silicon wafer. The vacuum control subsystem and electrostatic comb drive control subsystem are also designed for the microgripper controlling. At a driving voltage of 80V, a deflection of 25µm at the arm tip of the gripper is achieved. The objects ranging from 100µm to 200µm are picked and placed successfully in the experiment, and the experiment results indicate that this microgripper can realize the steady manipulation.

Supply chain transformation is to fundamentally rethink and radically redesign supply chain strategy, organization, processes and IT system(s) with the target of improving supply chain performance. This paper presents an integrated platform and a comprehensive methodology IBM Supply Chain Process Modeler (SCPM) for process-centric supply chain transformation. In the this methodology, the supply chain transformation is progressing along with the life cycle of business process management, from business process modeling, analyzing to optimizing or restructuring. The other contribution of the paper is the seamless integration of business process management tool and process execution IT system. Based on the integration, the huge gap between process model and execution system is closed and supply chain transformation decisions are realized in operation.

This paper describes a miniature MEMS-based piezoresistive displacement sensor with micrometric range and nanometric resolution. This kind of sensor should be integrated in a precision positioning mechanism (PPM) able to make accurate micro/nano-positioning tasks. We firstly provide a overview of the MEMS-based piezoresistive sensor principle. The theoretical modeling, dimension analysis and stress simulation of the sensor are carried out subsequently. Furthermore, the geometrical parameters of piezoresistance are designed and the main MEMS process of the sensor is developed. Finally, the experimental results are very good and the mini-sensor is able to measure a displacement with a 5 nm resolution over a 10µm range

In this paper we study the dynamics and kinematics of serial flexible link manipulators using the finite segment method. In our analysis the flexible manipulators are divided into a set of rigid bodies coupled with springs and dampers. The dynamic equations of multibody system with flexible manipulators are developed by means of spatial operator algebra which expression forms are the basis for a recursive O(n) dynamics algorithm, and the structure is a hybrid of forward and inverse dynamics algorithms for flexible manipulators using the finite segments method. Since the advantages of this method are shown to be automatic considering the geometrical nonlinear and dynamic stiffness terms in control equation and the characteristic of recursive approach of flexibly and conveniently modeling, it is useful and efficient to some certain flexible systems. Finally, to show the validation of this method is an efficient recursive algorithm, an example of two flexible link manipulators is presented, and the numerical results are favorably closed to the fact

Workforce is crucial to the success of service industry, especially for those software and consulting firms, where much of their revenue comes from the assignment of highly-skilled professionals. This is even more critical in the outsourcing environment, under which some of the professionals are outsourced from external vendors. Thus it is important to ensure hiring\assigning right people at the right time for the right project with the right cost. iRDM (intelligent Resource Deployment Manager) is an integrated solution developed by IBM Research division to support the full lifecycle management of highly-skilled professionals in the outsourcing environment. Based on the concept of "workforce supply chain", iRDM covers both the demand side and the supply side, and the matching between demand and supply. Besides the framework and software platform of iRDM, the demand and supply matching problem is mainly discussed. This solution has already been successfully deployed in an IBM consulting division.