Today, software development approach has been shifted from structure-oriented to object-oriented. UML has become a standard modeling language in object-oriented systems analysis and design. Over the past few years a considerable effort has been made in the model driven architecture transformation area. However, little attention has been paid to the issue of transformation from sequence diagram to class diagram for modeling platform independent model (PIM). This study, therefore, presents a systematic methodology, extended from Selonen et al (2003), for sequence diagram to class diagram transformation. A case is used to illustrate the concept, application, and advantage of using the proposed methodology. With this methodology, the system developer can construct PIM efficiently and thereby enhance the efficiency and quality of system development.

Abstract Processor-in-memory (PIM) architectures have been proposed in the recent years. One major objective of PIM is to reduce the performance gap between the CPU and memory. To exploit the potential benefits of PIM, we designed a statement-base parallelizing system ¡V SAGE(Statement-Analysis-Grouping-Evaluation) in [1][2][3]. From our pervious research, we find that the execution schedule is a critical factor to the performance of PIM systems. In this paper, we provide new scheduling mechanism for one-host and one-memory processors (1H-1M) and one-host and n-memory processors (1H-nM), respectively, to fully utilize all of the memory processors in PIM architectures. The experimental results of these two mechanisms are also discussed.

PIM (Processor-In-Memory) architectures have been proposed in recent years. One major objective of PIM is to reduce the performance gap between the CPU and memory. To exploit the potential benefits of PIM, we designed a statement base parallelizing system ¡VSAGE in [1, 2]. In order to make all processors take the best-fit workload in PIM, iteration base analysis is another research issue in this paper. We extend this system to achieve better performance by devising several comprehensive optimizing techniques, which include IMOP (Intelligent Memory Operation) recognition, tiling for PIM, and a precise mechanism to get workload balance execution schedule. The experimental results are also presented and discussed.

SAGE (Statement Analysis Grouping Evaluation) system is a statement-based parallelizing compiler designed for PIM (Processor-in-Memory) architectures. By extracting the potential parallelism and utilizing the capabilities of PIM architectures, this system can achieve better performance of the applications. It analyzes the source codes, transforms the codes, makes good executing schedule and generates the codes which can execute on PIM architectures. This thesis focuses on the implementation of the code generator in the SAGE system. The code generator can transforms the original source codes into subroutines according to the corresponding HBG (Hyper Block Graph) and execution schedule which is determined by the other modules of SAGE system. It generates subroutines, inserts necessary arguments, determines the positions of caller and callee and generates the interface program for executing on PIM simulator. The generated codes after our code generator are also discussed.

The Pim family of serine/threonine protein kinases (Pim 1, 2, and 3) contribute to cellular transformation by regulating glucose metabolism, protein synthesis, and mitochondrial oxidative phosphorylation. Drugs targeting the Pim protein kinases are being tested in phase I/II clinical trials for the treatment of hematopoietic malignancies. The goal of these studies was to identify Pim substrate(s) that could help define the pathway regulated by these enzymes and potentially serve as a biomarker of Pim activity. To identify novel substrates, bioinformatics analysis was carried out to identify proteins containing a consensus Pim phosphorylation site. This analysis identified the insulin receptor substrate 1 and 2 (IRS1/2) as potential Pim substrates. Experiments were carried out in tissue culture, animals, and human samples from phase I trials to validate this observation and define the biologic readout of this phosphorylation. Our study demonstrates in both malignant and normal cells using either genetic or pharmacological inhibition of the Pim kinases or overexpression of this family of enzymes that human IRS1S1101 and IRS2S1149 are Pim substrates. In xenograft tumor experiments and in a human phase I clinical trial, a pan-Pim inhibitor administered in vivo to animals or humans decreased IRS1S1101 phosphorylation in tumor tissues. This phosphorylation was shown to have effects on the half-life of the IRS family of proteins, suggesting a role in insulin or IGF signaling. These results demonstrate that IRS1S1101 is a novel substrate for the Pim kinases and provide a novel marker for evaluation of Pim inhibitor therapy.

¡@¡@With the rapid advance of information technology, embedded systems are becoming mature technology and have penetrated into most of the equipment. On the other hand, the unified modeling language (UML) has become a standard modeling language in object-oriented systems analysis and design. Over the past few years a considerable effort has been made in modeling the platform independent model (PIM) with UML for business information systems. However, the detailed guideline for modeling the PIM of embedded systems is lacking. ¡@¡@To alleviate the forgoing problem, this study proposed a PIM modeling methodology for embedded systems. This includes the use of sequence diagram, state machine diagram, timing diagram and class diagram for behavior and static modeling. For each diagram, the detailed modeling processes and rules are described. The research methodology is articulated using the design science research methodology. A usability evaluation is performed to demonstrate its applicability with a real-world embedded system case. With this proposed method, the system developer can easily model the PIM of embedded systems, and thereby enhance the efficiency and effectiveness of system development.

¡@¡@With the rapid advance of information technology, Service Oriented Architecture (SOA) is becoming a mature technology. Many software firms have implemented it to enhance the flexibility and efficiency of system development. Several systems analysis and design methods have been proposed to develop the SOA systems. Among them, however, the detailed guideline and procedure for identifying the services and components in PIM (Platform Independent Model) stage are lacking. ¡@¡@To alleviate the forgoing problem, this study proposed a PIM modeling methodology for developing an SOA system. The methodology which includes the business process, service, and component layers, uses the unified modeling language to describe the detailed modeling processes and rules for each layer. The research methodology is articulated using the design science research methodology. A usability evaluation is performed to demonstrate its applicability with a testing case. With this proposed method, the system developer can easily model the PIM of SOA systems, and thereby enhance the efficiency and effectiveness of system development.

This thesis presents different Internet Protocol (IP) multicasting design models and together with their implementation methods in OPNET Modeler 14.5. The models presented in this report are designed for handling various types of advanced multimedia applications and services when using the multicast communication paradigm. The architectural models presented in this thesis are independent of the routing protocols used as well as of the underlying networking environment. Several exiting challenges related to high bandwidth requirements, low network delay, and jitter are discussed together with these design models, and are simulated using the OPNET Modeler. The emerging demands for new multicast capable applications, for both real-time and non real-time traffic are considered together in this report in order to better evaluate multicast traffic requirements. Efficient routing methods, group management policies, and shortest path algorithms are also discussed in the proposed models. The pros and cons of IP multicasting are described using the Protocol Independent Multicast Sparse Mode (PIM-SM) method, in the highlight of the proposed design models. Several results related to link utilization, end-to-end delay, packet loss, PIM control traffic, Internet Group Management Protocol (IGMP), and network convergence are obtained by using already available statistical methods and modules from OPNET Modeler. Our major contribution with this thesis work relates to components associated with network convergence, routing delay, and processing delay. These components are illustrated in this thesis using various OPNET metrics. Moreover, several issues related to protocol scalability, supportability, performance optimization and efficiency are also presented by using OPNET’s build in reports, i.e., executive flow and survivability analysis reports. / 0735609277

Die antiapoptotischen Effekte der Pim-1 Kinase im Rahmen der ischämischen und Desfluran-induzierten Postkonditionierung / Antiapopiotic Effects of ishemic- and Desflurane-induced postconditioning are mediated by Pim-1 Kinase

No description available.