Hrischuk, C.E., Woodside, C.M., Rolia, J.A.: Trace-Based Load Characterization for Generating Software Performance Models. IEEE Trans. on Software Eng., V.25, No.1 (1999) 122--135  Home/Search   Document Details and Download   Summary   Related Articles   Check

....DCE RPC [35] CORBA[33] Java [13] and mobile agents [38] are technologies which are being used to build these types of distributed operations. The Model Making Automation Process is a generalization of a process for the automated construction of a performance model of distributed applications [18, 19]. MMAP starts with a special application level trace (called an ANGIOTRACE ) which is converted into a graph grammar model for analysis (the graph grammar is called software proper time or just 1. If a process is the unit of concurrency of the operating system, a task is a single threaded ....

....events only when a software dye is present. An ANGIOTRACE event record has special time stamp values so that a set of partial ordering relations can be used to reconstruct the cause and effect relationships between them. The ANGIOTRACE was introduced in [16] and has been further described in [18, 19]. The reader is referred to Appendix A for a more detailed description of the ANGIOTRACE. The ANGIOTRACE makes no assumptions about network structure or task behavior so it is network and environment transparent, being useful for a uni processor, multi processor, loosely coupled systems (e.g. ....

[Article contains additional citation context not shown here]

C. Hrischuk, C. M. Woodside, J. Rolia, and R. Iversen. Trace-based load characterization for generating software performance models. Technical Report SCE-97-05, Dept. of Systems and Computer Engineering, Carleton University, Ottawa, Canada K1S 5B6, 1997. To appear in IEEE Trans. on Software Engineering.

....changing the design, or boosting the hardware capability can accommodate it. To be effective, the analysis should be carried forward into the development of the system. As the specification is refined into more detailed forms, PASD can use models created from those forms too. Other work in [5] 10][23] describes a suitable technique for building layered models, based on traces created from the specification, or even from an execution of a prototype. In [5] and [20] it was applied with commercial CASE tools that specify behavior by communicating state machines. 2. Budget Analysis Road Map ....

C. E. Hrischuk, C. M. Woodside, and J. A. Rolia, "Trace-Based Load Characterization for Generating Software Performance Models," IEEE Trans. on Software Engineering, vol. v 25, n 1, pp. 122-135, Jan 1999.

....a client in requesting service from some lower level resource (e.g. 28] 32] 21] This is natural behaviour for a software server, and leads to a model with abstract entities that correspond to software resources. Layered queueing has been applied to web servers [5] transaction processing [10], data routers [18] and distributed databases [26] Besides performance related properties, resource architecture also identifies possibilities of resource deadlock due to resource request patterns. This paper investigates the connection between a software architecture and its resources, in ....

....because two concurrent programs within the same architecture request their resources in the same order, they cannot deadlock in a situation where each is waiting for a resource already held by the other. Applications of LQNs include client server systems, web servers [5] transaction processing [10], distributed databases [26] connection management [11] and intelligent network servers in telecommunications [28] A single layer LQN, such as would arise from Scenario 1 in Figure 3, is just an ordinary queueing network model. The Program resource in Figure 3 is the customer or customers, ....

[Article contains additional citation context not shown here]

C.E. Hrischuk, C.M. Woodside, J.A. Rolia, "TraceBased Load Characterization for Generating Software Performance Models", IEEE Trans. on Software Engineering, v 25, n 1, pp 122-135, Jan. 1999.

....this barrier. Such a technique has just become available and it is called the Model Making Automation Process (MMAP) MMAP is described briefly below. The reader is referred to [5] for a little larger description. MMAP is an extension to the performance model building technique described in [7]. 2.0 The Model Making Automation Process The Model Making Automation Process (MMAP) is a well formed, traceable model building process for automatic software execution model generation. It is well formed because a chain of formal transformations is defined from the input model domain (an ....

.... environment (MLOG [10] a distributed software system simulator called Parasol [13] coarse grained UNIX objects [9] and in the DCE RPC environment [15] using data collected by the POET debugger [20] The ANGIOTRACE was introduced in [6] by the author and has been further described in [7, 8]. 1. Angiotrace is a trademark of Angiograms for Software Analysis Inc. who can be contacted at angio istar.ca. ANGIOTRACE appears in capital letters to signify it is a trademark. V1.2 Page 3 22 July 1998 Using an ANGIOTRACE as the input description has several benefits. Traces are a low cost ....

C. Hrischuk, C. M. Woodside, J. Rolia, and R. Iversen. Trace-based load characterization for generating software performance models. Technical Report SCE-97-05, Dept. of Systems and Computer Engineering, Carleton University, Ottawa, Canada K1S 5B6, 1997. To appear in IEEE Trans. on Software Engineering.

No context found.

Hrischuk, C.E., Woodside, C.M., Rolia, J.A.: Trace-Based Load Characterization for Generating Software Performance Models. IEEE Trans. on Software Eng., V.25, No.1 (1999) 122--135

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC