Commissariat a l Energie Atomique-CEA/LIST (Mr. Huascar Espinoza, huascar.espinoza@cea.fr)
[GRM: Examples] Fig 10.21 shows a composite diagram, which is not conform to UML: parts are shown with attribute and operation compartments. This should be revised.
Instead of using as an example a composite structure, it will be used a class diagram. There are no implications in the text, so the only change is in the diagram shown in Figure 10.21. The diagram is changed by the next one:
> in this diagram is simplified to the minimum to avoid cluttering the picture.
Revised Text:
-
Modify Fig.10.21 by the one above.
Disposition: Resolved
Disposition: Resolved
OMG Issue No: 11772
Title: HRM: Examples
Source:
Commissariat a l Energie Atomique-CEA/LIST (Mr. Huascar Espinoza, huascar.espinoza@cea.fr)
Summary:
[HRM: Examples] Figure 14-78, 14-82, 14-83, 14-84 specify tagged values with a syntax that does not correspond to VSL (e.g., 166MHz, 64 MB). It should be clarified the syntax used, or align to VSL.
Resolution:
Figures 14-78, 14-82, 14-83 and 14-84 are modified in order to be consistent with VSL syntax, and with the syntax which has been allowed in the resolution of issue 11781.
Revised Text:
Figure 14-78 is modified as follow:
Figure 14-82 is modified as follow:
Figure 14-83 is modified as follow:
Figure 14-84 is modified as follow:
Disposition: Resolved
Disposition: Resolved
OMG Issue No: 11773
Title: GQAM General
Source: Commissariat a l Energie Atomique-CEA/LIST (Mr. Huascar Espinoza, huascar.espinoza@cea.fr)
Summary:
The references (publications) along the text in gqam should be moved to Annex G. (This is also true for chapter pam)
Resolution:
Replace each reference by suitable words which name the authors of documents, and insert the references in Annex G.
Revised Text:
Revisions are given both for gqam and pam.
1. Add to Annex G these references, inserted alphabetically by first author:
Jin, Jingwen and Nahrstedt, Klara, : “QoS Specification Languages for Distributed Multimedia Applications: A Survey and Taxonomy”, IEEE Multimedia, July/September 2004 (Vol. 11, No. 3) pp. 74-87.
Franaszek, P.A. and Nelson, R.D. , “Properties of delay-cost scheduling in time-sharing systems”, IBM J. of Research and Development, Volume 39, Number 3, 1995.
Skene, James, Lamanna, David and Wolfgang Emmerich, "Precise Service Level Agreements", Proc 26th International Conference on Software Engineering (ICSE'04), pp. 179-188).
Cruz, R.L. and Arvind V. Santhanam , "Optimal Routing, Link Scheduling and Power Control in Multi-hop Wireless Networks", Infocom 2003.
Smith, C.U. and L. Williams, "Performance Solutions", Addison- Wesley 2000
Lavenberg, S, "Performance Modeling Handbook”, Academic Press, 1983).
Menasce, Daniel, and Virgilio Almeida, “Capacity Planning for Web Services: metrics, models, and methods”, Prentice Hall, 2001
Xu, J, M. Woodside, and D. Petriu, "Performance Analysis of a Software Design using the UML Profile for Schedulability, Performance and Time," Proc. 13th Int. Conf. on Modeling Techniques and Tools for Computer Performance Evaluation (TOOLS 03), Urbana, USA, Sept. 2003.
TPC (Transaction Processing Council), “TPC Benchmark W (Web Commerce) Specification”, Version 1.8, Feb 19, 2002
2. Complete this reference that is already in Annex G
Jain, Raj, The Art of Computer Performance Modeling, Wiley, 1991.
3. Modify the text where the references were made, in 11 places
11773-1 p 259, no change
11773-2 p 260, sec Time-Related NFPs, para 2, sentence 2:
original text: A recent survey is given in (Lui Sha, Tarek F. Abdelzaher, Karl-Erik Årzén, Anton Cervin, Theodore P. Baker, Alan Burns, Giorgio C. Buttazzo, Marco Caccamo, John P. Lehoczky, Aloysius K. Mok, "Real Time Scheduling Theory: A Historical Perspective", Real-Time Systems, Volume 28, Number 2-3, November 2004, pp. 101-155).
new text: A recent survey is given by Sha et al.
11773-3 p 260 second-last paragraph
original text: Soft deadline: a stated percentage of responses must be complete within this delay. Quality-of-service specifications often are stated in these terms (QoS Specification Languages for Distributed Multimedia Applications: A Survey and Taxonomy, Jingwen Jin Klara Nahrstedt, IEEE Multimedia, July/September 2004 (Vol. 11, No. 3) pp. 74-87, esp Fig 4.).
new text: Soft deadline: a stated percentage of responses must be complete within this delay. Quality-of-service specifications often are stated in these terms (see e.g. Jin and Nahrstedt, esp. Fig 4.).
11773-4 p 260 last paragraph
original text: Delay cost function: a function of delay should be within a target value, or should be minimized. This is useful for trading off delays of multiple streams of requests that compete for resources. (see e.g. Franaszek and Nelson).
new text: Delay cost function: a function of delay should be within a target value, or should be minimized. This is useful for trading off delays of multiple streams of requests that compete for resources. (e.g. P. A. Franaszek and R. D. Nelson , Properties of delay-cost scheduling in time-sharing systems, IBM J. of Research and Development, Volume 39, Number 3, 1995).
11773-5 p 261 para2
original text: The expression of such measures in service level agreements was discussed and surveyed recently in: James Skene, D. Davide Lamanna, Wolfgang Emmerich, "Precise Service Level Agreements", Proc 26th International Conference on Software Engineering (ICSE'04), pp. 179-188). They point out the value of analysis of the execution path of the software.
new text: The expression of such measures in service level agreements was discussed and surveyed recently by Skene et al. They point out the value of analysis of the execution path of the software.
11773-6 p 261 last para of Sec 15.1
original text: In wireless networks the power of transmission may be controlled, affecting messaging speed. Optimal policies take into account competition between nodes ("Optimal Routing, Link Scheduling and Power Control in Multi-hop Wireless Networks", R. L. Cruz and Arvind V. Santhanam, Infocom 2003)
new text: In wireless networks the power of transmission may be controlled, affecting messaging speed. Optimal policies take into account competition between nodes (see e.g. Cruz and Santhanam).
11773-7 p 264, second last para
original text: BehaviorScenarios in similar forms are widely used for timing analysis. In schedulability analysis they are called "task sequences" (Jane Liu, "Real Time Systems", Wiley), and specifications of timing normally refer to certain scenarios. Performance models are also created from scenarios (C.U. Smith and L. Williams, "Performance Solutions", Addison- Wesley 2000)....
new text: BehaviorScenarios in similar forms are widely used for timing analysis. In schedulability analysis they are called "task sequences" (Jane Liu, "Real Time Systems"), and specifications of timing normally refer to certain scenarios. Performance models are also created from scenarios (see e.g. Smith and Williams, "Performance Solutions")...
11773-8 p 327, second para after Fig 17.11
original text: The QN model ignores the performance impact of the process thread pool sizes. To represent this we require an extended queueing network or layered network, that models the simultaneous possession of two resources (threads and processor). (S. Lavenberg, "Performance Modeling Handbook, Academic Press, 1983).
new text: The QN model ignores the performance impact of the process thread pool sizes. To represent this we require an extended queueing network or layered network, that models the simultaneous possession of two resources (threads and processor). (see, e.g. Lavenberg, "Performance Modeling Handbook”).
11773-9 p 328 para 1
original text: The service time of the logical server is the holding time of the thread. Solution of an EQN is approximate, using various strategies (see eg, Jain or Menasce).
new text: The service time of the logical server is the holding time of the thread. Solution of an EQN is approximate, using various strategies (see eg, Jain, or Menasce and Almeida).
11773-10 p 329 , first para after the list:
original text: The example is elaborated from the Transaction Processing Council standard scalable benchmark TPC-W for electronic commerce, by putting ...
new text: The example is elaborated from the Transaction Processing Council standard scalable benchmark TPC-W for electronic commerce (see TPC for the specification), by putting ...
11773-11 p 331 para 1.
original Text: This is a soft real-time embedded system with that a set of cameras that must be scanned at least once every second.
new: This is a soft real-time embedded system with that a set of cameras that must be scanned at least once every second, as described by Xu et al.
Disposition: Resolved