ML20202J230

Forwards Ballots for ANS-10.2, Portability of Scientific & Engineering Software & ANS-15.4, Selection & Training of Personnel for Research Reactors. Author Voted Not Approved to ANS-10.2 & Approved to ANS-15.4

ML20202J230
Person / Time
Site:	National Bureau of Standards Reactor
Issue date:	02/02/1999
From:	Weiss S NRC (Affiliation Not Assigned)
To:	Raby T AMERICAN NATIONAL STANDARDS INSTITUTE
References
NUDOCS 9902090071
Download: ML20202J230 (9)

Text

,a g

February 2, 1999 Mr. Tawfik M. Raby Chairman, N-17 Building 235, Room A141 NationalInstitute of Standards and Technology Gaithersburg, Maryland 20899

Dear Mr. Raby:

h

SUBJECT:

ANS-10.2 and ANS-15.4 N-17 COMMITTEE BALLOTS Enclosed are my ballots for ANS-10.2, " Portability of Scientific and Engineering Software," and ANS-15.4, " Selection and Training of Personnel for Research Reactors." I have voted "not approved" for ANS-10.2 and " approved" for ANS-15.4. Furthermore, I

' nave reviewed the Scopes of the standards, and have no objections.

If you have any questions concerning these ballots, please contact Al Adams at 301-415-1127 or me at 301-415-2170.

Sincerely, ORIGINAL SIGNED PY, Seymour H. Weiss, Director Non-Power Reactors and Decommissioning Project Directorate Division of Reactor Program Management Office of Nuclear Reactor Regulation NEIosbr$e.50-184 As stated cc w/ enclosure:

j See next page 1.

]

I 9N DISTRIBUTION:

E-Mall HARD COPY PDoyle TBurdick PUBLIC AAdams TDragoun MMendonca CGC (015-818) EHylton TMichaels CBassett DMatthews PDND r/f SHolmes Plsaac SWeiss WEresian (DocketFile"50184 PDN M P, ' ;LA HICB -

A on Stewart SWeiss A/ 9 / 99 .2,/ J /99 "),/ 3/99 OFFICIAL RECORD COPY DOCUMENT NAME: G:\SECY\ ADAMS \ BALLOT 2 -

9902090071 990202 PDR ADOCK 05000184 P PDR

(.me ,

3pf t"G (*, C[p,j7 k t.LL W- = ' ~ ~ ~ [ ')

m o

ps** %q, y- 4 UNITED STATES

1. 8 ]

NUCLEAR REGULATORY COMMISSION ,

, WASHINoToN. D.c. 2055Hm01

% ,, # February 2, 1999 Mr. Tawfik M. Raby Chairman, N-17 Building 235, Room A141 NationalInstitute of Standards and Technology Gaithersburg, Maryland 20899 g

SUBJECT:

ANS-10.2 and ANS-15.4 N 17 COMMITTEE Ball.OTS i

Enclosed are my ballots for ANS-10.2, " Portability of Scientific and Engineering Software," and ANS 15.4, " Selection and Training of Personnel for Research Reactors." I have voteu "not approved" for ANS-10.2 and " approved" for ANS-15.4. Furthermore, I have reviewed the Scopes of the standards, and have no objections.

If you have any questions concerning these ballots, please contact Al Adams at 301-415-1127 or me at 301-415-2170.

Sincerely,

, {l Seymour H Weiss, Director Non-Power Reactors and Decommissioning Project Directorate Division of Reactor Program Management Office of Nuclear Reactor Regulation

Enclosure:

As stated cc w/ enclosure:

See next page

7,.

q ,

. 3 '. '

u 1.w

~

LNationalInstitute of Standards Docket No. 50-184

' 'and Technology y.

cc:

!Wontgomery County Executive .

Donty Office Building Rockville, MD 20858

Director

' Department of State Planning ,

301.. West Preston Street Baltimore, MD 21201 ,

Director Department of Nature! Resources Power Plant Siting Program Energy & Coastal Zone Administration Tawes State Office Building Annapolis, MD 21401 Mr.' Tawfik Raby, Deputy Chief Reactor Radiation Division NationalInstitute of Standards and Technology U.S. Department of Commerce Gaithersburg, MD 20899 Honorable Michael L. Subin Montgomery County Council

.. Stella S. Werner Council Office Building i Rockville, MD 20850 Dr. William Vernetson  !

- Director of Nuclear Facilities I

Department of Nuclear Engineering Sciences l University of Florida l Gainesville, FL 32611-8300- 1 Mr. Jim Torrence l Reactor Radiation Division NationalInstitute of Standards and Technology {

U.S. Department of Commerce  !

Gaithersburg, MD 20899

.m J

l l

ANSI N-17 COMMITTEE BALLOT DOCUMENT:

ANS-15.4 " Selection and Training of Personnel for Research Reactors". (Reaffirmation)

BALLOT DUE: February 17.1999 Y Approved Approved with Comments Not Approved Not Voting Comments:

(NOTE - all " Approved with Comments" and "Not Approved" ballots must include comments.

Seymour H. Weits-P int Name Retum ballot to:

1. Signature 4M/

Tawfik M. Raby USNRC Chairman, N-17 Representing NationalInstitute of Standards & Technology Mailstop 8561 Building 235, Room A141 Gaithersburg, MD 20899-8561 N7 !ff l date'

l l

l ANSI N-17 COMMITTEE BALLOT DOCUMENT: ANS-10.2 " Portability of Scientific and Engineering Software". (Revision)

BALLOT DUE: February 17.1999 Approvad Approved with Comments X Not Approved Not Voting Comments: (NOTE all " Approved with Comments" and "Not Approved" ballots must include comments.

See next page.

Seymour H. Weiss Print Name Return ballot to:

/ 8ignature

~'

h Tawfik M. Raby USNRC Chairman, N-17 Representing NationalInstitute of Standards & Technology Mailstop 8561 Building 235, Room A141 / f Gaithersburg, MD 20899-8561 2/L/97 I Dath ' '

l

NRC Comments on ANS-10.2, " Portability of Scientific and Engineering Software" General Comments

1. We agree that it is good practice to design for portability - even though this is not an NRC requirement.

2. The proposed standard is vague in many areas but the greatest shortcoming is the lack of reference to the lEF.E software engineering standards or equivalent for software development. These standards were not written just for safety-critical real time applications but are well suited to development of computational code. j Several sources quote that the greatest contribution to portability / reuse is the use 1 of good general design practices which are described in these standards. The use I of data flow diagrams, entity relationship diagrams, data dictionaries and a structured life cycle (especially a good requirements specification) are useful in i

promoting portability. I

3. Additional good practices are available from many sources and would, perhaps, be worth referencing.

4. Perhaps the standard could discuss the pros and cons of commercial products that support multiple platforms, has a large user base, and has been at least if not more stable than several of the options. i

)

5. Some of the recommendations for portability / reuse include the generalizing of modules / objects as discussed in some of the work on domain analysis. The working group might wish to consider adding a discussion of the pros and cons of this.

6 Some more discussion of some of the potential trouble spots may be helpful. For example, computational modules in C and Fortran can be designed for stable portability but a similar high degree of portability in the remainder of the modules (graphics, interprocessor communication, network naming rules, and database interfaces) are in a state of flux. Some of these issues may require programming methods (such as using message passing interface for databases) that would not be the first inclination for the designer.

Specific comments

7. ' Foreword. The first paragraph states ".... practices recommended." It is not stated who provides the recommendation.

l l

4

t

8. Foreword. The first paragraph states "...have not been assigned sufficient importance." Consider changing this to "..have not usually been assigned sufficient -

importance." so as to not condemn the few that have done a good job.

9. Foreword. Consider placing the three referenced standards in either numerical or alphabetical order.

10. Definitions. Consider using IEEE 610.12 definitions when possible.

11. Definitions.' " Localization"is usually referred to as modularity or objects.

12. Introduction. The list of referenced standards and guidelines should reference the IEEE or other international software development standards and guides.

13. Section 4.1, Source and Revision Control. Source code and revision control should be referred to as configuration management and is required.  ;

14. .Section 4.2.1, Program Structure and Flow. Output processing should be included in the last sentence.

15. Section 4.2.2, Subprograms, it is not clear what guidance is being provided here.

The same statement could apply to programs in general. .

16. Section 4.3, Programming Languages. There should be reference to some of the better guidelines that currently exist for programming languages.

17. Section 4.3.6, Assembly Language. This section should be reworded. The use of assembly language is commonly needed, especially for high speed requirements.

18. Section 4.4.3, Data Libraries. The guidance of the last sentence will be difficult to meet. How do you provide detailed instructions on how to generate a library for an unknown future hardware / software environment? The decision has alreaJy been rnade by this point that the existing library is not portable to the current known systems, i l

19. Section 4.5.1, initialization and input processing, item (6). In addition to the consistency checks, there should be checks on the standard items (out-of-range, null sets, wrong type, too many or not enough data points, etc.).

20. Section 4.5.4, Output and Termination Processing. Explain (here or in the definitions) what is meant by " edited." Does it mean putting them into a nice output format or actually doing calculations.

Section 4.6.1, Variable Names. This section could mention methods such as data 21.

dictionaries as tools to assist in this.

I

i

22. Section 4.6.3, Source Documentation.1his section should also include the need to keep the design and test tools current and available. The documentation should be required rather than recommended. The manuals and handbooks should also be required if they include any meaningful information not available from the source code.

23. Section 4.7.1, Interfacing with External Packages. Packages provided other than the computer program source should also include the items standardized by the program language. Test runs of different compilers, both set for standard C for the same target machine will not always yield identical code. It may not matter but should not be assumed.

1 1

1 I

t