Future Trends of Software Technology and Applications: Software Architecture

Transcription

1 Pittsburgh, PA Future Trends of Software Technology and Applications: Software Architecture Paul Clements Software Engineering Institute Carnegie Mellon University Sponsored by the U.S. Department of Defense 2006 by Carnegie Mellon University 1

2 Report Documentation Page Form Approved OMB No Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE JAN REPORT TYPE 3. DATES COVERED to TITLE AND SUBTITLE Future Trends of Software Technology and Applications: Software Architecture 5a. CONTRACT NUMBER 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Carnegie Mellon University,Software Engineering Institute (SEI),Pittsburgh,PA, PERFORMING ORGANIZATION REPORT NUMBER 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR S ACRONYM(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES 14. ABSTRACT 11. SPONSOR/MONITOR S REPORT NUMBER(S) 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT a. REPORT unclassified b. ABSTRACT unclassified c. THIS PAGE unclassified Same as Report (SAR) 18. NUMBER OF PAGES 17 19a. NAME OF RESPONSIBLE PERSON Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

3 Software Engineering Institute Applied R&D laboratory situated as a college-level unit at Carnegie Mellon University, Pittsburgh, PA, USA Established in 1984 Technical staff of 335 Offices in Pittsburgh, Pennsylvania, Arlington, Virginia, and Frankfurt, Germany Purpose: Help others make measured improvements in their software engineering practices 2005 by Carnegie Mellon University 2

4 The ascendance of software architecture The software architecture of a program or software system is the structure or structures of the system, which comprise elements, the externally visible properties of those elements, and the relationships among them. --- Software Architecture in Practice, 2 nd ed., 2003 Software architecture as a field of study has enjoyed a steady ascendance since the 1980s (although its foundational principles go back much farther: e.g., Parnas on Information Hiding,1972). Dedicated conferences: WICSA About CFPs mentioning it are active at any particular time Journal of Systems and Software (Elsevier) section 2005 by Carnegie Mellon University 3

5 The ascendance of software architecture In practice, software architecture is has been embraced organizationally. Well defined practitioner roles and career paths Well defined architecture-based development methods Codification of duties, skills, and knowledge A recent search revealed 29 university courses with on-line descriptions * 22 industrial courses 7 certificate/certification programs 75 books * by Carnegie Mellon University 4

6 The ascendance of software architecture Architecture is seen as essential for the construction of large, complex software systems It is the primary carrier of a system s quality attributes It structures the development project as well as the software being developed It is a capital investment that can be reused, powering (for example) software product lines It is an essential communication vehicle among the system s stakeholders 2005 by Carnegie Mellon University 5

7 Architecture s role to date In order to predict its future, it helps to understand architecture s role up to this point. When programs became too large and complex to understand (or to engineer) by working only with source code, architecture became indispensable to exert intellectual control over the software. It helped us think in terms of problem-domain abstractions unsupported by programming languages of the day. Architecture provided large scale conceptual chunks of functionality wired together as a solution, chunks that addressed particular parts of the problem. Generally useful or commonly recurring chunks led to prepackaging and pre-wiring of those chunks, which in turn led to languages in which those chunks became primitives by Carnegie Mellon University 6

8 Examples of chunks Architectures and architectural thinking, driven by principles of modular design, gave rise to Clients and servers Middleware Services Whole-enterprise IT solutions Over time, the chunks have become simultaneously more specialized, more general, and more complex. 1960s chunk: SQRT subroutine Intermediate chunk: database, transaction processing 2000s chunk: shopping cart, on-line auction, B2B solutions by Carnegie Mellon University 7

9 Here s an idea Programs are very complex and they require programmers to think too much in terms of the underlying computational platform, and not the problem space. Let s give programmers a way to express themselves in the language of the problem space. Let s introduce software to automatically translate from platform-independent expressions into platform-dependent solutions. Wouldn t that be great? We could call this concept 2005 by Carnegie Mellon University 8

10 FORTRAN. FORTRAN was designed for mathematicians and scientists, and remains the preeminent programming language in these areas today. It allows people to work with their computers without having to understand how the machines actually work, and without having to learn the machine s assembly language. John Backus -- The History of Computing Project by Carnegie Mellon University 9

11 FORTRAN FORTRAN was not thought of by its creators as a programming language. It was designed to be the input to a translator. People initially threw out their FORTRAN code and pored over the assembly language generated from it. On the day people stopped looking at the assembly language and started treating the FORTRAN code itself as the artifact of value, FORTRAN became a programming language by Carnegie Mellon University 10

12 So what? Axioms: 1. There is nothing inherently different between a specification language and a programming language. 2. There is no stark dividing line between problem space and solution space. Rather, it is a continuum by Carnegie Mellon University 11

13 So what? Model-driven architecture is remarkable (to me) in that its descriptions seem to have very little to do with architecture. I prefer the term model driven development. The fundamental concept is in the translation from models to solutions. So if this important new paradigm can sidestep architecture, can software architecture be all that important? 2005 by Carnegie Mellon University 12

14 So what? In some IT communities, software architecture seems to be de-emphasized today. The chunks are so large and sophisticated, and go together in pre-wired ways. Creating the software architecture is not a complex task. Some IT firms say their architectural decisions consist of choosing between two competing vendors solutions. Once they do that, everything else follows. Can software architecture be all that important? 2005 by Carnegie Mellon University 13

15 Software architecture s role in software engineering The history of software engineering can be viewed as a steadily increasing spiral of language expressiveness, into problem spaces and away from platform details. When a new way to express problems arrives on the scene, programs are initially very simple their structure is very simple. But we always learn to express more and more complex solutions using the languages we have by Carnegie Mellon University 14

16 Final word As the solutions written using these new mega-programming languages become more and more complex, architecture will be there to help solution architects (the current term for those who program in these super-languages) to impose order on the chaos make sure quality attributes are achieved to structure the solution being given to the underlying computing systems. We will see continued movement towards supporting larger, more complex, and more varied problem-space abstractions, backed up by more sophisticated translation or compiler technology that will allow today s specification languages or modeling languages to continue to be tomorrow s programming languages by Carnegie Mellon University 15

17 Specific predictions Architecture-based practices (e.g., formal evaluations, standardized documentation) will continue to be codified. Tooling to support architecture design and development will improve greatly. We ll see true round trip engineering in which the current gulf between architectural design and downstream design and implementation languages and representation is bridged by Carnegie Mellon University 16

18 Questions Now or Later Paul Clements Product Line Systems Program U.S. Mail: Software Engineering Institute Carnegie Mellon University 4500 Fifth Avenue Pittsburgh, PA World Wide Web: architecture SEI Fax: by Carnegie Mellon University 17