Transcript Presentaci n (198.7Kb)

Usage of architectural styles and
technologies in IT companies and
organizations
David Ameller and Xavier Franch
GESSI investigation group, Universitat Politècnica de Catalunya (UPC)
Outline
1.
2.
3.
4.
2
Problem statement
Empirical method used
Summary findings
Challenges and lessons learned
EASA'09, Cambridge
7/23/2016
Problem statement

We are currently driving a survey to find out which
architectural styles and technologies are being in use and their
relation with NFRs

Non-Functional Requirements (NFR)

Selection of … for a software system


Architectural style
Technologies
We want to apply this knowledge to Model-Driven Software
Development (MDSD)


3
We are currently developing a framework that considers NFR in the
MDSD process
EASA'09, Cambridge
7/23/2016
Empirical method used
The survey has been developed following an iterative
methodology


Each iteration was revised by IT experts and researchers of the area
We chose LimeSurvey, a good and cheap software for
developing electronic surveys


Rule-based questions allow a better design of the survey
For the dissemination of the survey



Personal contact with software architects
Advertisement in IT communities, using the power of social
networks such as LinkedIn and Facebook

4
International Association of Software Architects
EASA'09, Cambridge
7/23/2016
Summary findings (I)
We have 60 responses


When we have at least 100 responses we will publish the final results
Almost all (88%)
respondents use
relational DB
5
EASA'09, Cambridge
7/23/2016
Summary findings (II)

While 96% of respondents consider NFR (73% at the same
level as functional requirements), only 57% use NFR to take
architectural and technological decisions

80% of respondents do not use development tools that
analyze the NFR compliance, but 70% would like to use such
kind of tools

The most accepted answer for all questions about interaction
is that developers want to be asked, but only for the most
important decisions

A great part of the respondents (79%) do not use MDSD in
their software projects, so it is difficult to carry out a good
evaluation of this part of the survey at this moment
6
EASA'09, Cambridge
7/23/2016
Summary findings (III)
Non-Functional Requirements importance

35
Maintainability
Reusability
30
Efficiency
25
20
15
Reliability
Ma
Usability
Re
Portability
Eff
Cost
Re
Standards compliance
Us
Organizational
Po
Co
Sta
10
Org
5
Marginal
Medium
Important
Critical
No answer
0
None
7
Marginal
Medium
Important
EASA'09, Cambridge
Critical
7/23/2016
No answer
Challenges and lessons learned

This survey can be seen as a particular instrument that
addresses some of the questions raised in the EASA’09 cfp

This was an overview of the current answers to the survey, for
the final report we will do a deeper statistical data analysis

We will analyze which are the most important NFR for each
architectural style
Empirical evidence about the current state of software
architectures usage by asking the involved actors


8
We found some difficulties in terminology (academia vs. IT)
EASA'09, Cambridge
7/23/2016
Usage of architectural styles and
technologies in IT companies and
organizations
Questions and answers