@techreport{TR-IC-PFG-24-31,
   number = {IC-PFG-24-31},
   author  =  {Matheus  Pierre  {Sforça} and Breno Bernard Nicolau {de
                   França}},
   title   =  {{Maintainability  Analysis  of  Open-Source  Serverless 
                   Repositories}},
   month = {December},
   year = {2024},
   institution = {Institute of Computing, University of Campinas},
   note = {In English, 35 pages.
    \par\selectlanguage{english}\textbf{Abstract}
       Serverless  computing  is  a  considerable  recent  paradigm in
       application  development.  This  type  of  application  aims to
       reduce  infrastructure and development costs by abstracting the
       configuration  needed  to  scale and deploy them. Also, as they
       are  only  executed  when  there is some demand, it promises to
       optimize the cost of running projects. \par As it is considered
       a  recent paradigm, good practices and its most recommended use
       cases are still under research. This study examines open-source
       serverless  applications’  structure,  organization,  and  code 
       quality  by  mining  software  repositories  on GitHub. We also
       compare  our  findings  with related work to see if we identify
       the same problems found in the existing literature in the area.
       In  this  study, we selected relevant open-source repositories,
       analyzed   their  content  with  SonarQube  and  internal  code 
       metrics,  such  as  the  number  of  functions  per  file,  and 
       consolidated  its  findings to highlight common characteristics
       and possible maintainability problems in this paradigm. In this
       study,  we  also  discuss  the limitations of this approach and
       propose  the  next  steps  for  future  research.  \par  In our
       analysis,  we  found  out  that  Serverless projects are indeed
       smaller  in  terms  of  lines  of  code  and usually have great
       indices  of  maintainability when checking for Code Smells. The
       amount  of  configuration  needed  also has great advantages by
       this  paradigm,  and it was stated a great incentive on writing
       focused  functions  with a low number of lines. However, a lack
       of   consensus   was  identified  regarding  structuring  these 
       projects.  In  terms  of  scalability,  we stated that as these
       applications  grow in size, problems related to maintainability
       are  more  perceptible due to unused code, large functions, and
       infrastructure constraints that lead to poor coding practices.
  }
}