sapienza

Software Analysis and Optimization Laboratory

Sapienza University of Rome

 

Welcome to the SoftwarE AnalysiS and OptimizatioN (aka SEASON) Laboratory at Sapienza University of Rome.

Team


camil demetrescu

Camil Demetrescu

demetres [at] dis.uniroma1.it
Associate Professor at DIAG
Sapienza University of Rome
[website] [dblp]

irene finocchi

Irene Finocchi

finocchi [at] di.uniroma1.it
Associate Professor at DI
Sapienza University of Rome
[website] [dblp]

daniele d elia

Daniele Cono D'Elia

delia [at] dis.uniroma1.it
PostDoc at DIAG
Sapienza University of Rome
[website] [dblp]

emilio coppa

Emilio Coppa

coppa [at] dis.uniroma1.it
PostDoc at DIAG
Sapienza University of Rome
[website] [dblp]


Former students:
  • Bruno Aleandri (M.Sc.), Valerio Bellizia (M.Sc.), Raffaele Di Fazio (M.Sc.), Luca Di Maio (M.Sc.), Alessandro Macchioni (M.Sc.), Romolo Marotta (B.Sc.)


Projects


aprof

aprof - input-sensitive profiling

aprof is a Valgrind tool for performance profiling designed to help developers discover hidden asymptotic inefficiencies in the code. From one or more runs of a program, aprof measures how the performance of individual routines scales as a function of the input size, yielding clues to its growth rate.
Related papers: [CDF-PLDI12] [CDFM-CGO14] [CDF-TSE14] [C-VAL14]


nearestfit

NearestFit - predicting MapReduce performance

The NearestFit progress indicator targets accuracy of progress predictions for MapReduce jobs in the presence of data skewness and super-linear computations. This is achieved combining performance profiling, machine learning techniques, and data streaming algorithms.
Related papers: [CF-SOCC15]


Research interests


  • Malware Analysis
  • Performance Profiling
  • Adaptive Optimization
  • Big Data Performance Prediction


Publications


Our latest publications:
  • [DD-CGO16] Daniele Cono D'Elia and Camil Demetrescu. Flexible On-Stack Replacement in LLVM. 2016 International Symposium on Code Generation and Optimization (CGO 2016), 2016.
  • [DDF-SPE15] Daniele Cono D'Elia, Camil Demetrescu, and Irene Finocchi. Mining Hot Calling Contexts in Small Space. Software: Practice and Experience, John Wiley & sons, 2015.
  • [FFF-JEA15] Irene Finocchi, Marco Finocchi, and Emanuele G. Fusco. Clique counting in MapReduce: Algorithms and experiments. ACM Journal of Experimental Algorithmics (ACM JEA), 2015. [DOI]
  • [CF-SOCC15] Emilio Coppa and Irene Finocchi. On data skewness, stragglers, and MapReduce progress indicators. Sixth ACM Symposium on Cloud Computing (SoCC'15), pp 139-152, 2015. [DOI] [PDF] [SLIDES]
  • [DFR-TOPLAS14] Camil Demetrescu, Irene Finocchi, and Andrea Ribichini. Reactive Imperative Programming with Dataflow Constraints. ACM Transactions on Programming Languages and Systems (ACM TOPLAS), 37(1), pp 1-53, 2014.
  • [C-VAL14] Emilio Coppa. An interactive visualization framework for performance analysis. 8th International Conference on Performance Evaluation Methodologies and Tools (VALUETOOLS 2014), pp 159-164, 2014. [DOI] [PDF] [SLIDES] [PROJECT SITE]
  • [CDF-TSE14] Emilio Coppa, Camil Demetrescu, and Irene Finocchi. Input-Sensitive Profiling. IEEE Transactions on Software Engineering (IEEE TSE'14), 40(12), pp 1185-1205, 2014.[DOI] [PDF] [PROJECT SITE]
  • [CDFM-CGO14] Emilio Coppa, Camil Demetrescu, Irene Finocchi, and Romolo Marotta. Estimating the Empirical Cost Function of Routines with Dynamic Workloads. 12th IEEE/ACM International Symposium on Code Generation and Optimization (CGO 2014), pp 230-239, 2014. [DOI] [PDF] [SLIDES] [PROJECT SITE]
  • [DDF-OOPSLA13] Daniele Cono D'Elia and Camil Demetrescu. Ball-Larus Path Profiling Across Multiple Loop Iterations. 28th ACM SIGPLAN Conference on Object-Oriented Programming, Systems, Languages, and Applications (OOPSLA'13), pp 373-390, 2013.
  • [CDF-PLDI12] Emilio Coppa, Camil Demetrescu, and Irene Finocchi. Input-Sensitive Profiling. 33rd ACM SIGPLAN conference on Programming Language Design and Implementation (PLDI 2012), pp 89-98, 2012. [DOI] [PDF] [SLIDES] [PROJECT SITE]
  • [ADFF-OOPSLA12] Giorgio Ausiello, Camil Demetrescu, Irene Finocchi, and Donatella Firmani. k-Calling context profiling. 27th ACM International Conference on Object Oriented Programming Systems Languages and Applications (OOPSLA'12), pp 867-878, 2012.
  • [DDF-PLDI11] Daniele Cono D'Elia, Camil Demetrescu, and Irene Finocchi. Mining Hot Calling Contexts in Small Space. 32nd ACM SIGPLAN conference on Programming Language Design and Implementation (PLDI'11), pp 516-527, 2011.
  • [DFR-OOPSLA11] Camil Demetrescu, Irene Finocchi, and Andrea Ribichini. Reactive Imperative Programming with Dataflow Constraints. 26th ACM International Conference on Object Oriented Programming Systems Languages and Applications (OOPSLA'11), pp 407-426, 2011.


Grants


amanda
Amanda: Algorithmics for MAssive and Networked DAta - MIUR, PRIN 2012


algodeep
AlgoDEEP: Algorithmic Challenges for Data-intensive Processing on Emerging Computing Platforms - MIUR, PRIN 2008


mainstream
MainStream: Algorithms for Massive Information Structures and Data Streams - MIUR, PRIN 2006


algonext
Algo-Next: Algorithms for the Next Generation Internet and Web: Methodologies, Design and Applications - MIUR, PRIN 2004