LUISS

Curriculum Vitae

PERSONAL DATA:
Name: Simone Sbaraglia.
Date of Birth: June 27, 1972.
Place of Birth: Rome, Italy.
E-mail: ssbaraglia@luiss.it

EDUCATION:
March 2003: Ph.D. in “Mathematical Methods and Models for Technology and Society”, Faculty of Engineering, University of Rome “La Sapienza”.
Thesis title: “Efficient Optimization Techniques in Finance and Economics”.
October 1998: graduated in mathematics “summa cum laude”.
University of Rome “La Sapienza”
Thesis title: “Hamilton-Jacobi equations on bounded domains with weak boundary conditions”.

PROFESSIONAL CAREER:
Since October 2005: Tenured Professor (“Professore Associato”) at “University of Cagliari”, Italy.
Job Responsibilities: Tenured professor in Mathematics and Computer Science.

October 1, 2003 – October 31, 2005: Research Staff Member at the “Advanced Computing and Technology Center (ACTC)”, I.B.M. T.J. Watson Research Center, Yorktown Heights, NY.
Job Responsibilities:
– Chief architect and developer of the pSigma performance-oriented binary instrumentation infrastructure.
– Chief architect and developer of the Sigma infrastructure for memory analysis.
– Chief architect of the IBM HPC Toolkit v2.
– Chief architect of the IBM HPCS Toolkit for productive computing.
– Member of the IBM Productivity Group on the DARPA Percs/HPCS project for the development of High Performance

March 1, 2003 – October 8, 2003: Post-doctoral position at the “Advanced Computing and Technology Center (ACTC)”, I.B.M. T.J. Watson Research Center, Yorktown Heights, NY.
Job Responsibilities:
– Chief architect and developer of the Sigma infrastructure for memory analysis.
– Data-centric performance tools.
– Efficient compression and analysis of memory traces.

November 01, 2002 – February 28, 2003: Supplemental position at the “Advanced Computing and Technology Center (ACTC)”, I.B.M. T.J. Watson Research Center, Yorktown Heights, NY.
Job Responsibilities:
– Design and development of memory simulators for the IBM Power3 and Power4 systems.
– Simulation and performance analysis of prefetching algorithms on IBM systems.

January 02, 2002 – October 28, 2002: research grant with the Italian National Research Council – Institute for Applied Computing. Rome, Italy.
Job Responsibilities:
– Design and efficient implementation of algorithms for the approximation of optimality problems arising in financial theory.

July 02, 2001 – January 02, 2002: Supplemental position at the “Advanced Computing and Technology Center (ACTC)”, I.B.M. T.J. Watson Research Center, Yorktown Heights, NY.
Job Responsibilities:
– Development of memory simulators and tools to analyze and improve the performance of a memory subsystem.

April 01, 2000 – June 30, 2001: research grant with Italian National Research Council – Institute for Applied Computing. Rome, Italy.
Job Responsibilities:
– Design and implementation of a software infrastructure for the optimal asset-liability management for insurance companies.
– Development of effective approximation methods for Hamilton-Jacobi equations of first and second order.

August 01, 1999 – April 01, 2000: research grant with Italian National Research Council – Institute for Applied Computing. Rome, Italy.
Job Responsibilities:
– Quantitative analysis of Mortgage-Backed Securities.
– Development of mathematical methods and numerical algorithms for pricing Mortgage-Backed Securities and Life Insurance Products.

March 01, 1999 – August 01, 1999: supplemental position at National Insurance Association, Italia – Capital Markets Department.
Job Responsibilities:
– Mathematical models for assets and liabilities management for life insurance companies.
– Software package to perform Monte Carlo simulations on the evolution of a life insurance company liabilities.

January 01, 1999 – June 06, 1999: research grant with Italian National Research Council – Institute for Applied Computing. Rome, Italy.
Job Responsibilities:
– Design and implementation of a software infrastructure for the simulation of the propagation of brakes and the diffusion of chemical polluting in porous materials.

PUBLICATIONS:
S. Sbaraglia, I. Chung, G. Cong, D. Klepacki, S. Seelam, H. Wen, An Extensible Bottleneck Discovery Framework, Proceedings of the 2008 ACM/IEEE international conference on Supercomputing, November 15-21, 2008, Austin, TX.

S. Sbaraglia, I. Chung, G. Cong, D. Klepacki, S. Seelam, H. Wen, A Framework for Automated Performance Bottleneck Detection, 13th International Workshop on High-Level Parallel Programming Models and Supportive Environments, IPDPS 2008,April 14th 2008, Miami, FL.

H. Wen, S. Sbaraglia, S. Seelam, I. Chung, G. Cong, D. Klepacki, A Productivity Centered Application Performance Tuning Framework, Proceedings of the II International Conference on Performance Evaluation Methodologies and Tools (Valuetools), Nantes, France. Oct. 23-25 2007.

H. Wen, S. Sbaraglia, S. Seelam, I. Chung, G. Cong, D. Klepacki, A Productivity Centered Tools Framework for Application Performance Tuning, Proceedings of the IV International Conference on the Quantitative Evaluation of Systems (QEST07), Edinburgh, Scotland. Sep. 16-19 2007.

G. Cong, S. Sbaraglia, A study on the locality behavior of parallel and sequential algorithms for connectivity problems, Lecture Notes in Computer Science, Vol. 4297, 2006, pp. 583-596. ISSN:0302-9743.

G. Cong, S. Sbaraglia, A study on the locality behavior of parallel and sequential algorithms for connectivity problems, Proceedings of the International Conference on High Performance Computing,Bangalore, India, Dec.18-21 2006.

M. Papi, S. Sbaraglia, Optimal Asset-Liability Management with Constraints: A Dynamic Programming Approach, Applied Mathematics and Computation, Vol. 173, n.1, 2006. ISSN: 0096-3003.

S. Sbaraglia, J. Odom, L. Derose, K. Ekanadham, J. Hollingsworth, Using Dynamic Tracing Sampling to Measure Long Running Programs, proceedings of the 2005 ACM/IEEE international conference on Supercomputing, November 12-18, 2005, Seattle, WA.

S. Sbaraglia, K. Ekanadham, S. Crea, S. Seelam, pSigma: An Infrastructure for Parallel Application Performance Analysis using Symbolic Specifications, EWOMP, October 18-22, 2004, Stockholm,Sweden.

M. Papi, S. Sbaraglia, Lipschitzian Estimates in Discrete-Time Constrained Optimal Control, “Dynamics of Continuous, Discrete and Impulsive Systems?, Series A: Mathematical Analysis”, Vol.13, n.1, 2006. ISSN: 1201-3390.

L.Derose, K.Ekanadham, S.Sbaraglia An Approach for Symbolic Mapping of Memory References, accepted for publication at the international conference Euro-Par 2004, Pisa, 31 Aug-3 Sep 2004.

M. Adamo, A. Amadori, M. Bernaschi, C. La Chioma, A. Marigo, B. Piccoli, S. Sbaraglia, A. Uboldi, D. Vergni Optimal Strategies for the Issuance of Public Debt Securities, “International Journal of Theoretical and Applied Finance”, vol.7, 2004.

M. Papi, S. Sbaraglia, Regularity Properties of Constrained Set-Valued Maps, “Nonlinear Analysis:
Theory, Methods & Applications”, Vol. 54, n. 7, 2003. ISSN: 0362-546X.

R. Natalini, C. Nitsch, G.Pontrelli, S.Sbaraglia, A numerical study of a nonlocal model of damage propagation under chemical aggression, “European Journal of Applied Mathematics”, Volume 14,Issue 4, (2003) p. 447-464

M. Bernaschi, M. Briani, F. Gozzi, M. Papi, S. Sbaraglia, A model for the optimal asset liability management for insurance companies, “International Journal of Theoretical and AppliedFinance”, vol. 6, 227, 2003.

L.Derose, K.Ekanadham, J.K.Hollingsworth, S.Sbaraglia SIGMA: A Simulator Infrastructure to Guide
Memory Analysis, Proceedings of the 2002 ACM/IEEE conference on Supercomputing, p.1-13, November 16, 2002, Baltimore, Maryland.

GRANTS, AWARDS AND OTHER RESEARCH ACTIVITIES :
2008: US Patent Invention: A Method for an Extensible and Programmable Framework for Automating Performance Analysis and Tuning of Software Applications”. IBM Invention Achievement Award, IBM Research.

2008: US Patent Invention: “Profiling Application Performance According to Data Structures”. IBM Invention Achievement Award in recognition for the research conducted on data-centric profiling, IBM Research.

2007: US Patent Invention: “An Extensible Infrastructure for Automated Detection of Application Performance Bottlenecks”. IBM Invention Achievement Award in recognition for the research conducted on the automatic characterization of application performance bottlenecks, IBM Research.

2007: US Patent Invention: “A Programmable Binary Method for Application Performance Data Collection”. IBM Invention Achievement Award in recognition for the research conducted on binary instrumentation and performance data collection, IBM Research.

Since August 2006: Member of the IBM Productivity Group on the DARPA Percs/HPCS project for the development of High Performance Productive Computing Systems

May 2006: Organization of the II International Workshop Petascale Tools Strategy Workshop, IBM
Research, May 16,17 2006, Yorktown Heights, NY.

May 2005: Organization of the international workshop Petascale Tools Strategy Workshop, IBM
Research, May 3,4 2005, Yorktown Heights, NY.

Since May 2004: Principal Investigator for IBM Corporation in the “National Science Foundation” funded research effort
2004-2007 “Performance Measurements and Modeling of Deep Memory Hierarchy Systems”, Grant for $800.000. The research is conducted in collaboration with “University of Maryland at College Park” and the “San Diego Supercomputing Center”.

2003: US Patent Invention Control-Flow Based Compression of Execution Traces, IBM Research. IBM Invention Achievement Award in recognition for the research conducted on the effective compression of execution instruction and memory traces, IBM Research.

Since October 2003: Chief architect and manager of the IBM Research Project “SIGMA” for the simulation of shared and distributed-memory architectures and deep memory hierarchies.

TECHNICAL REPORTS:
March 2003: S. Sbaraglia, D. Skinner Performance Prediction and Scalability Analysis Using SIGMA, NERSC Technical Report, Oakland Supercomputing Facility, Lawrence Berkeley National Lab, Oakland, CA.

October 2002: L.Derose, K.Ekanadham, J.K.Hollingsworth, S.Sbaraglia SIGMA: A Simulator Infrastructure to Guide Memory Analysis, IBM Research Technical Report.

January 2001: M. Papi, S. Sbaraglia, Optimal Asset-Liability Management with Constraints: A Dynamic Programming Approach, IAC Technical Report Q13/001.

May 1999: R. Natalini, G. Pontrelli, S. Sbaraglia, Studio della propagazione di fratture e la diffusione di inquinanti in materiali porosi, IAC Technical Report, 5/1999.

June 1998: M. Falcone, A. Siconolfi, S. Sbaraglia, Analysis and Approximation of some Optimal Control Problems, technical report and issue for the course in “Control Theory” at “Università degli Studi di Roma LA SAPIENZA”.

TALKS AND CONFERENCE PARTICIPATION:
08/2007: Dagstuhl Seminar “Code Instrumentation and Modeling for Parallel Performance Analysis”, Aug. 18-24 2007, Dagstuhl, Germany.

05/2006: The IBM HPC Toolkit, IBM Petascale Workshop, May 16, 17 2006. IBM T.J. Watson Res. Center, Yorktown Heights, NY.

08/2005: demo of the pSigma instrumentation infrastructure for the DARPA PERCS review, IBM T.J. Watson Res. Center, Yorktown Heights, NY.

06/2005: speaker on the topic “Latest news from ACTC: The New HPCT Toolkit”, at the SCICOMP 11 Meeting, June 1-3 2005, Edinburgh, Scotland.

05/2005: speaker on the topic “The IBM HPC Toolkit and Performance Analysis of Scientific Applications” at the SCICOMP 11 meeting, University of Edinburgh, Edinburgh UK.

05/2005: speaker on the topic “Current Issues in Instrumentation of Parallel Applications”, at the IBM Petascale Tools Strategy Workshop, IBM Research, May 3,4 2005, Yorktown Heights, NY.

05/2005: invited speaker on the topics: “pSigma: A Parallel Instrumentation Infrastructure for Parallel Applications”, “Totalview and the Art of Parallel Debugging”, “The IBM HPC Toolkit for Performance Analysis of Scientific Applications” (3 seminars) at the Arctic Region Supercomputing Center, Fairbanks, Alaska, USA.

03/2004: “Totalview and Parallel Debugging”, Linux Cluster Institute, University of Illinois at Urbana, Champaing, Champaign, IL, USA.

10/2004: “pSigma: An Infrastructure for Parallel Application Performance Analysis using Symbolic Specifications”, Univ. of Stockholm, Sweden.

07/2004: invited speaker on “ACTC Tools for Performance Tuning”, SCICOMP 10 & SP-XXL IBM Scientific Computing User Group, Austin, Texas.

04/2004: invited speaker on “SIGMA: A Simulator Infrastructure to Guide Memory Analysis, Performance Prediction and Scalability Analysis Using SIGMA”, Lawrence Berkeley Lab Seminars, Berkeley, California.

03/2004: “SIGMA: A Simulator Infrastructure to Guide Memory Analysis”, CINECA, Bologna.

11/2003: “Demo of the SIGMA memory simulator and overview of the performance tools developed at the “Advanced Computing and Technology Lab”, IEEE Supercomputing Conference, Phoenix, California, Nov. 15-21, 2003.

10/2003: invited speaker on “Optimal Asset-Liability Management with Constraints”, IBM Research Seminars, NY.

5/2003: invited speaker on “Performance Modeling Tools: An Overview of SIGMA, Metasim and Dimemas”, HPC Monthly Meeting, IBM Research, NY.

3/2003: invited speaker on “Optimal Asset-Liability Management with Constraints”, Univ. of Rome “La Sapienza”.

11/2002: Participation to the international conference “Supercomputing”, Baltimore, Nov. 16-22, 2002.

9/2002: speaker on “Optimal Asset-Liability Management with Constraints: A Dynamic Programming Approach” at the AMASES 2002 workshop, Verona 11-14 September, 2002.

06/2000: speaker on “A Model for the Optimal Asset-Liability Management for Insurance Companies” in the symposium on Mathematical Finance during the S.I.M.A.I. (Italian Society for Applied and Industrial Mathematics) Workshop (Ischia, 2-9 June 2000).

COMPUTER SCIENCE SKILLS:
Deep knowledge of the following operating systems, programming languages and tools:

Operating Systems: Unix, Linux, Windows, Mac OS X.

Programming Languages: C, C++, Fortran90, Fortran77, Perl, Matlab, Java, Python, Lex&Yacc.

Programming Models: parallel programming on distributed-memory and shared-memory architectures (MPI, OpenMP), Pthread programming, network programming and clientserver design.

Architectures: Power architecture, Power3 and Power4 assembly and binary language, RISC systems, XCOFF & ELF binary formats, AIX and Linux Kernel internals.

Numerical Methods: Optimization algorithms, finite difference schemes, finite element methods, stochastic methods, linear algebra, statistical analysis.

Other: Compiler principles, operating system principles and design, memory hierarchy design, binary instrumentation, performance tools.

Al fine di migliorare la tua esperienza di navigazione, questo sito utilizza i cookie di profilazione di terze parti. Chiudendo questo banner o accedendo ad un qualunque elemento sottostante acconsenti all’uso dei cookie.