Pierre Jolivet


I am a research scientist at CNRS. The Toulouse Institute of Computer Science Research (IRIT) is my host laboratory.
I work in the field of high-performance computing (especially for the design of fast and robust solvers in computational sciences).

HPDDM is a framework for high-performance domain decomposition methods I am developing with Frédéric Nataf.

Right-hand side figure: Poisson's equation solved with BDD-GenEO on 16 subdomains using Feel++.



pierre · jolivet @ enseeiht · fr
+33 5 34 32 21 65
Office F-316
2 rue Charles Camichel
31071 Toulouse Cedex 7


  1. Ph.D. in applied mathematics — Université de Grenoble. Link to thesis (8.3 MB).
  2. M.Sc. in Computational Science & Engineering (with high honor)Ensimag, Grenoble.
  3. M.Sc. in applied mathematics (with highest honor)Université Joseph Fourier, Grenoble.

Selected awards



  1. P. J., P.-H. Tournier. Block Iterative Methods and Recycling for Improved Scalability of Linear Solvers. Acceptance rate: 18% (82/446).
  2. S. Di Girolamo, P. J., K. D. Underwood, T. Hoefler. Exploiting Offload-Enabled Network Interfaces.
  3. V. Dolean, P. J., F. Nataf. An Introduction to Domain Decomposition Methods: Algorithms, Theory, and Parallel Implementation.
  4. R. Haferssas, P. J., F. Nataf. A robust coarse space for optimized Schwarz methods: SORAS-GenEO-2.
  5. S. Di Girolamo, P. J., K. D. Underwood, T. Hoefler. Exploiting Offload Enabled Network Interfaces. Best Paper Awardee at HotI ‘15.
  6. V. Dolean, P. J., F. Nataf, N. Spillane, H. Xiang. Two-Level Domain Decomposition Methods for Highly Heterogeneous Darcy Equations. Connections with Multiscale Methods.
  7. P. J., F. Hecht, F. Nataf, C. Prud'homme. Scalable domain decomposition preconditioners for heterogeneous elliptic problems.
  8. P. J., F. Hecht, F. Nataf, C. Prud'homme. Scalable Domain Decomposition Preconditioners For Heterogeneous Elliptic Problems. Acceptance rate: 20% (92/457). Best Paper Finalist at SC13.
  9. P. J., F. Hecht, F. Nataf, C. Prud'homme. Overlapping domain decomposition methods with FreeFem++.
  10. P. J., V. Dolean, F. Hecht, F. Nataf, C. Prud'homme, N. Spillane. High-performance domain decomposition methods on massively parallel architectures with FreeFem++.
  11. S. Allassonnière, P. J., C. Giraud. Detecting Long Distance Conditional Correlations Between Anatomical Regions Using Gaussian Graphical Models.