1 1  Capabilities: Dynamic load balancing and static data partitioning -Geometric, graph-based, hypergraph-based -Interfaces to ParMETIS, PT-Scotch, PaToH Graph coloring Graph/matrix fill-reducing or locality-preserving ordering iZoltan interface supports ITAPS mesh interfaces  Download via the Trilinos toolkit:  Further information: Karen Devine, Erik Boman, Siva Rajamanickam, Zoltan Combinatorial Scientific Computing Toolkit

2 2 Example: LCLS RF gun, colors indicate distribution to different CPUs. Fields are computed only in causal region, using p-refinement. (Courtesy SLAC National Accelerator Laboratory.) Particle Partitioning Field Partitioning  SLAC’s Pic3P accelerator simulation solves Maxwell’s equations Field computation on fixed mesh Particles moving through domain  Load balance using two different data decompositions Fields partitioned with graph-based methods (ParMETIS) Particles partitioned geometrically (Zoltan RCB 3D)  Enables scalable solution of larger problems: 24k CPUs, 750M DOFs, 5B particles Zoltan example: Partitioning for Particle-in-Cell methods

3 3 Zoltan’s Data-Structure Neutral Design Supports a Wide Range of Applications and Data Structures. Multiphysics simulations Adaptive mesh refinement Crash simulations Particle methods Parallel electronics networks 1 2 Vs SOURCE_VOLTAGE 1 2 Rs R 1 2 Cm012 C 1 2 Rg02 R 1 2 Rg01 R 1 2 C01 C 1 2 C02 C 12 L2 INDUCTOR 12 L1 INDUCTOR 12 R1 R 12 R2 R 1 2 Rl R 1 2 Rg1 R 1 2 Rg2 R 1 2 C2 C 1 2 C1 C 1 2 Cm12 C Linear solvers & preconditioners