Update Trilinos homepage

Author: maherou

_data/definitions.yml

@@ -4,14 +4,24 @@ linear_solver: "Software that finds a solution vector x for Ax = b, given a know
 
 sparse_linear_solver: "Software that solves systems Ax = b, where A has enough zeros that it makes sense to keep track of what is zero or day_introductory_course_on_trilinos."
 
+sparse_problem: "A problem such that dependencies, if expressed as a matrix, has enough zeros that it makes sense to keep track of what is zero or not."
+
 nonlinear_solver: "Software that finds a solution vector u for F(u) = 0, where F() is a nonlinear function."
 
 transient_solver: "Software that finds a solution vector x(t) at time t to satisfy dx/dt = f(x, t), x(0) = x0."
 
-linear_algebra_kernels: "Software that computes results such as vector dot products (alpha = x*x') and matrix vector multiplication y = Ax."
+parallel_linear_algebra_kernels: "Software that computes results such as vector dot products (alpha = x*x') and matrix vector multiplication y = Ax on a parallel computer."
 
-sparse_kernels: "Software that takes into account matrix or graph sparsity, such as sparse matrix-vector multiplication and graph partition for parallel efficiency."
+parallel_sparse_kernels: "Software that takes into account matrix or graph sparsity, such as sparse matrix-vector multiplication and graph partition for parallel efficiency."
 
 optimization_solver: "Software that finds u such that u maximizes or minimizes F(u) for some function F()."
 
 uq_solver: "Software that quantifies the uncertainty of a given solution with a certain probability."
+
+mpi: "The Message Passing Interface (MPI) is the primary interface for realizing parallel execution on distributed memory computers.  MPI libraries such as MPICH, OpenMPI and MVAPICH enable parallel execution on dozens to millions of processors."
+
+multicore: "Processor with multiple cores that enable parallel execution, usually with access to the same pool of shared memory."
+
+accelerator: "A computing device designed for high-throughput parallel execution.  Sometimes called GPUs."
+
+vectorization: "A form of parallelism particularly well suited to performing the same sequence of operations on arrays of values."

index.md

@@ -4,25 +4,52 @@ keywords: homepage scientific libraries parallel computing
 permalink: index.html
 ---
 
+### The Trilinos Community
 The Trilinos Project is a community of developers, users and
 <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.user_developer}}">user-developers</a>
 focused on collaborative creation of algorithms and enabling technologies within an object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems on new and emerging high-performance computing (HPC) architectures.
 
+### Trilinos Software
+Trilinos is also a collection of reusable scientific software libraries, known in particular for
+ <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.linear_solver}}">linear solvers</a>,
+  <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.nonlinear_solver}}">non-linear solvers</a>,
+ <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.transient_solver}}">transient solvers</a>,
+ <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.optimization_solver}}">optimization solvers</a>, and
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.linear_solver}}">uncertainty quantification (UQ) solvers</a>.
+
+### Parallel Computing Using Trilinos
+Most Trilinos algorithms and software are built upon its abilities to construct and solve
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.sparse_problem}}">sparse problems</a>, using
+
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.sparse_linear_solver}}">sparse linear solvers</a>. These solvers rely on a collection of data structure classes and functions (kernels) for
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.parallel_linear_algebra_kernels}}">parallel linear algebra</a>, especially
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.parallel_sparse_kernels}}">parallel sparse kernels</a>.
+
+### Trilinos Portable Parallel Execution
+Trilinos is targeted for all major parallel architectures, including distributed memory using
+<a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.mpi}}">the Message Passing Interface (MPI)</a>, <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.multicore}}">multicore</a> using a variety of common approaches,
+href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.accelerator}}">accelerators</a> using common and emerging approaches, and
+ href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.vectorization}}">vectorization</a>.
+
+ Trilinos parallel functionality is written on top of libraries that support compile-time polymorphism, such that, as long as a given algorithm and problem size contain enough latent parallelism, the same Trilinos source code can be compiled and execution on any reasonable combination of distributed, multicore, accelerator and vectorizing computing devices.
+
+
+### Trilinos Packages
 Trilinos is organized around fundamental software elements called <a href="#" data-toggle="tooltip" data-original-title="{{site.data.glossary.trilinos_package}}">packages</a>.  The Trilinos package architecture enables simultaneous development of many new capabilities in a federated system.  Each package has its own name and identity within the research community, giving the package team recognition outside of Trilinos itself.
 
-## Trilinos Product Suites
+### Trilinos Product Suites
 Trilinos contains many packages, organized in [five product suites](product.html):
 - Data services: Vectors, matrices, graphs and similar data containers, and related operations.
 - Linear and Eigenproblem solvers: For large, distributed systems of equations.
 - Nonlinear solvers and analysis tools: Includes basic nonlinear approaches, continuation methods and similar.
 - Discretizations: Many finite element and similar approaches.
 - Framework: Tools for building, testing and integrating Trilinos capabilities.
 
-## Trilinos is on GitHub
+### Trilinos GitHub Location
 The [primary Trilinos repository](https://github.com/trilinos/Trilinos) is  hosted on GitHub.
 
-## TUG 2020
+### TUG 2020
 [The 2020 Trilinos User-Developer Group (TUG) Meeting](https://trilinos.github.io/trilinos_user-developer_group_meeting_2020.html) will be virtual.  Details available in August, 2020.
 
-## TUG 2019
+### TUG 2019
 [The 2019 Trilinos User-Developer Group Meeting (TUG)](https://trilinos.github.io/trilinos_user-developer_group_meeting_2019.html) was held in October in Albuquerque.