Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Cosmin Petra Project Name: stoch_prog Division: Mathematics and Computer Science Division Project title: Stochastic Programming and Application for Energy Systems Associated funding: DOE-ASCR: Mathematics of Petascale Data Program. Scalable Algorithms for Gaussian Processes. Mihai Anitescu, PI. Other Systems: Science: This project is an optimization framework that exploits the weather forecast information in the operation of energy systems. Our objective is to illustrate how the use of forecast information and high accuracy physics and computations can translate into lower operating costs. We will solve the stochastic unit commitment/energy dispatch problem with network transmission constraints and wind power generation. Project description: In the previous year, we were able to obtain a high level of parallelization in solving stochastic optimization problems. The half-arrow shaped structure these problems have was exploited at the linear algebra layer in an optimization framework based on interior-point methods. We devised a scenario-based decomposition that uses the Schur-complement technique. The code we implemented, PIPS, was able to solve problems up to 189 million variables using 2,048 cores on Fusion with a (strong) scaling efficiency of 78%. The efficiency of the 1024 cores run was slightly better than 90%. We also used Elemental to parallelize dense linear solves with the Schur complement matrix. When solving more complex stochastic optimization problems like the stochastic unit commitment problem with network transmission constraints, the scenario (second stage) subproblems become really large and impose memory related constraints on how many MPI processes can be run on one node. We will follow a mixed approach MPI+SMP to overcome this limitation. On the other hand, the first stage solves involving the Schur complement matrix will be also performed in the same mixed programming model using the built-in features of Elemental. We expect to perform tests involving all Fusion's cores. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 300000 Justification: Thank You, The LCRC Accounts System