Ashutosh Mahajan

Mixed-integer nonlinear optimization

Many optimal decision problems in scientific, engineering, and public sector applications involve both discrete decisions and nonlinear system dynamics that affect the quality of the final design or plan. These decision problems lead to mixed-integer nonlinear programming (MINLP) problems that combine the combinatorial difficulty of optimizing over discrete variable sets with the challenges of handling nonlinear functions. We review models and applications of MINLP, and survey the state of the art in methods for solving this challenging class of problems. Most solution methods for MINLP apply some form of tree search. We distinguish two broad classes of methods: single-tree and multitree methods. We discuss these two classes of methods first in the case where the underlying problem functions are convex. Classical single-tree methods include nonlinear branch-and-bound and branch-and-cut methods, while classical multitree methods include outer approximation and Benders decomposition. The most efficient class of methods for convex MINLP are hybrid methods that combine the strengths of both classes of classical techniques. Non-convex MINLPs pose additional challenges, because they contain non-convex functions in the objective function or the constraints; hence even when the integer variables are relaxed to be continuous, the feasible region is generally non-convex, resulting in many local minima. We discuss a range of approaches for tackling this challenging class of problems, including piecewise linear approximations, generic strategies for obtaining convex relaxations for non-convex functions, spatial branch-and-bound methods, and a small sample of techniques that exploit particular types of non-convex structures to obtain improved convex relaxations. We finish our survey with a brief discussion of three important aspects of MINLP. First, we review heuristic techniques that can obtain good feasible solution in situations where the search-tree has grown too large or we require real-time solutions. Second, we describe an emerging area of mixed-integer optimal control that adds systems of ordinary differential equations to MINLP. Third, we survey the state of the art in software for MINLP.

Heuristic static load-balancing algorithm applied to the fragment molecular orbital method

In the era of petascale supercomputing, the importance of load balancing is crucial. Although dynamic load balancing is widespread, it is increasingly difficult to implement effectively with thousands of processors or more, prompting a second look at static load-balancing techniques even though the optimal allocation of tasks to processors is an NP-hard problem. We propose a heuristic static load-balancing algorithm, employing fitted benchmarking data, as an alternative to dynamic load balancing. The problem of allocating CPU cores to tasks is formulated as a mixed-integer nonlinear optimization problem, which is solved by using an optimization solver. On 163,840 cores of Blue Gene/P, we achieved a parallel efficiency of 80% for an execution of the fragment molecular orbital method applied to model protein-ligand complexes quantum-mechanically. The obtained allocation is shown to outperform dynamic load balancing by at least a factor of 2, thus motivating the use of this approach on other coarse-grained applications.

Sequential computation of elementary modes and minimal cut sets in genome-scale metabolic networks using alternate integer linear programming

Motivation: Elementary (flux) modes (EMs) have served as a valuable tool for investigating structural and functional properties of metabolic networks. Identification of the full set of EMs in genome‐scale networks remains challenging due to combinatorial explosion of EMs in complex networks. It is often, however, that only a small subset of relevant EMs needs to be known, for which optimization‐based sequential computation is a useful alternative. Most of the currently available methods along this line are based on the iterative use of mixed integer linear programming (MILP), the effectiveness of which significantly deteriorates as the number of iterations builds up. To alleviate the computational burden associated with the MILP implementation, we here present a novel optimization algorithm termed alternate integer linear programming (AILP). Results: Our algorithm was designed to iteratively solve a pair of integer programming (IP) and linear programming (LP) to compute EMs in a sequential manner. In each step, the IP identifies a minimal subset of reactions, the deletion of which disables all previously identified EMs. Thus, a subsequent LP solution subject to this reaction deletion constraint becomes a distinct EM. In cases where no feasible LP solution is available, IP‐derived reaction deletion sets represent minimal cut sets (MCSs). Despite the additional computation of MCSs, AILP achieved significant time reduction in computing EMs by orders of magnitude. The proposed AILP algorithm not only offers a computational advantage in the EM analysis of genome‐scale networks, but also improves the understanding of the linkage between EMs and MCSs. Availability and Implementation: The software is implemented in Matlab, and is provided as supplementary information. Contact: hyunseob.song@pnnl.gov Supplementary information: Supplementary data are available at Bioinformatics online.

Finite-Element Modeling of Retention in Nanocrystal Flash Memories With High-

The analysis of retention characteristics is essential for reliability assessment of any type of flash memory. A high-

k

k

Interpoly Dielectric Stack

(HK) material in the Interpoly dielectric (IPD) stack is known to improve the program–erase characteristics and makes the control oxide thicker for better retention time; however, it also brings-in defects along with it, creating undesirable charge loss path that plays a significant role in retention. Accurate modeling of this leakage process is important for optimizing the architecture that is less prone to oxide defects and exhibiting improved charge retention. In this paper, we present a physical model considering all possible charge leakage mechanisms that include inelastic tunneling through multiple traps in addition to direct tunneling (DT). The trap capture process is modeled as an inelastic process from nanocrystal (NC) to trap state, while the trap-to-trap tunneling is calculated by Bardeen’s Transfer Hamiltonian approach. The trap emission process and DT from the spherical NC are modeled by computing the lifetime of quasi-bound states with very small decay widths in the finite-element setup. We show that a single trap model in HK cannot explain the retention characteristics. We propose a new asymmetric IPD structure that show significant improvement in retention time.

A Branch-and-Estimate Heuristic Procedure for Solving Nonconvex Integer Optimization Problems

We present a method for solving nonconvex mixed-integer nonlinear programs using a branch-and-bound framework. At each node in the search tree, we solve the continuous nonlinear relaxation multiple times using an existing non-linear solver. Since the relaxation we create is in general not convex, this method may not find an optimal solution. In order to mitigate this difficulty, we solve the relaxation multiple times in parallel starting from different initial points. Our preliminary computational experiments show that this approach gives optimal or near-optimal solutions on benchmark problems, and that the method benefits well from parallelism.

Decoupling network optimization in high speed systems by mixed-integer programming

Power Integrity is maintained in a high speed system by designing an efficient decoupling network. This paper provides a generic formulation for decoupling capacitor selection and placement problem which is solved by mixed-integer programming. A real-world example is presented for the same. The minimum number of capacitors that could achieve the target impedance over the desired frequency range are found along with their optimal locations. In order to solve an industrial problem, the s-parameters data of power plane geometry and capacitors are used for the accurate analysis including bulk capacitors and VRM.