Optimal Reservoir Operation for Flood Control Using Folded Dynamic Programming

Folded Dynamic Programming (FDP) is adopted for developing optimalnreservoir operation policies for flood control. It is applied to a case study of Hirakud Reservoir in Mahanadi basin, India with the objective of deriving optimal policy for flood control. The river flows down to Naraj, the head of delta where a major city is located and finally joins the Bay of Bengal. As Hirakud reservoir is on the upstream side of delta area in the basin, it plays an important role in alleviating the severity of the flood for this area. Data of 68 floods such as peaks of inflow hydrograph, peak of outflow from reservoir during each flood, peak of flow hydrograph at Naraj and d/s catchment contribution are utilized. The combinations of 51, 54, 57 thousand cumecs as peak inflow into reservoir and 25.5, 20, 14 thousand cumecs respectively as,peak d/s catchment contribution form the critical combinations for flood situation. It is observed that the combination of 57 thousand cumecs of inflow into reservoir and 14 thousand cumecs for d/s catchment contribution is the most critical among the critical combinations of flow series. The method proposed can be extended to similar situations for deriving reservoir operating policies for flood control.

Software Pipelined Execution of Stream Programs on GPUs

The StreamIt programming model has been proposed to exploit parallelism in streaming applications on general purpose multi-core architectures. This model allows programmers to specify the structure of a program as a set of filters that act upon data, and a set of communication channels between them. The StreamIt graphs describe task, data and pipeline parallelism which can be exploited on modern Graphics Processing Units (GPUs), as they support abundant parallelism in hardware. In this paper, we describe the challenges in mapping StreamIt to GPUs and propose an efficient technique to software pipeline the execution of stream programs on GPUs. We formulate this problem - both scheduling and assignment of filters to processors - as an efficient Integer Linear Program (ILP), which is then solved using ILP solvers. We also describe a novel buffer layout technique for GPUs which facilitates exploiting the high memory bandwidth available in GPUs. The proposed scheduling utilizes both the scalar units in GPU, to exploit data parallelism, and multiprocessors, to exploit task and pipelin parallelism. Further it takes into consideration the synchronization and bandwidth limitations of GPUs, and yields speedups between 1.87X and 36.83X over a single threaded CPU.

Robust heuristics for scalable optimization of complex SQL queries

Modern database systems incorporate a query optimizer to identify the most efficient "query execution plan" for executing the declarative SQL queries submitted by users. A dynamic-programming-based approach is used to exhaustively enumerate the combinatorially large search space of plan alternatives and, using a cost model, to identify the optimal choice. While dynamic programming (DP) works very well for moderately complex queries with up to around a dozen base relations, it usually fails to scale beyond this stage due to its inherent exponential space and time complexity. Therefore, DP becomes practically infeasible for complex queries with a large number of base relations, such as those found in current decision-support and enterprise management applications. To address the above problem, a variety of approaches have been proposed in the literature. Some completely jettison the DP approach and resort to alternative techniques such as randomized algorithms, whereas others have retained DP by using heuristics to prune the search space to computationally manageable levels. In the latter class, a well-known strategy is "iterative dynamic programming" (IDP) wherein DP is employed bottom-up until it hits its feasibility limit, and then iteratively restarted with a significantly reduced subset of the execution plans currently under consideration. The experimental evaluation of IDP indicated that by appropriate choice of algorithmic parameters, it was possible to almost always obtain "good" (within a factor of twice of the optimal) plans, and in the few remaining cases, mostly "acceptable" (within an order of magnitude of the optimal) plans, and rarely, a "bad" plan. While IDP is certainly an innovative and powerful approach, we have found that there are a variety of common query frameworks wherein it can fail to consistently produce good plans, let alone the optimal choice. This is especially so when star or clique components are present, increasing the complexity of th- e join graphs. Worse, this shortcoming is exacerbated when the number of relations participating in the query is scaled upwards.

Solving MDPs using two-timescale simulated annealing with multiplicative weights

We develop extensions of the Simulated Annealing with Multiplicative Weights (SAMW) algorithm that proposed a method of solution of Finite-Horizon Markov Decision Processes (FH-MDPs). The extensions developed are in three directions: a) Use of the dynamic programming principle in the policy update step of SAMW b) A two-timescale actor-critic algorithm that uses simulated transitions alone, and c) Extending the algorithm to the infinite-horizon discounted-reward scenario. In particular, a) reduces the storage required from exponential to linear in the number of actions per stage-state pair. On the faster timescale, a 'critic' recursion performs policy evaluation while on the slower timescale an 'actor' recursion performs policy improvement using SAMW. We give a proof outlining convergence w.p. 1 and show experimental results on two settings: semiconductor fabrication and flow control in communication networks.

Solution of MDPS using simulation-based value iteration

This article proposes a three-timescale simulation based algorithm for solution of infinite horizon Markov Decision Processes (MDPs). We assume a finite state space and discounted cost criterion and adopt the value iteration approach. An approximation of the Dynamic Programming operator T is applied to the value function iterates. This 'approximate' operator is implemented using three timescales, the slowest of which updates the value function iterates. On the middle timescale we perform a gradient search over the feasible action set of each state using Simultaneous Perturbation Stochastic Approximation (SPSA) gradient estimates, thus finding the minimizing action in T. On the fastest timescale, the 'critic' estimates, over which the gradient search is performed, are obtained. A sketch of convergence explaining the dynamics of the algorithm using associated ODEs is also presented. Numerical experiments on rate based flow control on a bottleneck node using a continuous-time queueing model are performed using the proposed algorithm. The results obtained are verified against classical value iteration where the feasible set is suitably discretized. Over such a discretized setting, a variant of the algorithm of [12] is compared and the proposed algorithm is found to converge faster.

A hybrid computer for X-ray crystallography

A hybrid computer for structure factor calculations in X-ray crystallography is described. The computer can calculate three-dimensional structure factors of up to 24 atoms in a single run and can generate the scatter functions of well over 100 atoms using Vand et al., or Forsyth and Wells approximations. The computer is essentially a digital computer with analog function generators, thus combining to advantage the economic data storage of digital systems and simple computing circuitry of analog systems. The digital part serially selects the data, computes and feeds the arguments into specially developed high precision digital-analog function generators, the outputs of which being d.c. voltages, are further processed by analog circuits and finally the sequential adder, which employs a novel digital voltmeter circuit, converts them back into digital form and accumulates them in a dekatron counter which displays the final result. The computer is also capable of carrying out 1-, 2-, or 3-dimensional Fourier summation, although in this case, the lack of sufficient storage space for the large number of coefficients involved, is a serious limitation at present.

Incremental attribute evaluation through recursive procedures

Incremental semantic analysis in a programming environment based on Attribute Grammars is performed by an Incremental Attribute Evaluator (IAE). Current IAEs are either table-driven or make extensive use of graph structures to schedule reevaluation of attributes. A method of compiling an Ordered Attribute Grammar into mutually recursive procedures is proposed. These procedures form an optimal time Incremental Attribute Evaluator for the attribute grammar, which does not require any graphs or tables.

A four-timescale algorithm for constrained stochastic optimization of RED

The overall performance of random early detection (RED) routers in the Internet is determined by the settings of their associated parameters. The non-availability of a functional relationship between the RED performance and its parameters makes it difficult to implement optimization techniques directly in order to optimize the RED parameters. In this paper, we formulate a generic optimization framework using a stochastically bounded delay metric to dynamically adapt the RED parameters. The constrained optimization problem thus formulated is solved using traditional nonlinear programming techniques. Here, we implement the barrier and penalty function approaches, respectively. We adopt a second-order nonlinear optimization framework and propose a novel four-timescale stochastic approximation algorithm to estimate the gradient and Hessian of the barrier and penalty objectives and update the RED parameters. A convergence analysis of the proposed algorithm is briefly sketched. We perform simulations to evaluate the performance of our algorithm with both barrier and penalty objectives and compare these with RED and a variant of it in the literature. We observe an improvement in performance using our proposed algorithm over RED, and the above variant of it.

Dynamic programming based optimum non-uniform samples for speech reconstruction and coding

Non-uniform sampling of a signal is formulated as an optimization problem which minimizes the reconstruction signal error. Dynamic programming (DP) has been used to solve this problem efficiently for a finite duration signal. Further, the optimum samples are quantized to realize a speech coder. The quantizer and the DP based optimum search for non-uniform samples (DP-NUS) can be combined in a closed-loop manner, which provides distinct advantage over the open-loop formulation. The DP-NUS formulation provides a useful control over the trade-off between bitrate and performance (reconstruction error). It is shown that 5-10 dB SNR improvement is possible using DP-NUS compared to extrema sampling approach. In addition, the close-loop DP-NUS gives a 4-5 dB improvement in reconstruction error.

A generalized Linear Programming based approach to optimal divisible load scheduling

In this paper we propose a general Linear Programming (LP) based formulation and solution methodology for obtaining optimal solution to the load distribution problem in divisible load scheduling. We exploit the power of the versatile LP formulation to propose algorithms that yield exact solutions to several very general load distribution problems for which either no solutions or only heuristic solutions were available. We consider both star (single-level tree) networks and linear daisy chain networks, having processors equipped with front-ends, that form the generic models for several important network topologies. We consider arbitrary processing node availability or release times and general models for communication delays and computation time that account for constant overheads such as start up times in communication and computation. The optimality of the LP based algorithms is proved rigorously.

Computer modelling approach to study the modes of binding of alpha- and beta-anomers of D-galactose, D-fucose and D-glucose to L-arabinose-binding protein

The modes of binding of alpha- and beta-anomers of D-galactose, D-fucose and D-glucose to L-arabinose-binding protein (ABP) have been studied by energy minimization using the low resolution (2.4 A) X-ray data of the protein. These studies suggest that these sugars preferentially bind in the alpha-form to ABP, unlike L-arabinose where both alpha- and beta-anomers bind almost equally. The best modes of binding of alpha- and beta-anomers of D-galactose and D-fucose differ slightly in the nature of the possible hydrogen bonds with the protein. The residues Arg 151 and Asn 232 of ABP from bidentate hydrogen bonds with both L-arabinose and D-galactose, but not with D-fucose or D-glucose. However in the case of L-arabinose, Arg 151 forms hydrogen bonds with the hydroxyl group at the C-4 atom and the ring oxygen, whereas in case of D-galactose it forms bonds with the hydroxyl groups at the C-4 and C-6 atoms of the pyranose ring. The calculated conformational energies also predict that D-galactose is a better inhibitor than D-fucose and D-glucose, in agreement with kinetic studies. The weak inhibitor D-glucose binds preferentially to one domain of ABP leading to the formation of a weaker complex. Thus these studies provide information about the most probable binding modes of these sugars and also provide a theoretical explanation for the observed differences in their binding affinities.

Computer simulation of protein—carbohydrate complexes: application to arabinose-binding protein and pea lectin

The CCEM method (Contact Criteria and Energy Minimisation) has been developed and applied to study protein-carbohydrate interactions. The method uses available X-ray data even on the native protein at low resolution (above 2.4 Å) to generate realistic models of a variety of proteins with various ligands.The two examples discussed in this paper are arabinose-binding protein (ABP) and pea lectin. The X-ray crystal structure data reported on ABP-β-l-arabinose complex at 2.8, 2.4 and 1.7 Å resolution differ drastically in predicting the nature of the interactions between the protein and ligand. It is shown that, using the data at 2.4 Å resolution, the CCEM method generates complexes which are as good as the higher (1.7 Å) resolution data. The CCEM method predicts some of the important hydrogen bonds between the ligand and the protein which are missing in the interpretation of the X-ray data at 2.4 Å resolution. The theoretically predicted hydrogen bonds are in good agreement with those reported at 1.7 Å resolution. Pea lectin has been solved only in the native form at 3 Å resolution. Application of the CCEM method also enables us to generate complexes of pea lectin with methyl-α-d-glucopyranoside and methyl-2,3-dimethyl-α-d-glucopyranoside which explain well the available experimental data in solution.

A computer study of the miscibility limits of Pd-Cu-Si

Metallic glasses are of interest because of their mechanical properties. They are ductile as well as brittle. This is true of Pd77.5Cu6Si16.5, a ternary glassy alloy. Actually, the most stable metallic glasses are those which are alloys of noble or transition metals A general formula is postulated as T70–80G30-20where T stands for one or several 3d transition elements, and includes the metalloid glass formers. Another general formula is A3B to A5B where B is a metalloid. A computer method utilising the MIGAP computer program of Kaufman is used to calculate the miscibility gap over a range of temperatures. The precipitation of a secondary crystalline phase is postulated around 1500K. This could produce a dispersed phase composite with interesting high temperature-strength properties.

A computer study of the miscibility limits of Sb-Te-Bi

The quaternary system Sb1bTe1bBi1bSe with small amounts of suitable dopants is of interest for the manufacture of thermoelectric modules which exhibit the Peltier and Seebeck effects. This property could be useful in the production of energy from the thermoelectric effect. Other substances are bismuth telluride (Bi2Te3) and Sb1bTe1bBi and compounds such as ZnIn2Se4. In the present paper the application of computer programs such as MIGAP of Kaufman is used to indicate the stability of the ternary limits of Sb1bTe1bBi within the temperature ranges of interest, namely 273 K to 300 K.