Automatic Compilation of MATLAB Programs for Synergistic Execution on Heterogeneous Processors

MATLAB is an array language, initially popular for rapid prototyping, but is now being increasingly used to develop production code for numerical and scientific applications. Typical MATLAB programs have abundant data parallelism. These programs also have control flow dominated scalar regions that have an impact on the program's execution time. Today's computer systems have tremendous computing power in the form of traditional CPU cores and throughput oriented accelerators such as graphics processing units(GPUs). Thus, an approach that maps the control flow dominated regions to the CPU and the data parallel regions to the GPU can significantly improve program performance. In this paper, we present the design and implementation of MEGHA, a compiler that automatically compiles MATLAB programs to enable synergistic execution on heterogeneous processors. Our solution is fully automated and does not require programmer input for identifying data parallel regions. We propose a set of compiler optimizations tailored for MATLAB. Our compiler identifies data parallel regions of the program and composes them into kernels. The problem of combining statements into kernels is formulated as a constrained graph clustering problem. Heuristics are presented to map identified kernels to either the CPU or GPU so that kernel execution on the CPU and the GPU happens synergistically and the amount of data transfer needed is minimized. In order to ensure required data movement for dependencies across basic blocks, we propose a data flow analysis and edge splitting strategy. Thus our compiler automatically handles composition of kernels, mapping of kernels to CPU and GPU, scheduling and insertion of required data transfer. The proposed compiler was implemented and experimental evaluation using a set of MATLAB benchmarks shows that our approach achieves a geometric mean speedup of 19.8X for data parallel benchmarks over native execution of MATLAB.

Spin State Selective Excitation of Single Quantum Transitions Using Multiple Quantum Spectroscopy- an Aid to Analyse Complex Proton NMR spectra of Oriented Molecules

NMR spectra of molecules oriented in liquid-crystalline matrix provide information on the structure and orientation of the molecules. Thermotropic liquid crystals used as an orienting media result in the spectra of spins that are generally strongly coupled. The number of allowed transitions increases rapidly with the increase in the number of interacting spins. Furthermore, the number of single quantum transitions required for analysis is highly redundant. In the present study, we have demonstrated that it is possible to separate the subspectra of a homonuclear dipolar coupled spin system on the basis of the spin states of the coupled heteronuclei by multiple quantum (MQ)−single quantum (SQ) correlation experiments. This significantly reduces the number of redundant transitions, thereby simplifying the analysis of the complex spectrum. The methodology has been demonstrated on the doubly 13C labeled acetonitrile aligned in the liquid-crystal matrix and has been applied to analyze the complex spectrum of an oriented six spin system.

Register Allocation and Optimal Spill code Scheduling in Software Pipelined Loops using 0-1 Integer Linear Programming Formulation

In achieving higher instruction level parallelism, software pipelining increases the register pressure in the loop. The usefulness of the generated schedule may be restricted to cases where the register pressure is less than the available number of registers. Spill instructions need to be introduced otherwise. But scheduling these spill instructions in the compact schedule is a difficult task. Several heuristics have been proposed to schedule spill code. These heuristics may generate more spill code than necessary, and scheduling them may necessitate increasing the initiation interval. We model the problem of register allocation with spill code generation and scheduling in software pipelined loops as a 0-1 integer linear program. The formulation minimizes the increase in initiation interval (II) by optimally placing spill code and simultaneously minimizes the amount of spill code produced. To the best of our knowledge, this is the first integrated formulation for register allocation, optimal spill code generation and scheduling for software pipelined loops. The proposed formulation performs better than the existing heuristics by preventing an increase in II in 11.11% of the loops and generating 18.48% less spill code on average among the loops extracted from Perfect Club and SPEC benchmarks with a moderate increase in compilation time.

Application-oriented adaptive MPI_Bcast for grids

Due to the importance of collective communications in scientific parallel applications, many strategies have been devised for optimizing collective communications for different kinds of parallel environments. There has been an increasing interest to evolve efficient broadcast algorithms for computational grids. In this paper, we present application-oriented adaptive techniques that take into account resource characteristics as well as the application's usage of broadcasts for deriving efficient broadcast trees. In particular, we consider two broadcast parameters used in the application, namely, the broadcast message sizes and the time interval between the broadcasts. The results indicate that our adaptive strategies can provide 20% average improvement in performance over the popular MPICH-G2's MPI_Bcast implementation for loaded network conditions.

Collocated and Distributed STBCs with Partial Interference Cancellation Decoding, Part II: Code Construction

In this second part of a two part series of papers, we construct a new class of Space-Time Block Codes (STBCs) for point-to-point MIMO channel and Distributed STBCs (DSTBCs) for the amplify-and-forward relay channel that give full-diversity with Partial Interference Cancellation (PIC) and PIC with Successive Interference Cancellation (PIC-SIC) decoders. The proposed class of STBCs include most of the known full-diversity low complexity PIC/PIC-SIC decodable STBCs as special cases. We also show that a number of known full-diversity PIC/PIC-SIC decodable STBCs that were constructed for the point-topoint MIMO channel can be used as full-diversity PIC/PIC-SIC decodable DSTBCs in relay networks. For the same decoding complexity, the proposed STBCs and DSTBCs achieve higher rates than the known low decoding complexity codes. Simulation results show that the new codes have a better bit error rate performance than the low ML decoding complexity codes available in the literature.

Improving Superscalar Instruction Dispatch And Issue By Exploiting Dynamic Code Sequences

Superscalar processors currently have the potential to fetch multiple basic blocks per cycle by employing one of several recently proposed instruction fetch mechanisms. However, this increased fetch bandwidth cannot be exploited unless pipeline stages further downstream correspondingly improve. In particular,register renaming a large number of instructions per cycle is diDcult. A large instruction window, needed to receive multiple basic blocks per cycle, will slow down dependence resolution and instruction issue. This paper addresses these and related issues by proposing (i) partitioning of the instruction window into multiple blocks, each holding a dynamic code sequence; (ii) logical partitioning of the registerjle into a global file and several local jles, the latter holding registers local to a dynamic code sequence; (iii) the dynamic recording and reuse of register renaming information for registers local to a dynamic code sequence. Performance studies show these mechanisms improve performance over traditional superscalar processors by factors ranging from 1.5 to a little over 3 for the SPEC Integer programs. Next, it is observed that several of the loops in the benchmarks display vector-like behavior during execution, even if the static loop bodies are likely complex for compile-time vectorization. A dynamic loop vectorization mechanism that builds on top of the above mechanisms is briefly outlined. The mechanism vectorizes up to 60% of the dynamic instructions for some programs, albeit the average number of iterations per loop is quite small.

Dielectric properties of c-axis oriented Zn1-xMgxO thin films grown by multimagnetron sputtering

Zn1−xMgxO (x = 0.3) thin films have been fabricated on Pt/TiO2/SiO2/Si substrates using multimagnetron sputtering technique. The films with wurtzite structure showed a (002) preferred orientation. Ferroelectricity in Zn1−xMgxO films was established from the temperature dependent dielectric constant and the polarization hysteresis loop. The temperature dependent study of dielectric constant at different frequencies exhibited a dielectric anomaly at 110 °C. The resistivity versus temperature characteristics showed an anomalous increase in the vicinity of the dielectric transition temperature. The Zn1−xMgxO thin films exhibit well-defined polarization hysteresis loop, with a remanent polarization of 0.2 μC/cm2 and coercive field of 8 kV/cm at room temperature.

Dielectric properties of (110) oriented PbZrO3 and La-modified PbZrO3 thin films grown by sol-gel process on Pt(111)/Ti/SiO2/Si substrate

Highly (110) preferred orientated antiferroelectric PbZrO3 (PZ) and La-modified PZ thin films have been fabricated on Pt/Ti/SiO2/Si substrates using sol-gel process. Dielectric properties, electric field induced ferroelectric polarization, and the temperature dependence of the dielectric response have been explored as a function of composition. The Tc has been observed to decrease by ∼ 17 °C per 1 mol % of La doping. Double hysteresis loops were seen with zero remnant polarization and with coercive fields in between 176 and 193 kV/cm at 80 °C for antiferroelectric to ferroelectric phase transformation. These slim loops have been explained by the high orientation of the films along the polar direction of the antiparallel dipoles of a tetragonal primitive cell and by the strong electrostatic interaction between La ions and oxygen ions in an ABO3 perovskite unit cell. High quality films exhibited very low loss factor less than 0.015 at room temperature and pure PZ; 1 and 2 mol % La doped PZs have shown the room temperature dielectric constant of 135, 219, and 142 at the frequency of 10 kHz. The passive layer effects in these films have been explained by Curie constants and Curie temperatures. The ac conductivity and the corresponding Arrhenius plots have been shown and explained in terms of doping effect and electrode resistance.

Digital Simulation of Field-Oriented Control of Induction Motor

This paper describes the simulation of a control scheme using the principle of field orientation for the control of a voltage source inverter-fed induction motor. The control principle is explained, followed by an algorithm to simulate various components of the system in the digital computer. The dynamic response of the system for the load disturbance and set-point variations have been studied. Also, the results of the simulation showing the behavior of field coordinates for such disturbances are given.

Microprocessor-Based Field-Oriented Control of A CSI-Fed Induction Motor Drive

This paper describes the method of field orientation of the stator current vector with respect to the stator, mutual, and rotor flux vectors, for the control of an induction motor fed from a current source inverter (CSI). A control scheme using this principle is described for orienting the stator current with respect to the rotor flux, as this gives natural decoupling between the current coordinates. A dedicated micro-computer system developed for implementing this scheme has been described. The experimental results are also presented.

Algebraic Approaches to Space-Time Code Construction for Multiple-Antenna Communication

A major challenge in wireless communications is overcoming the deleterious effects of fading, a phenomenon largely responsible for the seemingly inevitable dropped call. Multiple-antennas communication systems, commonly referred to as MIMO systems, employ multiple antennas at both transmitter and receiver, thereby creating a multitude of signalling pathways between transmitter and receiver. These multiple pathways give the signal a diversity advantage with which to combat fading. Apart from helping overcome the effects of fading, MIMO systems can also be shown to provide a manyfold increase in the amount of information that can be transmitted from transmitter to receiver. Not surprisingly,MIMO has played, and continues to play, a key role in the advancement of wireless communication.Space-time codes are a reference to a signalling format in which information about the message is dispersed across both the spatial (or antenna) and time dimension. Algebraic techniques drawing from algebraic structures such as rings, fields and algebras, have been extensively employed in the construction of optimal space-time codes that enable the potential of MIMO communication to be realized, some of which have found their way into the IEEE wireless communication standards. In this tutorial article, reflecting the authors’interests in this area, we survey some of these techniques.