Three parallel programming paradigms: Comparisons on an archetypal PDE computation

Three paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation -- a nonlinear, structured-grid partial differential equation boundary value problem -- using the same algorithm on the same hardware. All of the paradigms -- parallel languages represented by the Portland Group's HPF, (semi-)automated serial-to-parallel source-to-source translation represented by CAP-Tools from the University of Greenwich, and parallel libraries represented by Argonne's PETSc -- are found to be easy to use for this problem class, and all are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under any paradigm includes specification of the data partitioning, corresponding to a geometrically simple decomposition of the domain of the PDE. Programming in SPMD style for the PETSc library requires writing only the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global-to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm as a starting point, introduction of concurrency through subdomain blocking (a task similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Programming with CAPTools involves feeding the same sequential implementation to the CAPTools interactive parallelization system, and guiding the source-to-source code transformation by responding to various queries about quantities knowable only at runtime. Results representative of "the state of the practice" for a scaled sequence of structured grid problems are given on three of the most important contemporary high-performance platforms: the IBM SP, the SGI Origin 2000, and the CRAYY T3E.

Semi-Lagrange time integration for PDE models of asian options

Semi-Lagrange time integration is used with the finite difference method to provide accurate stable prices for Asian options, with or without early exercise. These are combined with coordinate transformations for computational efficiency and compared with published results

Image denoise by fourth-order PDE based on the changes of laplacian

Fourth-order partial differential equation (PDE) proposed by You and Kaveh (You-Kaveh fourth-order PDE), which replaces the gradient operator in classical second-order nonlinear diffusion methods with a Laplacian operator, is able to avoid blocky effects often caused by second-order nonlinear PDEs. However, the equation brought forward by You and Kaveh tends to leave the processed images with isolated black and white speckles. Although You and Kaveh use median filters to filter these speckles, median filters can blur the processed images to some extent, which weakens the result of You-Kaveh fourth-order PDE. In this paper, the reason why You-Kaveh fourth-order PDE can leave the processed images with isolated black and white speckles is analyzed, and a new fourth-order PDE based on the changes of Laplacian (LC fourth-order PDE) is proposed and tested. The new fourth-order PDE preserves the advantage of You-Kaveh fourth-order PDE and avoids leaving isolated black and white speckles. Moreover, the new fourth-order PDE keeps the boundary from being blurred and preserves the nuance in the processed images, so, the processed images look very natural.

Excellent correlation between cathepsin B inhibition and cytotoxicity for a series of palladacycles

The reaction of the five- or six-membered C,N or C,S-palladacycles [(L)PdCl](2) with PTA (1,3,5-triaza-7-phosphaadamantane) led to the monomeric complexes [(L)Pd(PTA)Cl] 6a, 6b and 7 where LH= N,N-dimethyl-1-phenylmethanamine, benzyl(methyl)sulfane or 1-methyl-5-phenyl-1H-benzo[e][1,4]diazepin-2(3H)-one respectively. Dimeric complexes have also been synthesised: [Pd(2)L(2)(mu-dppe)Cl(2)], where LH = 1-methyl-5-phenyl-1H-benzo[e][1,4]diazepin-2(3H)-one (1a), (R)- or (S)-3-isopropyl-1-methyl-5-phenyl-1H-benzo[e][1,4]diazepin-2(3H)-one (1b, 1c), [Pd(2)L(2)(mu-dppf)Cl(2)], where L= 1-methyl-5-phenyl-1H-benzo[e][1,4]diazepin-2(3H)-one (4a) or (R)-3-isopropyl-1-methyl-5-phenyl-1H-benzo[e][1,4]diazepin-2(3H)-one (4b), respectively, and dppe = 1,2-bis(diphenylphosphino)ethane, dppf = 1,1'-bis(diphenylphosphino)ferrocene. The complexes were characterised in solution, by (1)H and (31)P NMR spectroscopy, and single crystals of complexes 6b and 7 were studied in the solid state by X-ray crystallography. The palladacycles were evaluated for in vitro activity as cytotoxic agents on A2780/S cells and also as cathepsin B inhibitors, an enzyme implicated in a number of cancer related events.

Inhibition of prandial and waterspray-induced rat grooming by 8-OH-DPAT

In addition to its hyperphagic effect in rats, 8-OH-DPAT also reduces grooming, but it is uncertain whether the inhibition of grooming is a specific effect or a consequence of response competition from eating. The present experiments explored the effects of 8-OH-DPAT on periprandial grooming and grooming elicited by spraying rats with water. Momentary time sampling over 30 or 60 min, with behaviour scored in one of 6 or 7 (depending on food availability) mutually exclusive categories (feeding, active, scratching, face grooming, body grooming, genital grooming and resting) at 15s intervals, was used for data collection. Non-deprived rats were tested in the presence and absence of food and baseline grooming levels were manipulated by spraying the dorsal surface of the back with water. Data were submitted to ANOVA. The first experiment confirmed that 8-OH-DPAT increased food intake and that this was associated with a parallel increase in feeding observations; active observations were also increased, but resting and total grooming observations were reduced: scratching was reduced even at 0.003mg/kg, face- and body-grooming were reduced at doses > 0.03mg/kg and genital-grooming was least sensitive, only being reduced at 0.1mg/kg. The second experiment revealed that spraying with water had no effect on food intake, feeding or resting observations, but increased total grooming (largely due to increased body-grooming) and reduced activity observations. In rats sprayed with water, 8-OH-DPAT increased food intake (0.1mg/kg) and observations of feeding (0.003 & 0.1mg/kg), but total grooming was dose-dependently inhibited, with genital-grooming most sensitive(> 0.003mg/kg), followed by face-grooming (> 0.01mg/kg) and body-grooming (>0.03mg/kg), whilst low levels of scratching were unaffected. The final experiment tested water-sprayed rats in the absence of food: 8-OH-DPAT increased resting and reduced total grooming, mostly as a consequence of reductions in face- and body-grooming, but there were also modest reductions in scratching. These results confirm that 8-OH-DPAT has a suppressant effect on all aspects of grooming, except where there are probable floor effects, and that this is independent of response competition from increased eating.

Inhibition of prandial and waterspray-induced rat grooming by 8-OH-DPAT

The effects of 8-OH-DPAT treatment on rat grooming behaviour, elicited either prandially or in response to spraying with water were investigated. Dose (≤0.1 mg/kg s.c.) response studies employed momentary time sampling over 30 or 60 min with behaviour being scored in one of 6 or 7 (depending on food availability) mutually exclusive categories (feeding, active, scratching, face-grooming, body grooming, genital-grooming and resting) at 15 s intervals. In non-deprived rats, tested with wet mash available, feeding and activity frequencies were increased, but resting and total grooming were inhibited by 8-OH-DPAT. Face-, body- and genital-grooming occurred at higher levels than scratching, but all categories were reduced with reductions in scratching occurring at a lower dose (0.01 mg/kg). Misting rats with a fine water spray selectively increased body grooming and decreased activity without altering feeding, while 8-OH-DPAT increased feeding and reduced face-, body- and genital-grooming, without affecting already low levels of scratching. In misted rats, tested without food, 8-OH-DPAT reduced face-, body- and genital-grooming and increased resting. These results confirm i) that the water spray technique is a useful method for increasing grooming and ii) that 8-OH-DPAT has a suppressant effect on grooming independent of response competition from enhanced feeding.