Automated Meta-Control for Adaptable Real-Time Software

The software meta-controller is an online agent responsible for dynamically adapting an application's software configuration, e.g. altering operational modes and migrating tasks, to best accommodate varying runtime circumstances. In distributed real-time applications such adaptations must be carried out in a manner which maintains the schedulability of all critical tasks while maximizing some notion of system value for all other tasks. For large-scale real-time applications, considering all possible adaptations at the task-level is computationally intractable. This paper presents an automated aggregate approach to software meta-control, appropriate for large-scale distributed real-time systems. The aggregate automated meta-control problem is still NP-hard, but it has very practical approximate solutions. Introduced, here, are two very-effective approximation algorithms, QDP and GG, with very reasonable polynomial time complexity. Both algorithms also provide us with upper bounds for optimum system values, useful for deriving absolute, albeit somewhat pessimistic, measures of actual performance. Extensive Monte Carlo analysis is used to illustrate that expected performance for both algorithms is generally suboptimal by no more than a few percent. Our flexible software meta-control model is also shown to be readily applied to a wide range of time-sensitive applications.     

Cementing properties of oil shale ash: II. Moist curing of cast and compacted samples

This second paper in the series describes the results of an investigation with bed and cyclone oil shale ash samples prepared in a pilot plant fluidized bed at 650° and 760°C. Cast pastes and mortars have hydraulic properties and can attain strength values in the range of 50 to 150 kg/cm² after 28 days moist curing. The fact that this strength is lower than that of portland cement pastes and mortars, is mainly the result of the high water requirement of the ash, due to its high inherent fineness of about 15,000 cm²/gm. Relatively high strength values are obtained in compacted samples, by controlling their water content. 28 days values of about 300 and 400 kg/cm² of specimens compacted at pressure at 200 and 1500 kg/cm², respectively were achieved.     

Properties of type K expensive cement of pure components II. Proposed mechanism of ettringite formation and expansion in unrestrained paste of pure expansive component

A model is proposed to account for the behaviour of unrestrained paste hydrated pure K expansive component, based on growth of a porous ettringite layer around the C₄A₃S? grains, resulting in an expanding sphere. At the critical degree of hydration, α_cr, first contact is formed between the expanding spheres and they start pushing against each other, causing expansion. The model accounts quantitatively for the course of expansion and variation in porosity upon hydration and the variations of these parameters with temperature, as well as for the pore size distribution.     

Incipient fracture angle,fracture loci and critical stress for mixed mode loading

A prediction of the direction of incipient crack growth in brittle-like materials and the associated fracture loci under mixed mode loading is proposed. It is postulated that the direction of unstable crack propagation is determined by the “weakest” near-tip element defined as the one which would relax maximum potential energy upon prospective crack extension. Starting from the energy rate principle of crack extension (Eshelby energy-momentum tensor and Rice $\text{J-}\text{internal}$ vector) it is deduced that a crack will extent in the direction along which the following stress criterion is satisfied, $\text{(δ}{}_{\mathrm{\theta \theta }}{}^2\text{? δ}{}_{\mathrm{rr}}{}^2\text{) →}\text{maximum (for}\text{δ}{}_{\mathrm{\theta \theta \; >\; 0)}}$ The fracture angle in pure Mode II (70.4° away from the original straight path) is shown to be unstable in the sense that any slight tension along the crack (non-singular at the crack tip) affects considerably (up to 22%) the directionality of crack extension. It appears to be sensitive to the extent of the near-tip zone ($\text{r}{}_0$) in which linear elasticity does not hold and the non-singular stress term (squared).The fracture loci in mixed mode loading (generated by projecting the $\text{J-}\text{integral}$ vector along the prospective fracture path and letting this scalar function attain a critical value) is quadratic in $\text{K}{}_1$ and $\text{K}{}_2$ with an interactive cross product term $\text{K}{}_1\text{× K}{}_2$.The suggested criterion with its implication in predicting critical fracture load, exhibits behavior which is consistent with experimental observations collected from several sources. The common and uncommon features with respect to other known criteria are compared and discussed.     

Fracture conditions of a crack approaching a disturbance

The conditions which drive a crack to propagate in the vicinity of a cylindrical disturbance (cavity, bonded or unbonded inclusion) is expressed by linear fracture mechanics terms. A basic elastic solution with three levels of converging approximations is proposed and compared with finite element solution. The major finding is that a cavity under a given transverse (to cylindrical axis) tensile load attracts a nearby crack and promotes local fracture, whereas a rigid perfectly bonded inclusion repels the crack advancement and arrests local fracture. Unbonded or weakly bonded inclusion fall between those limits. Quantitative values of the relative stress intensity factor are given for large spectrum of geometries and elastic constants. A complementary numerical study is focused on the effect of a thin-layered binder around the inclusion on the stress intensity factor at the crack tip while approaching the binder. The conclusion is that the binder should be as thin as possible and with high shear modulus in order not to neutralize the local toughening effect of the inclusion.

---

Résumé Les conditions qui gouvernent la propagation d'une fissure au voisinage d'une discontinuité cylindrique (cavité, inclusion soudée ou non) peuvent être décrites par la théorie linéaire de la mécanique de rupture. On propose, et on compare à une solution par éléments finis, une solution élastique de base comportant trois niveaux d'approximations convergentes. On trouve essentiellement qu'une cavité sujette à une charge de traction transversalement à son axe cylindrique a pour effet d'attirer une fissure en son voisinage et provoquer une rupture locale. Par contre, une inclusion rigide et parfaitement soudée repousse l'avance de la fissure et provoque un arrêt de la rupture.Des inclusions non soudées ou partiellement soudées ont un effet situé entre ces deux limites. On fournit des valeurs quantitatives du facteur d'intensité des contraintes, pour un large spectre de géométries et de constantes élastiques. Une analyse numérique complémentaire est centrée sur l'effet d'un film de liaison autour de l'inclusion sur le facteur d'intensité des contraintes à l'extrémité de la fissure lorsque celle-ci avoisine cette zône de liaison.On en conclut que le film de liaison devrait être aussi mince que possible et posséder un module de cisaillement élevé, afin de neutraliser l'effet de durcissement local de l'inclusion.

     

Assessments of the kinetic and dynamic loads sustained by stationary tools during high rate plastic forming

A systematic method for evaluating the kinetic and dynamic loads sustained by stationary tools (as opposed to moving tools for which methods already exist) during high rate plastic forming is examined and exemplified by examples. It is essentially based on the momentum theorem for continua for incompressible flow, utilizing kinematically admissible velocity fields. In steady state forming processes (such as rolling, wire drawing, etc.), the difference between the active load (imposed or calculated a priori) and the reactive load, is formulated rigorously, whereas for non-steady processes (forging, impact extrusion, etc.) the formulation gives merely an approximation to the dynamic effects on the tools. The resulting velocity-dependent reactions on the tools are given in terms of two nondimensional numbers, namely, the “kinetic head” (u₀²/σ₀) (called the Euler Number) and the “dynamic head” (ú₀L/σ₀), which includes the machine speed (u₀), machine acceleration ( ), material density , yield strength ₀ and a characteristic dimension of the product, L. The same two non-dimensional heads emerged previously from energy-balance consideration in Ref. [1], while approximating dynamic loads on moving tools, hence a consistency is demonstrated. These heads are unavoidably multiplied by geometrical functions, which typify the specific process under consideration and may amplify (or diminish) the intensity of the dynamic effects. The present work is focussed on quantifying, by the above method, the inherent difference between the reactive load sustained by the non-moving tool (say, a die) and the acting load carried by the moving tool (piston, ram, etc.) In particular cases of very slow processes, these loads are equal by static equilibrium. In some practical processes (like rolling) their difference appears to be relatively small, whereas in others (like impact extrusion) it appears extremely large.     

Cementing properties of oil shale ash III. Curing in CO2 atmosphere

The third paper in this series describes the results of an investigation with oil shale ash compacts cured in CO₂ atmosphere to enhance initial and final strength. It was found that half hour compressive strength can be increased by a factor of three in CO₂ curing. Additional CO₂ curing was not followed by increase in strength, but further moist curing resulted in significant strengthening. Addition of 10% portland cement resulted in marked improvement in strength of the samples cured for 1/2 hr. in CO₂ and then in moist environment, giving values of 300 and 450 kg/cm² after 1/2 hour and 28 days respectively.     

Optical properties of solvent‐cast polarizer films for liquid‐crystal displays: A viscoelastic modeling framework

Abstract— Many of the films used in polarizer assemblies in LCDs (e.g., triacetyl cellulose or TAC) are produced by a solvent‐casting process, which is known to impart optical anisotropy to the film expressed as finite out‐of‐plane birefringence. This feature of the film could have a significant impact on the optical performance of the display and it needs to be accounted for in any compensation scheme for the LC cell. This paper reviews the origin of this optical anisotropy, and it presents a viscoelastic model that links this property to the solvent‐casting process and to some key material parameters. The model results are compared with experimental data generated for polystyrene films cast from toluene, and generally good agreement is demonstrated.     

Properties of type K expansive cement of pure components I. Hydration of unrestrained paste of expansive component — Results

A mixture of $\text{C}{}_4\text{A}{}_3\text{S}\text{, C}\text{S}\text{H}{}_2\text{and CH}$, in stoichiometric ratio to form ettringite only was paste hydrated at temperatures between 20–50°C. Unrestrained specimens were tested for expansion, porosity, strength and hydration at different times. Expansions started at a critical degree of hydration α_cr, at which total porosity was minimum and strength was maximum. Raising curing temperature reduced α_cr and enhanced expansion and total porosity. Mercury intrusion porosimetry revealed a bimodal pore size distribution.