Sharif Digital Repository

Data prefetching is a well-known technique to hide the memory latency in the last-level cache (LCC). Among many prefetching methods in recent years, the Global History Buffer (GHB) proves to be efficient in terms of cost and speedup. In this paper, we show that a fixed value for detecting patterns and prefetch degree makes GHB to (1) be conservative while there are more opportunities to create new addresses and (2) generate wrong addresses in the presence of constant strides. To resolve these problems, we separate the pattern length from the prefetching degree. The result is an aggressive prefetcher that can generate more addresses with a given pattern length. Furthermore with a variable... 

Emerging cloud workloads in today's modern data centers have large memory footprints that make the processor's caches to be ineffective. Since L1 data cache is in the critical path, high data cache miss rates degrade the performance. To fix the issue in traditional workloads, data prefetchers predict the needed data to hide the memory latency and ultimately improve performance. In this paper, we focus on the L1 data cache to answer the question on why state-of-the-art prefetching methods are inefficient for modern workloads in terms of performance and energy consumption? This is because L1 cache is the most important player affecting the processor performance. Results show that, on the one... 

The energy consumption for processing modern workloads is challenging in data centers. Due to the large datasets of cloudworkloads, the miss rate of the L1 data cache is high, andwith respect to the energy efficiency concerns, suchmisses are costly formemory instructions because lower levels ofmemory hierarchy consume more energy per access than the L1. Moreover, large last-level caches are not performance effective, in contrast to traditional scientific workloads. The aim of this article is to propose a large L1 data cache, called Domino, to reduce the number of accesses to lower levels in order to improve the energy efficiency. In designing Domino, we focus on two components that use the... 

Overlay multicast has gained much attention in recent years as an alternative method to network layer multicast, especially for mobile ad hoc networks (MANETs). In this paper, we propose a new overlay multicast protocol to achieve simplicity of deployment, rapid adaptation of overlay structure when nodes move, and reduced delay. In our algorithm, to join or leave a multicast group, it is only sufficient for a member node to inform its first upstream member node. This updates the tree structure more rapidly when nodes move. In addition, join and leave delays are reduced and this makes the protocol suitable for multimedia multicasting in MANETs. Simulation results compared to that of ODMRP and... 

In this paper, a new method for path planning is proposed using a genetic algorithm (GA). Our method has two key advantages over existing GA methods. The first is a novel environment representation which allows a more efficient method for obstacles dilation in comparison to current cell based approaches that have a tradeoff between speed and accuracy. The second is the strategy we use to generate the initial population in order to speed up the convergence rate which is completely novel. Simulation results show that our method can find a near optimal path faster than computational geometry approaches and with more accuracy in smaller number of generations than GA methods. © 2009 IEEE  

In this paper, we propose a new method for mobile robot path planning in dynamic environment when the trajectories of obstacles are unknown. Our algorithm first utilizes a global approach called clearance based probabilistic roadmap method to find a suitable path and then locally apply evolutionary algorithm to keep the structure of the path when obstacles collide with the path. As a result, the path will act like an elastic band. To reach real time applicability, a light fitness function is proposed compare to other genetic algorithms to reduce the computation time needed for calculating and repairing path. Simulation results show that our method not only can keep the original structure of... 

This paper investigates the efficiency and performance of several of the fiooding techniques popular in secondary and enhanced oil recovery. These processes include water fiooding, immiscible and miscible gas fiooding, and several forms of water-alternating-gas (WAG) injection. The study is carried out numerically on a typical one-dimensional domain using an advanced high-resolution central scheme. Some performance indices such as oil cut drop, breakthrough time, and sweep efficiency were employed to evaluate the appropriateness of each injection strategy  

Purpose - The purpose of this paper is to investigate the performance of a specific class of high-resolution central schemes in conjunction with the black oil models for hydrocarbon reservoir simulation. Design/methodology/approach - A generalized black oil model is adopted, in which the solubility of gas in both oil and water and evaporation of oil are considered, leading to a system of equations prone to degeneracy. A computer code is generated and three test cases are solved to evaluate the performance of various schemes in terms of accuracy and discontinuity handling. Findings - It is shown that, although some of the central schemes are highly sensitive to the choice of... 

In this paper, two different important quality metrics, clearance and execution time, for motion planning of mobile robots are studied. Then a new approach, called Clearance-Based Probabilistic Roadmap Method (CBPRM (c)), is proposed which in contrast to VV(c) method, Visibility-Voronoi with clearance c, beside satisfaction of a predefined clearance is very efficient in running time. This provides the CBPRM (c) to be applicable in dynamic environments which require realtime approaches. Experiments showed that the CBPRM(c) is able to generate paths with very low execution time considering determined clearance. © 2008 IEEE  

Server workloads operate on large volumes of data. As a result, processors executing these workloads encounter frequent L1-D misses. In a many-core processor, an L1-D miss causes a request packet to be sent to an LLC slice and a response packet to be sent back to the L1-D, which results in high overhead. While prior work targeted response packets, this work focuses on accelerating the request packets. Unlike aggressive OoO cores, simpler cores used in many-core processors cannot hide the latency of L1-D request packets. We observe that LLC slices that serve L1-D misses are strongly temporally correlated. Taking advantage of this observation, we design a simple and accurate predictor. Upon... 

Four 5-, 10-, 20- and 30-story moment frames, representing low-, mid-, and two high-rise structures, were subjected to a great number of idealized directivity pulses. The amplitudes and periods of pulses vary from 0.02 g to 1.0 g and 0.5 to 12 sec, respectively. Over 1400 nonlinear dynamic analyses of low- to high-rise moment frames were performed which were feasible through using modified fish-bone model. The distribution of interstory drift along the height was studied and two applied contours were proposed: (i) the maximum interstory drift contour, and (ii) the critical story contour. These contours were demonstrated versus the ratio of natural period of the structure to the pulse period... 

In this article singular points of a parallel manipulator are obtained based on a novel geometrical method. Here we introduce the constrained plain method (CPM) and some of its application in parallel mechanism. Given the definition of constraint plane (CP) and infinite constraint plane (ICP) the dependency conditions of constraints is achieved with the use of a new theorem based on the Ceva geometrical theorem. The direction of angular velocity of a body is achieved by having three ICPs with the use of another theorem. Finally, with the use of the above two novel theorems singularities of the 3UPF_PU mechanism are obtained. It should be emphasized that this method is completely geometrical,... 

An analytical model is proposed to study the effect of particle size on melting enthalpy and entropy of metallic nanoparticles (NPs). The Mott's and Regel's equations for melting entropy in the combination of core average coordination number (CAC) and surface average coordination number (SAC) of freestanding NPs are considered. Clusters of icosahedral (IC), body centered cubic (BCC), and body centered tetragonal (BCT) structure without any vacancies and defects are modeled. Using the variable coordination number made this model to be in good agreement with experimental and molecular dynamic (MD) results of different crystal structures. The model predicts melting entropy and enthalpy of... 

Grinding is widely used for manufacturing of components that require fine surface finish and good dimensional accuracy. In this study a thermo-mechanical finite element analysis is conducted to find out how grinding parameters can affect temperature and residual stress distribution in the workpiece. Results of parametric study presented in this work indicate, by carefully selecting the grinding parameters, minimum thermal and mechanical damage can be achieved. Higher workpiece velocities produce higher surface residual stress. By increasing depths of cut, depth of tensile residual stresses increases. Convection heat coefficient does not have any considerable effect on surface residual stress... 

Adaptive Noise Cancellation (ANC) is a well-known technique for background noise reduction in automobile and vehicular environments. The noise fields in automobile and other vehicle interior obey the diffuse noise field model closely. On the other hand, the ANC does not provide sufficient noise reduction in the diffuse noise fields. In this paper, a new multistage post-filter is designed for ANC as a solution to diffuse noise conditions. The designed post-filter is a single channel Linear Prediction (LP) based speech enhancement system. The LP is performed by an adaptive lattice filter and attempts to extract speech components by using intermediate ANC signals. The post-filter has no... 

Shot peening is widely used to improve the fretting fatigue strength of critical surfaces. Fretting fatigue occurs in contacting parts that are subjected to fluctuating loads and sliding movements at the same time. This paper presents a sequential finite element simulation to investigate the shot peening effects on normal stress, shear stress, bulk stress and slip amplitude, which are considered to be the controlling parameters of fretting damage. The results demonstrated that among the modifications related to shot peening, compressive residual stress has a dominant effect on the fretting parameters  

One-dimensional (1D) MgO structures were successfully synthesized via carbothermic reduction of mechanically activated mixture of MgO and graphite. Mechanical activation of source materials before carbothermic reduction can substantially enhance the formation of MgO products at a temperature (1000 °C) relatively lower than that required in previous approaches (≥1200 °C). However, the morphology of MgO formed is dependent on the degree of mechanical activation and the condition of the subsequent carbothermic reduction. Two distinctive morphologies were found for MgO products synthesized using our method: single crystalline nanorods with rectangular cross-sections whose diameters range from 50... 

Attitude and vibration control of a general form of flexible satellites is addressed in this paper. Partial differential dynamic equations are derived considering new details such as multi sectioned solar panels and elastic connections between main hub and solar panels. Boundary control approach is adopted to eliminate simplification errors of discrete models, using just one actuator in the hub. Asymptotic stability of attitude dynamics is proved for a group of boundary controllers and necessary conditions for asymptotic stability of vibrations are discussed. Being independent of modeling accuracy and using easily measurable feedbacks are among advantages of the proposed class of... 

Characterization and calcination behavior of a low-grade manganese ore, as a part of Mn ferroalloys production, was studied by XRF, ex-situ XRD, in-situ XRD, and SEM-EDS techniques. Calcination experiments were carried out at and up to 900 °C (1173 K) in air and argon atmospheres. The samples were in particles and powder forms. The results indicated that both quartz and calcite phases in the ore exhibit a bimodal spatial distribution; as relatively large regions and finely distributed in the Mn- and Fe-containing phases. By Rietveld analysis of the in-situ XRD data, the reactions occurring upon heating during the calcination process were deduced. Thermal decomposition and reactive diffusion...