ONE POPULAR HARDWARE DEVICE FOR PERFORMING FAST ROUTING LOOKUPS AND PACKET CLASSIFICATION IS A TERNARY CONTENT-ADDRESSABLE MEMORY (TCAM). WE PROPOSE TWO algorithms TO MANAGE THE TCAM SUCH THAT INCREMENTAL UPDATE TIMES...
详细信息
ONE POPULAR HARDWARE DEVICE FOR PERFORMING FAST ROUTING LOOKUPS AND PACKET CLASSIFICATION IS A TERNARY CONTENT-ADDRESSABLE MEMORY (TCAM). WE PROPOSE TWO algorithms TO MANAGE THE TCAM SUCH THAT INCREMENTAL UPDATE TIMES REMAIN SMALL IN THE WORST CASE.
Peaking filters enhance the sharpness impression in video images, but are often complex and have some image-quality drawbacks. We compare two spatial peaking filters and investigate the one that combines a simple stru...
详细信息
Peaking filters enhance the sharpness impression in video images, but are often complex and have some image-quality drawbacks. We compare two spatial peaking filters and investigate the one that combines a simple structure with good image quality. This simple structure makes our proposal a good candidate for software implementation on a programmable architecture. We evaluated the performance on a 32-bit(1) and a new 64-bit multimedia processor.
We present a user-level thread scheduler for shared-memory multiprocessors, and we analyze its performance under multiprogramming. We model multiprogramming with two scheduling levels: our scheduler runs at user-level...
详细信息
We present a user-level thread scheduler for shared-memory multiprocessors, and we analyze its performance under multiprogramming. We model multiprogramming with two scheduling levels: our scheduler runs at user-level and schedules threads onto a fixed collection of processes, while below this level, the operating system kernel schedules processes onto a fixed collection of processors. We consider the kernel to be an adversary, and our goal is to schedule threads onto processes such that we make efficient use of whatever processor resources are provided by the kernel. Our thread scheduler is a non-blocking implementation of the work-stealing algorithm. For any multithreaded computation with work T-1 and critical-path length T-infinity, and for any number P of processes, our scheduler executes the computation in expected time O(T-1 / P-A + T-infinity P / P-A), where P-A is the average number of processors allocated to the computation by the kernel. This time bound is optimal to within a constant factor, and achieves linear speedup whenever P is small relative to the parallelism T-1/T-infinity.
In this paper we present a simple parallel sorting algorithm and illustrate its application in general sorting, disk sorting, and hypercube sorting. The algorithm (called the (I, m)-mergesort (LMM)) is an extension of...
详细信息
In this paper we present a simple parallel sorting algorithm and illustrate its application in general sorting, disk sorting, and hypercube sorting. The algorithm (called the (I, m)-mergesort (LMM)) is an extension of the bitonic and odd-even mergesorts. Literature on parallel sorting is abundant. Many of the algorithms proposed, though being theoretically important, may not perform satisfactorily in practice owing to large constants in their time bounds. The algorithm presented in this paper has the potential of being practical. We present an application to the parallel disk sorting problem. The algorithm is asymptotically optimal (assuming that N is a polynomial in M, where N is the number of records to be sorted and M is the internal memory size). The underlying constant is very small. This algorithm performs better than the disk-striped mergesort (DSM) algorithm when the number of disks is large. Our implementation is as simple as that of DSM (requiring no fancy data structures or prefetch techniques.) As a second application, we prove that we can get a sparse enumeration sort on the hypercube that is simpler than that of the classical algorithm of Nassimi and Sahni [16]. We also show that Leighton's columnsort algorithm is a special case of LMM.
An adaptive control strategy for vehicle exhaust noise reduction considering acceleration characteristics is presented, and its simulation analysis system is developed in order to efficiently examine the performance o...
详细信息
An adaptive control strategy for vehicle exhaust noise reduction considering acceleration characteristics is presented, and its simulation analysis system is developed in order to efficiently examine the performance of the proposed control approach. The simulation system, which is implemented with the MATLAB/Simulink dynamic system analysis tools, and can be employed to evaluate the different control algorithms easily, plays an important role in the control system design. After many simulation tests, such as investigating the features of control trackability and stability under accelerated running condition, the practical application of the ANC system for vehicle exhaust noise reduction is introduced. (C) 2001 Society of Automotive Engineers of Japan, Inc. and Elsevier Science B.V. All rights reserved.
Uniform processing of V-shaped and tri-planar graphite/epoxy composite parts has been achieved in a single mode cavity using variable frequency microwave energy. A computer process control system was used to achieve u...
详细信息
Uniform processing of V-shaped and tri-planar graphite/epoxy composite parts has been achieved in a single mode cavity using variable frequency microwave energy. A computer process control system was used to achieve uniform heating and stable curing temperature. The control parameters were microwave frequency and power. The control algorithm was mainly composed of two parts: intelligent mode selection for uniform heating, and power control for constant curing temperature. Experimental results were compared between single mode heating, mode sweeping heating, and intelligent mode switching heating. The comparison showed significant improvement of temperature uniformity using the intelligent mode switching technique.
computer algorithms and graphical solutions for deflection-based closed-form back-calculation of rigid pavement properties are available. However, a solution in the form of regression equations would have useful pract...
详细信息
computer algorithms and graphical solutions for deflection-based closed-form back-calculation of rigid pavement properties are available. However, a solution in the form of regression equations would have useful practical applications because of the speed and convenience in computation. This work presents a regression model for this purpose based on the closed-form back-calculation algorithm NUS-BACK. The input parameters remain the same as the original closed-form algorithm, with no requirements for seed values or initial guess of pavement properties. The database for the development of the regression model was established by forward computation of deflections for practical ranges of pavement properties. The verification of the regression model is presented by checking against the solutions of the closed-form algorithm NUS-BACK. The reliability of the regression equations was assessed against the closed-form algorithm for two conditions of input deflections: one for exact deflection measurements with no errors, another for deflection measurements with random errors. The analysis showed that the regression model could provide comparable reliability and accuracy to the analytical closed-form solution by NUS-BACK.
We have developed an improved algorithm that allows us to enumerate the number of site animals on the square lattice LIP to size 46. We also calculate the number of lattice trees up to size 44 and the radius of gyrati...
详细信息
We have developed an improved algorithm that allows us to enumerate the number of site animals on the square lattice LIP to size 46. We also calculate the number of lattice trees up to size 44 and the radius of gyration of both lattice animals and trees up to size 42. Analysis of the resulting series yields an improved estimate, lambda =4.062570(8), for the growth constant of lattice animals, and, lambda (0)=3.795254(8), for the growth constant of trees, and confirms to a very high degree of certainty that both the animal and tree generating Functions have a logarithmic divergence. Analysis of the radius of gyration series yields the estimate, nu =0.64115(5), for the size exponent.
In this paper, we present, formalize and prove the correctness of recovery algorithm for our open and safe nested transaction model using I/O automaton model. Our nested transaction model uses the notion of a recovery...
详细信息
In this paper, we present, formalize and prove the correctness of recovery algorithm for our open and safe nested transaction model using I/O automaton model. Our nested transaction model uses the notion of a recovery point subtransaction in the nested transaction tree. It introduces a prewrite operation before each write operation to increase the potential concurrency. Our transaction model is termed as "open and safe" as prewrites allow early reads (before database writes on disk) without cascading aborts. The systems restart and buffer management operations are modelled as nested transactions to exploit possible concurrency during restart. Each non-access transaction, object, and the scheduler is modeled as I/O automaton. Each of these automata is specified with the help of some pre-and post-conditions. These pre- and post-conditions capture the operational semantics and the behavior of each automaton during recovery operations. Our proof technique makes use of assertional reasoning and provide many interesting invariant, thus gives a better understanding of our recovery algorithm.
The adaptive control system with reference models of local subsystems was considered. The proposed control algorithms were proved to provide zero errors of following the local reference models under a priori uncertain...
详细信息
The adaptive control system with reference models of local subsystems was considered. The proposed control algorithms were proved to provide zero errors of following the local reference models under a priori uncertainty of the parameters and interrelations of the subsystems. In doing so, only the measurable variables of the local subsystems were used to generate the control actions and in the adjustment algorithms, that is, control was totally decentralized.
暂无评论