Zhang, J. and H. Zhang: 2001, `SEM User's Guide'. Department of Computer Science, University of Iowa.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....tend to have high bandwidths [4, 16] This is thought to be primarily due to the startup overhead associated with the TCP slow start mechanism that probes the bandwidth at connection startup. Recent work has attempted to perform class based isolation of TCP flows [35] or startup optimizations [37, 38] to attempt to mitigate this problem. As a proof of concept, we found 5 10 percent improvement on average when using file size classification over when the entire history file was used to calculate a prediction. Graphs are presented in later sections to support these claims. For our GridFTP ....

Y. Zhang and L. Qiu and S. Keshav, Optimizing {TCP} Start-up Performance, Technical Report TR991731, Cornell University, Department of Computer Science, 1999.

....that generates test cases for unit testing and pair wise testing. Develop test drivers, test script driver for Java programs, test execution, and test result analysis. 19 Integrating the testing module with two other existing modules of TROMLAB framework: the Java code generation module [24], and the measurement tool that collects and validates the testing quality measurement [18] ....

L. Zhang. Master Thesis, Department of Computer Science, Concordia University, Montreal, Canada.

....to demand i or viceversa. Most applications of genetic algorithms to scheduling known to us avoid this problem by merely representing the order of jobs resp. tasks waiting for a machine and afterwards constructing the schedule by a separate module. 4] shows extensions to these representations but [26] cites various authors stating serious shortcomings of genetic algorithms applied to scheduling problems. Even when supplies are constructed, the search space remains very large. For a satisfactory solution it is mandatory that the search be guided by knowledge which is investigated further in ....

W. Zhang: Reinforcement Learning for Job-Shop Scheduling. PhD thesis, Oregon State University, Department of Computer Science, May 1996.

....The next two implementations, SATO (SAtisfiability Testing Optimized) 8] and LDPP linear list based Davis Putnam Prover) 8] came out in 1994. Another program released the same year was Mace (Models And CounterExamples) 5] Finally, SEM (System for Enumerating finite Models) was released in 1995[9, 10, 12]. The first four programs use the conversion based method, while SEM uses the direct approach. Recently, nonexhaustive search techniques have been successfully applied to some problems[11] THE MODEL GENERATION PROGRAM The new first order model generating program, called Stork, was designed to ....

Jian Zhang and Hantao Zhang. SEM user guide. Department of Computer Science, University of Iowa, 1995.

....ASM semantics for SDL (see [5] 6] provides a conceptual framework which is partly reused here. The dynamic semantics consists of four parts as can be seen in figure 11. These parts are: An SDL Abstract Machine (SAM) which is defined using ASMs (former work in this direction is reported in [13]) In the SAM there are basic features to express structure: ASM agents are enhanced by gates. These gates are the interfaces used by the SDL part. Moreover, connections are predefined here they will represent SDL channels and other SDL connections later. A last part of the SDL Abstract Machine ....

St. Lau and A. Prinz. BSDL: The Language -- Version 0.2. Department of Computer Science, Humboldt University Berlin, August 1995.

No context found.

F.C.M. Lau and S.H. Zhang, Optimal Gossiping in Paths and Cycles, Technical Report TR-97-10, Department of Computer Science, The University of Hong Kong, July 1997.

No context found.

H. Zhang: Effective Occlusion Culling for the interactive Display of Arbitrary Models, PhD. Thesis, University of North Carolina at Chapel Hill, Department of Computer Science, 1998.

....tend to have high bandwidths [4, 5, 19] This difference is thought to be primarily due to the startup overhead associated with the TCP start mechanism that probes the bandwidth at connection startup. Recent work has focused on class based isolation of TCP flows [43] and on startup optimizations [45, 46] to mitigate this problem. As a proof of concept, we found 5 10 percent improvement on average when using file size classification instead of the entire history file to calculate a prediction. This is shown in Figures 12 and 13, and discussed in Section 6. For our GridFTP transfer data we ran a ....

Y. Zhang. L. Qiu, and S. Keshav, Optimizing {TCP} Start-up Performance, Technical report TR99-1731, Department of Computer Science, Cornell University, 1999.

No context found.

M. Zhang, E. W. Chionh, R. N. Goldman. Hybrid Dixon Resultants, Technical Report TR 98-314, Rice University, Department of Computer Science, May, 1998.

....6M 1 6M D Gamma 1 for M label IRS, where M is from 2 to Theta( 3 p n) and from 2 to Theta( p n) respectively. In this paper, we improve on these results by proving two lower bounds using non planar graphs. The first lower bound, which is a generalization of the lower bound for 2 IRS in [14], is 3 2 D for any number of labels up to Theta(log n) the second lower bound is 5 4 D for any number of labels up to Theta( p n) It seems that for both 1 IRS and M IRS (M 1) the lower bound tends to be higher for non planar graphs. Together with the results presented in this paper, the ....

S.S.H. Tse and F.C.M. Lau, On Two-label Interval Routing, Technical Report TR-96-04, Department of Computer Science, The University of Hong Kong, 1996.

....at a black dot; if they cross, that means there is a clash on the use of the same edge. 3 The Mesh The algorithm we propose here for the mesh uses the first phase of the FujitaYamashita algorithm as its first phase (i.e. Fig. 1) For the second phase, the path gossiping algorithm introduced in [9] is used. By proving a lower bound on the gossiping time, we show that this path algorithm is optimal for the second phase here where a node has more than one token to distribute at the start of the second phase. An important feature of the combined algorithm is that it starts both phases at the ....

.... obtain an arbitrary token from v i 1 (L i 1 ) and store it in L i (if i 0) until done; each node v i on the right of the center node repeat ( similarly, except all the directions are reversed ) until done; Figure 3: The Phase 2 algorithm (for path) We use the path algorithm introduced in [9] here, which divides the row into two equal halves, with a center node, v c , in the middle, where c = bn=2c. The algorithm is as shown in Fig. 3. Let s focus on the left half; what goes on in the right half is symmetric. The operation of the algorithm can be seen as being composed of two gather ....

F.C.M. Lau and S.H. Zhang, Optimal Gossiping in Paths and Cycles, Technical Report TR-97-10, Department of Computer Science, The University of Hong Kong, July 1997.

No context found.

Zhang, J. and H. Zhang: 2001, `SEM User's Guide'. Department of Computer Science, University of Iowa.

No context found.

J. Zhang and H. Zhang. SEM User's Guide. Department of Computer Science, University of Iowa, 2001.

No context found.

T.L. Lau, Guided Genetic Algorithm, PhD Thesis, Department of Computer Science, University of Essex, Colchester, United Kingdom (1999).

No context found.

Lau, T.L., Guided Genetic Algorithm, PhD Thesis, Department of Computer Science, University of Essex, Colchester, UK, 1999

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC