### TABLE I ANALYSIS FOR SMALL NETS IN THE MANhATTAN PLANE

1992

Cited by 76

### Table 1: Input le

1996

"... In PAGE 32: ...answer no of space dimensions 2 coarse partition [2,2] re nement [2,2] sweeps f[4,0] amp; [3,1] amp; [2,2] amp; [1,3] amp; [0,4]g Table1 0: Pre- and Postsmoothing, test1.i, test1b.... In PAGE 33: ...answer element type ElmT3n2D smoother basic method SOR smoother renumber unknowns fRenumNoUnknowns amp; RedBlackg Table1 1: Red-black vs. lexicographic Gauss-Seidel, test5.... In PAGE 33: ... (table 12, test2.i) menu item answer coarse partition [8,8] sweeps [1,1] coarse grid basic method fSOR amp; ConjGrad amp; GaussElimg coarse grid max iterations f1 amp; 10 g Table1 2: Coarse grid solution, test2.i Take a coarse grid with several unknowns.... In PAGE 34: ... (table 13, test3.i) menu item answer matrix type MatSparse coarse partition [16,16] coarse grid basic method GaussElim coarse grid renumber unknowns fRenumNoUnknowns amp; AMDhat amp; AMDbarg Table1 3: Direct coarse grid solution, test3.i Take a coarse grid with several unknowns and use a direct Gaussian elimination solver for sparse matrices.... In PAGE 34: ... (table 14, test4.i) menu item answer no of grid levels f2 amp; 3 amp; 4 amp; 5g coarse partition f[2,2] amp; [4,4] amp; [8,8] amp; [16,16]g Table1 4: Direct coarse grid solution, test4.i We now look at a V1;1 multigrid cycle with a direct coarse grid solver.... In PAGE 35: ...3 Semi-coarsening and non-standard re nement Exercise 13 Semi-coarsening (table 15, test6.i) menu item answer coarse partition f[16,2] amp; [2,16]g re nement f[1,2] amp; [2,1]g Table1 5: Semi-coarsening, test6.i Figure 12: Anisotropic grid, derived by semi-coarsening We now have a look at di erent ways of grid re ning.... In PAGE 36: ...(table 16, test7.i) menu item answer no of grid levels f2 amp; 3g coarse grid basic method GaussElim re nement f[2,2] amp; [4,4]g Table1 6: Non-bisecting re nement, test7.i Up to now we only have considered re nement via bisection which means a factor of two.... In PAGE 37: ... (table 17, prec1.i) menu item answer basic method ConjGrad preconditioning type PrecDD smoother basic method fSOR amp; SSOR amp; Jacobi amp; ConjGradg Table1 7: The smoother, prec1.i We now compare the performance of di erent smoothers.... In PAGE 38: ... (table 18, prec2.i) menu item answer sweeps [2,0] smoother basic method SOR basic method fBiCGStab amp; ConjGrad amp; CGSg preconditioning type PrecDD Table1 8: Non-symmetric preconditioner, prec2.i We now turn to non-symmetric Krylov methods.... In PAGE 39: ...xercise 18 Additive vs. multiplicative preconditioner. (table 19, prec3.i) menu item answer sweeps [1,1] basic method ConjGrad preconditioning type PrecDD domain decomposition method fMultigrid amp; AddMultigrid amp; NestedMultigridg Table1 9: Additive vs. multiplicative preconditioner, prec3.... ..."

Cited by 6

### Tab le 2DTable

"... In PAGE 21: ...0 401 Geode 1318 6549 302709 14194 2.1 795 Table 1 Statistics for various object-oriented environments Table2 and Table 3 present the total time and memory requirements for each of the class libraries from Table 1, for each of the table-based dispatch techniques on the best, worst and natural input orderings. The framework is implemented in C++, was compiled with g++ -O2, and executed on a Sparc-Station 20.... In PAGE 21: ...7 174.1 Table2 Timing Results for the DT Framework in seconds Library Order STI SC RD ICT SCCT best 10.6 1.... ..."

### Table 7: Correlation coe cientbetween le popularity and le duration.

2003

"... In PAGE 13: ... To determine whether there is a correlation between le duration and le popularity,we compute the corre- lation coe cientbetween le popularity and le duration for both of our workloads. Table7 shows these results. We use both the le frequency and the le rank as the popularity metric to compute the correlation coe cient.... ..."

Cited by 6

### Table 3. Mean for NE, LE and BSE NE LE BSE

"... In PAGE 5: ... Table3 shows the mean error, bias and variance across all the data sets for each of NE, LE and BSE. Fig-... ..."

### Table 3. Mean outcomes for NE, LE and BSE NE LE BSE

"... In PAGE 5: ... We compare the performance of these three algorithms using the method mentioned in Section 6. Table3 shows the mean error, bias and variance across all the data sets for each of NE (plain AODE), LE and BSE. Table 4 presents the win/draw/loss records for LE against the alternative algorithms on fty-six data sets.... ..."

Cited by 3

### Table 1. Pro le Examples

1999

Cited by 48