### Table 3. Ternary Steiner problem: column-wise labelling.

"... In PAGE 16: ...able 3. Ternary Steiner problem: column-wise labelling. straints alone, with notable differences in the size of the search tree. Note for instance that n = 10 in Table3 could be proved to have no solution only with our algorithm, given the time limit 1 hour. Balanced Incomplete Block Designs (BIBD).... ..."

### Table 1. Time complexities among different methods Method Update time Prefix Sum[HAMS97] O(nd) Relative Prefix Sum[GAES99] O(nd/2)

2001

"... In PAGE 7: ...in the tree. Table1 shows the comparison of complexities among different methods. Here, we assume that N = nd and n is the number of cells in each dimension.... In PAGE 7: ...As an example of the comparison of time complexities shown in Table1 , Table 2 shows the number comparison of the update costs for various methods when the dimensionalities (d in Table) are 2, 4, and 8, and the size (n in Table 1) of each dimension is 101 and 102. For instance, when d = 4 and n = 102, the total size of a data cube N = nd is 108.... In PAGE 7: ...As an example of the comparison of time complexities shown in Table 1, Table 2 shows the number comparison of the update costs for various methods when the dimensionalities (d in Table) are 2, 4, and 8, and the size (n in Table1 ) of each dimension is 101 and 102. For instance, when d = 4 and n = 102, the total size of a data cube N = nd is 108.... ..."

Cited by 7

### Table 1. Time complexities among different methods Method Update time Prefix Sum[HAMS97] O(nd) Relative Prefix Sum[GAES99] O(nd/2)

2001

"... In PAGE 7: ...in the tree. Table1 shows the comparison of complexities among different methods. Here, we assume that N = nd and n is the number of cells in each dimension.... In PAGE 7: ...As an example of the comparison of time complexities shown in Table1 , Table 2 shows the number comparison of the update costs for various methods when the dimensionalities (d in Table) are 2, 4, and 8, and the size (n in Table 1) of each dimension is 101 and 102. For instance, when d = 4 and n = 102, the total size of a data cube N = nd is 108.... In PAGE 7: ...As an example of the comparison of time complexities shown in Table 1, Table 2 shows the number comparison of the update costs for various methods when the dimensionalities (d in Table) are 2, 4, and 8, and the size (n in Table1 ) of each dimension is 101 and 102. For instance, when d = 4 and n = 102, the total size of a data cube N = nd is 108.... ..."

Cited by 7

### Table 1: Speeds for matrix-vector multiplications in Fortran 77 (column-wise implementation) and for library routines.

### Table 9.1: Grouping of wavelengths to implement the row-wise and column-wise buses of a SBCH(4; 3) network. Rows/Columns Group1 Group2

65

### Table 2: Pointer Jumping Example

"... In PAGE 17: ... The algorithm is described in Figures 10 and 11. An example for p = 8 is shown in Table2 . The first step is to compute the parallel prefix sum from the number of bytes in each PE.... ..."

### Table 2. Ternary Steiner problem: row-and-column-wise labelling.

### Table 1. Benchmarking results for generic and problem-speciflc strategies for the SEND-MORE-MONEY crypto-arithmetic puzzle (column-wise deflnition; 25 solutions, i.e. leading zeros are allowed)

2003

"... In PAGE 10: ... The implemented example strategies have been tested with several difierent constraint problems exercising variations of constraint solver cooperation sets. Results for the SEND-MORE-MONEY problem runs behave representatively and are displayed in the upper half of Table1 . The strategies provide an optional parameter to allow the deflnition of a variable order that should be used for projection.... In PAGE 12: ... 3. Strategy speciflcation for the heuristic-solver- ow strategy The lower part of Table1 shows the performance data for four difierent speciflc strategies. The individual strategies all prioritize difierent types of con- straints and/or difierent solver instances and clearly show the advantage of problem-tailored strategies for our setting.... ..."

Cited by 3

### Table 2. Time steps spend and food items dropped in the different concentric zones. Simulation without spatial constraints. Table should be read column wise.

"... In PAGE 21: ...21 Relationship to world Conduct Cognitive agents (explicit awareness) Reactive agents (sub-explicit awareness) Teleonomic (goal driven) Intentional agents Drive-based agents Reflexes (perception driven) Module-based agents Tropistic agents Table2 -1.... ..."

### Tables 1 and 2 show the measured and predicted scalability of Jacobi algorithm with two di er- ent data distribution strategies: two-dimensional block distribution and column-wise distribution of all program arrays to a two-dimensional and one-dimensional processors array, respectively. The di erence in percentage between the predicted and measured values is given in the third column of the tables.

1999

Cited by 8