### Table 1. Worst Case Delays by Independent Methods

"... In PAGE 10: ... Performance Comparison In this section, we compare the worst case delay derived by our integrated method with that by the independent method. Table1 lists the worst case delays derived by the independent methods [CRU9 1, RAH951. 5.... ..."

### Table 3. Delay testing results for the worst case.

2004

"... In PAGE 6: ... In addition, there is a delay associated with sending a message over a socket, which is significant for messages that require a short delay. Table3 and Fig. 5 show short delays, which are difficult for the emulator to faithfully apply, therefore we call it the worst case.... ..."

Cited by 1

### Table 1. Worst Case Delays by Independent Methods

"... In PAGE 20: ... Performance Comparison In this section, we compare the worst case delay derived by our integrated method with that by the independent method. Table1 lists the worst case delays derived by the independent methods [CRU91,RAH95]. For the sake of completeness, we include the details of these derivations in Appendix... In PAGE 30: ...Appendix B Here, we derive the worst case delays at the FIFO server, the priority driven servers, and the regulator by using the independent methods. These delays have been listed in Table1 . We copy it as follows: Delay at the Scheduler Delay at the Regulator The Total Delay Regulated Priority Driven Server G71 G70 G70 i i i i I G2B GD7 G2D G2D G2D 1 1 1 G71 G70 G70 G62 G72 i i i i i i I G2B GD7 G2D G2B G2D G2D 1 1 1 2 2 1 1 1 GD7 G2B GD7 GD7 G2D G2B G2D G2D G71 G70 G70 G62 G72 i i i i i i I Regulated FIFO Server G77 G70 G2D G2D GD7 ( ) 1 n Imax G77 G70 G62 G72 G2D G2D GD7 G2B ( ) 1 n i i Imax 2 2 1 GD7 G2D GD7 G2D GD7 G2B G77 G70 G62 G72 ( ) n i i Imax where G77 G62 G70 G70 G72 G3D G3D G3D G3D G3D G3D GE5 GE5 k k n k k i k n 1 1 1 = 0 and , i = 1, 2, .... ..."

### TABLE I WORST-CASE AVERAGE DELAY COMPARISON Implementation style Critical path Worst-case average delay (ns)

### Table 1: Worst-case, average, and best-case delay for aggregation algorithm worst case average best case

in Tilman Wolf

2001

Cited by 5

### Table 1: Worst-case, average, and best-case delay for aggregation algorithm worst case average best case

2001

Cited by 5

### Table 9: Worst-case delays with all combinations of input o sets.

1999

"... In PAGE 15: ...s (0.5,0.5), while the traditional phase combination of (0.0,0.0) is actually the worst possible. The worst-case delay is reduced by anywhere from 25% to 30% when the repeaters are placed with optimum phase. Finally, Table9 shows the same worst-case delays for the middle line, this time taken over all eight rise/fall combinations and all nine combinations of input waveform o sets. Again, even when the inputs do not switch perfectly simultaneously, the best phase combination is (0.... ..."

Cited by 10

### Table 9: Worst-case delays with all combinations of input o sets. 17

"... In PAGE 13: ...s (0.5,0.5), while the traditional phase combination of (0.0,0.0) is actually the worst possible. The worst-case delay is reduced by anywhere from 25% to 30% when the repeaters are placed with optimum phase. Finally, Table9 shows the same worst-case delays for the middle line, this time taken over all eight rise/fall combinations and all nine combinations of input waveform o sets. Again, even when the inputs do not switch perfectly simultaneously, the best phase combination is (0.... ..."

### Table 2: Worst-case switching delay associated with various interactive operations.

2000

"... In PAGE 26: ... In [35] the authors report one-way propagation delays from 30 to 50 milliseconds for a wide-area ATM network, and less than 10 milliseconds for ATM LAN connections. Table2 summarizes the worst-case switching delay for various types of requests. This delay is measured in real time (not the logical time of the movie).... ..."

Cited by 3

### Table 1: Worst-case switching delay associated with various interactive operations.

2000

"... In PAGE 21: ... In [28] the authors report one-way propagation delays from 30 to 50 milliseconds for a wide-area ATM network, and less than 10 milliseconds for ATM LAN connections. Table1 summarizes the worst-case switching delay for various types of requests. This delay is measured in real time (not the logical time of the movie).... ..."

Cited by 3