### Table 1. Performance Characteristics of Different AM Implementations

1997

"... In PAGE 7: ...ficient, buffered writes in the SCI DSM only. Performance measurements on the UCSB SCI cluster show competitive performance behavior of the SCI AM system ( Table1 ). Our own implementation, depicted in the first row of Table 1, adds little over- head to the raw latency of 9.... ..."

Cited by 13

### Table 2 - Summary Statistics Comparison of Different Portfolio Selection

1996

"... In PAGE 7: ... The Markowitz critical line algorithm is also an undominated member of the second (SSD) and third (TSD) degree stochastic dominance sets. Place Table2 Here Using 120 months, the alternative screening methods that use mean-variance optimization (EV) or LPM optimization (LPM QP )after ranking and grouping exhibit higher R/SV and R/V ratios than the... ..."

Cited by 1

### Table 3 - Summary Statistics Comparison of Different Portfolio Selection

1996

"... In PAGE 8: ... When the number of observations is close to the number of assets, the alternatives are competitive with the Markowitz critical line algorithm and actually dominate it in the TSD efficient set. Place Table3 Here Tables 4 and 5 presents the results of breaking the 120 month period into two 60 month subperiods. Table 4 presents the R/SV results, while Table 5 provides the R/V results.... ..."

Cited by 1

### Table 2. Optimal parameter values of the Elman net

"... In PAGE 5: ... To find the optimal parameters of the net many simulations were performed. Table2 shows the investigated and the optimal values of the parameters separated in parameter of the network structure and the learning algorithm. Since the Elman net is a recurrent net- work, the information about previous locations is modelled in the context cells.... ..."

### TABLE I VALUES OF xi FOR RULE-BASE (AS A FRACTION OF THE LONG TERM AVERAGE RATE) We have not attempted to tune our fuzzy controller to pro- vide optimal performance because this appears to be very dif- ficult due to the many degrees of freedom associated with the membership functions, rule-base, and the parameters thereof. However, as we show later, any further tuning beyond the ba- sic intuitive ideas is not necessary and the fuzzy controller per- forms well. Tuning the fuzzy controller to provide optimal per- formance will be the subject of future research.

2002

Cited by 3

### Table 7 shows the number of virtual call sites for which the call graph contains more than one potential target method. Call sites with at most one potential target method can be converted to cheaper static instead of virtual calls, and they can be inlined, possibly enabling many other optimizations. Therefore, an analysis that proves that call sites are not polymorphic can be used to significantly improve run-time performance. object-sensitive call site

2006

"... In PAGE 11: ... Table7... ..."

Cited by 17

### Table VII shows the number of virtual call sites for which the call graph contains more than one potential target method. Call sites with at most one potential target method can be converted to cheaper static instead of virtual calls, and they can be inlined, possibly enabling many other optimizations. Therefore, an analysis that proves that call sites are not polymorphic can be used to significantly improve run-time performance.

2006

Cited by 17

### Table 5: Computational Requirements for Optimization

1997

"... In PAGE 10: ... 0 1 2 CSSO iteration 0 100 200 300 f Figure 12: Convergence History for case #10 Computational Resources The number of system and contributing anal- yses required to perform multidisciplinary design optimization is a primary concern. Table5 lists the average number of system analyses (SA apos;s) and contributing analyses (CA1, CA2) required to per- form optimization via the methods discussed in this paper. For CSSO/NN runs, it is indicated how many CA apos;s were required to perform the SA apos;s and how many were required during sub- space optimization.... In PAGE 10: ... The total number of CA apos;s required thus depends on how many iterations are performed. The results listed in Table5 demonstrate that the ability of CSSO to incorporate design data obtained from previous experience is bene cial. This is evidenced by the reduction in required SA apos;s and CA apos;s as more of such information is in- cluded.... ..."

Cited by 5

### Table 1: The overhead of event dispatching in secs. We use runtime code generation to implement a fast path to invoke guards and handlers from the dispatcher. Events are dispatched to 1, 5 and 10 handlers with and without guards. The rst column shows the overhad to directly invoke a Modula-3 procedure call. We nd that our runtime code generation and optimizations improve the performance of event dispatching by roughly a factor of four when compared to iterating over a table of func- tion pointers. Even with optimizations, though, dispatch overhead may be signi cant if many

1995

"... In PAGE 5: ... Finally, we use peephole optimizations to improve the quality of the generated code. Table1 shows the performance of our dispatcher under a variety of conditions. We show the performance of a dispatched event to between 1 and 10 handlers, with and without installed guards, and varying numbers of arguments.... ..."

Cited by 3

### Table 10: Optimization rules.

2001

"... In PAGE 17: ...15 The rewriting is performed modulo the following rules: x + y = y + x x +(y + z)=(x + y)+z (x y) z = x (y z) The optimization rules presented in Table10 are not needed to get the desired restricted syntactic form, but can be used to simplify the terms. They could be applied with higher priority than the rules in Table 9 to achieve possible reductions.... In PAGE 29: ... Of course, during the process of linearization many optimizations are conceivable, some of which can only be applied in a certain context. We have already mentioned some optimization rewrite rules ( Table10 ) that can be applied during one of the linearization steps. Another optimization can be performed in the cases where a new process name is introduced.... ..."

Cited by 19