"... In PAGE 3: ... 3.2 Fair Service Table1 compares with WFQ and WRR the ability of PP in providing fair service when a system is overloaded. A good fairness is said to be achieved if the server capacity is allocated in the max-min weighted fair share allocation manner, in which the server capacity is allocated in order of increasing demand (i.... In PAGE 3: ...apacity share (i.e. the experienced throughput) is simply CA CX BP D6 CX C8 C1 CYBPBD D6 CY BVBN (6) where BV is the server capacity, which is assumed to be 1 here, and D6 CX is the relative weight of class CX determined from (1) for PP, from (4) for WFQ, and from (5) for WRR. For Table1 , AQ BD BP AQ BF BP AQ BG BP BCBMBEBEBH and AQ BE BP BCBMBH; D6 CX BPBCBMBEBH for all classes. Clearly, for this case, although the system is overloaded, the load of some classes is less than the minimum average throughput which needs to be guar- anteed by a fair scheduling discipline.... In PAGE 3: ... Clearly, for this case, although the system is overloaded, the load of some classes is less than the minimum average throughput which needs to be guar- anteed by a fair scheduling discipline. Results for class 2 traffic in Table1... In PAGE 4: ... It can be verified that all WFQ, WRR and PP provide fair service in the max-min weighted fair share manner. Table1 also shows that a misbehaving class does not af- fect the average throughputs guaranteed to other classes if either WFQ or WRR or PP is adopted. Assume that class 2 for Table 1 had the same load intensity as other classes, i.... In PAGE 4: ... Table 1 also shows that a misbehaving class does not af- fect the average throughputs guaranteed to other classes if either WFQ or WRR or PP is adopted. Assume that class 2 for Table1 had the same load intensity as other classes, i.e.... In PAGE 4: ... Then, since AQBOBD, each class would have received an average throughput equal to BCBMBEBEBH. Comparing with the simulation results shown in Table1 , it is clear that although the load of class 2 traffic is increased to BCBMBH, it does not af- fect the average throughput of each other classes if any one of WFQ, WRR or PP is adopted. In contrast, if the adopted discipline is SP, there is a great reduction in rate provided to class 4 traffic.... ..."

"... In PAGE 9: ... 3DES has shown a better performance than IPSec under the same circumstances (see table 20). Table2 0: SSL Transfer Speed Algorithm Time (Sec) No Algorithm 2 3DES-EDE-CBC-SHA 9.8 DES-CBC-SHA 5.... In PAGE 10: ... In most cases IPSec doesn apos;t interoperate well, so both sides of the connection are required to havethe same vendor apos;s devices (see table 21). Table2 1: IPSec vs. SSL Function IPSec SSL Con guration hard easy Client Authentication must option Pre-Shared Key yes no Interoperability Problem yes no TCP Application Support all some UDP support yes no Throughput Rate high high Compression Support yes OpenSSL only HandshakeTime slow fast References [1] Sheila Frankel, quot;Demystifying the IPSec Puzzle quot;, Artec House Publisher, 2001 [2] Eric Rescorla, quot;SSL and TLS, Designing and Building Secure Systems quot;, Addison-Wesley, 3rd Printing, Aug.... ..."

"... In PAGE 41: ... To perform this part of the systems study, a list of ice protection systems and devices has been developed. Table4 shows the IPSs and sensors for which data have been found. Table 4 Ice Protection Systems and Sensors... ..."

"... In PAGE 10: ... 3.4 System calls APD operations are summarised in Table2 . These calls work as follows:... ..."

"... In PAGE 10: ... 3.4 System calls APD operations are summarised in Table2 . These calls work as follows:... ..."

"... In PAGE 32: ...2 Bottom-up approach: analysis of concurrent systems PN synthesis can also be used for the analysis of systems. Table3 describes the results of the ap- plication of our algorithms to the synthesis of Petri nets from TSs obtained from speed-independent circuits (all examples are described in [32]). This can be used to produce a user-readable description of the functionality of a circuit, in the form of a timing diagram-like labeled Petri net (a Signal Transi- tion Graph, STG).... In PAGE 32: ... Another potential application is to optimize the input to direct synthesis methods, that have been devised for Petri nets using both synchronous and asynchronous circuit design tech- niques ([39, 8, 44]). Figure 16 illustrates one of the examples shown in Table3 (a4 tflo1). Initially, a TS is derived by calculating all reachable states of the circuit (symbolic techniques can be used here).... ..."