Results 1 - 10 of 4,256

Table 2: BGP Update Matches for Loops using Sprint Link BGP Information

in On the Correlation between Route Dynamics and Routing Loops
by Ashwin Sridharan, Sue B. Moon, Christophe Diot 2003
"... In PAGE 6: ... This would preempt their association with any BGP event observed during the trace. From Table2 , it is evident that a significant number of persis- tent loops were present in some of the traces. For NYC-20 50.... ..."
Cited by 12

Table 6: Page download times for those pages where persistent connections were used. For comparison, page download times when using both serial and persistent connections are shown. Pages are categorized by the number of objects they contain, and divided by whether more or less than half of the objects in the pages were downloaded over persistent connections. All times are in milliseconds.

in Whole Page Performance
by Leeann Bent, Geoffrey M. Voelker 2002
"... In PAGE 13: ... Table 5 above showed that the primary reason is that most pages in the trace cannot benefit from the use of persis- tent connections. To focus on the impact of persis- tent connections on just those pages that can ben- efit from their use, Table6 shows the page down- load times for those pages that used persistent con- nections in the original trace. We categorize the pages according to the number of objects they con- tain, and compare the performance of using the se- rial and persistent connection optimization meth- ods.... In PAGE 13: ... We further divide the pages by whether more or less than half of the objects in the pages were downloaded over persistent connections. The page download times in Table6 show that the use of persistent connections does improve perfor- mance for those pages that can benefit from them. These are the pages one would expect: they have a substantial number of objects in the page, and most of those objects are downloaded over persis- tent connections.... ..."
Cited by 14

Table 4: Summary for the usage of HTTP persistent con- nections in ISI-1 traffic

in Rapid Model Parameterization from Traffic Measurements
by Kun-chan Lan, John Heidemann
"... In PAGE 9: ... Motivated by the increasingly important role of persis- tent connection in web traffic, as reported by previous study [50], we also model the persistent connection used in HTTP/1.1, As shown in Table4 , although only less than 20% of connections are persistent, they account for about 50% of all objects transferred and more than 20% of all bytes transferred. This clearly shows persistent connection... ..."

Table 1: Persistent Class Definitions

in Migrating a Leitstand System between Object-Oriented Database Systems - An Experience Report
by C. Huemer, G. Kappel, S. Vieweg, To Objectstore
"... In PAGE 12: ... In order to simulate this ONTOS functionality in ObjectStore each persis- tent class includes a static persistent class variable named extent of type os_Set containing all the instances of this class and of its subclasses, respectively. Table1 summarizes the persistent class definitions for ScheduleAgent in ONTOS and ObjectStore, respectively. 5.... ..."

Table 4: Summary for the usage of HTTP persistent con- nections in ISI-1 traffic

in Rapid Model Parameterization from Traffic Measurements
by Kun-chan Lan, John Heidemann 2002
"... In PAGE 9: ... Motivated by the increasingly important role of persis- tent connection in web traffic, as reported by previous study [50], we also model the persistent connection used in HTTP/1.1, As shown in Table4 , although only less than 20% of connections are persistent, they account for about 50% of all objects transferred and more than 20% of all bytes transferred. This clearly shows persistent connection plays an important role in the dynamics of TCP connections for the Web.... ..."

Table 5. Generic persistent step graph algorithm To use this algorithm we have to choose an instance, like for PG and CSG, for parameter functions A() and (). That is why from this generic algorithm a lot of exploration algorithm instances can be proposed. Some instances are studied in section 3.3 (PminSG), 3.4 (PSmaxG) and 3.6 (HPSG). But rst we prove that any instance of PSG preserves deadlocks. 3.2 Preservation of deadlocks The proof that the PSG preserves deadlocks of the state graph is similar to the one given for the CSG in [VAM96], it follows from a normalisation lemma. Normalisation operator N: Operator N extracts from all sequence w of en- abled transitions in a state s, a maximal step or a maximal pre x of a such step. N : S Step(T) T T 7! Step(T) T is de ned as follows:

in On combining the Persistent Sets Method with the Covering Steps Graph Method
by Pierre-olivier Ribet, Francois Vernadat, Bernard Berthomieu 2002
"... In PAGE 8: ... Persistent sets are subsets of transitions whose exploration is su cient to detect potential deadlocks, steps are used to re all these transitions \together quot; when possible. The algorithm skeleton is shown in Table5 , referred to as the PSG (Persis- tent Step Graph) algorithm in the sequel. Its layout is similar to that of algorithm CSG in Table 3.... In PAGE 10: ... PSG generalises PG Proof. Any PG can be seen as an PSG: Taking (P; o) = fftg j t 2 Pg in the PSG algorithm shown in Table5 , the graph generated is exactly that generated by the PG algorithm in Table 2, assuming both use the same function A(). So, clearly, the PSG may produce graphs whose size is smaller than those produced by PG.... ..."
Cited by 4

Table 6 Predicted total Kjeldahl N and forms of N and P ap plied to nine sprayfields based on effluent collected and pond effluent analysis in 4 months from seven north Florida dairies. Plant Nutrient Form and Type of Analysis Unfiltered Filtered Filtered Filtered Unfiltered Filtered Dairy N

in Normie Buehring
by William L. Kingery 1995
"... In PAGE 24: ... Conventional tillage incorporated YWC resulted in the highest okra yield on the Haufler farm. Transplanted okra tended to give greater yield compared to directseeded particularly in the incorpo rated YWC ( Table6 ). Okra establishmentwas very difficult because of seedling death and insect damage.... In PAGE 24: ...10 level of probability. Table6 . Cumulative total okra yield from application of 269 Mg/ha yard waste compost treatments, Green Acres, 1994.... In PAGE 94: ... Dick (1984) also found higher soil enzyme levels in NT soils than in CT soils. In the soybean study, both rye and vetch enhanced soil esterase, phosphatase, and aryl sul fatase activity compared to BG soils ( Table6 ). Soils from VC plots initially had significantly greater esterase and phos phatase activity than did soils from RC and BG plots; however, effects of the rye cover crop were more persis tent.... In PAGE 95: ...iol. Biochem. 22:1023-1027. Table6 . Effect of rye and vetch cover crops on soil enzyme ac tivities of a Dundee silt loam (0-2 cm), soybean study, 1994.... ..."

Table 6 Goal refinement predicates Goal refinement

in Requirements Engineering for Trust Management: Model, Methodology, and Reasoning
by Paolo Giorgini, Fabio Massacci, John Mylopoulos, Nicola Zannone
"... In PAGE 15: ... In partic- ular, our formal framework supports all phases of the require- ments analysis process described in the paper, including goal modeling. Thus, firstly, we introduce predicates for goal/task refinement and resource decomposition ( Table6 ). Predicate service(s) holds if s is a service.... ..."
Cited by 2

Table A1 Summative results of objective parameters Goal Evaluation parameter Group 1 (x) Group 2 (y) Variability

in 1 2 3 4 5 6 7 8
by Federica Cena, Ilaria Torre

Table 2: Agent Goals

in AN AGENT MODEL FOR DISTRIBUTED PART-SELECTION Timothy P. Darr and William P. Birmingham
by Advanced Technologies, Timothy P. Darr, William P. Birmingham
"... In PAGE 3: ... 2 Agent Goals In this agent model, goals achieve design properties2. Table2 shows the mapping from properties to goals. In most of the cases, this mapping is obvious.... ..."
Results 1 - 10 of 4,256