Lu, Z.; McKinley, K.S.: ¬The effect of collection organization and query locality on information retrieval system performance (2000)
0.00
0.0016913437 = product of:
0.0033826875 = sum of:
0.0033826875 = product of:
0.006765375 = sum of:
0.006765375 = weight(_text_:a in 33) [ClassicSimilarity], result of:
0.006765375 = score(doc=33,freq=8.0), product of:
0.053105544 = queryWeight, product of:
1.153047 = idf(docFreq=37942, maxDocs=44218)
0.046056706 = queryNorm
0.12739488 = fieldWeight in 33, product of:
2.828427 = tf(freq=8.0), with freq of:
8.0 = termFreq=8.0
1.153047 = idf(docFreq=37942, maxDocs=44218)
0.0390625 = fieldNorm(doc=33)
0.5 = coord(1/2)
0.5 = coord(1/2)
- Abstract
- The explosion of content in distributed information retrieval (IR) systems requires new mechanisms in order to attain timely and accurate retrieval of text. Collection selection and partial collection replication with replica selection are two such mechanisms that enable IR systems to search a small percentage of data and thus improve performance and scalability. To maintain effectiveness as well as efficiency, IR systems must be configured carefully to consider workload locality and possible collection organizations. We propose IR system architectures that incorporate collection selection and partial replication, and compare configurations using a validated simulator. Locality and collection organization have dramatic effects on performance. For example, we demonstrate with simulation results that collection selection performs especially well when the distribution of queries to collections is uniform and collections are organized by topics, but it suffers when particular collections are "hot." We find that when queries have even modest locality, configurations that replicate data outperform those that partition data, usually significantly. These results can be used as the basis for IR system designs under a variety of workloads and collection organizations
- Type
- a
