Gabriel Hernán Tolosa

Performance Improvements for Search Systems Using an Integrated Cache of Lists+Intersections

Modern information retrieval systems use several levels of caching to speedup computation by exploiting frequent, recent or costly data used in the past. In this study we propose and evaluate a static cache that works simultaneously as list and intersection cache, offering a more efficient way of handling cache space. In addition, we propose effective strategies to select the term pairs that should populate the cache. Simulation using two datasets and a real query log reveal that the proposed approach improves overall performance in terms of total processing time, achieving savings of up to 40% in the best case.

Analysis of Cost-Aware Policies for Intersection Caching in Search Nodes

We propose static, dynamic and hybrid cost-awarepolicies for intersection caching and we introduce three different strategies to solve a query computing list intersections. We run experiments over a simulation framework using real data for both document collection and text queries. We observe that: a) cost-aware policies outperform cost-oblivious policies, b) static policies are better than dynamic policies (as in posting list caching), c) hybrid policies outperform the previous ones achieving up to 29% of cost savings and d) computing strategies that try to maximize cache usage have a better performance than the a-priori most efficient ones.

3D inverted index with cache sharing for web search engines

Web search engines achieve efficient performance by partitioning and replicating the indexing data structure used to support query processing. Current practice simply partitions and replicates the text collection on the set of cluster processors and then constructs in each processor an index data structure. This paper proposes a different approach by constructing an index data structure that properly considers the fact that data is partitioned and replicated. This leads to a so-called 3D indexing strategy that outperforms current approaches. Performance is further boosted by introducing an application caching scheme devised to hold most frequently issued queries.

Revisión: tecnologia de agentes de software

Los agentes de software en la era de las redes globales son una herramienta vital para superar el fenomeno llamado “sobrecarga de informacion”. El grado de madurez alcanzado en esta tecnologia permite que hoy se puedan ver aplicaciones concretas funcionado en organizaciones, como asi tambien en el escritorio del usuario hogareno. El objetivo de este trabajo es presentar una revision bibliografica sobre la tecnologia de agentes de software, con orientacion a los modelos que permiten gerenciar la sobrecarga de informacion.

Performance improvements for search systems using an integrated cache of lists + intersections

Modern information retrieval systems use several levels of caching to speedup computation by exploiting frequent, recent or costly data used in the past. Previous studies show that the use of caching techniques is crucial in search engines, as it helps reducing query response times and processing workloads on search servers. In this work we propose and evaluate a static cache that acts simultaneously as list and intersection cache, offering a more efficient way of handling cache space. We also use a query resolution strategy that takes advantage of the existence of this cache to reorder the query execution sequence. In addition, we propose effective strategies to select the term pairs that should populate the cache. We also represent the data in cache in both raw and compressed forms and evaluate the differences between them using different configurations of cache sizes. The results show that the proposed Integrated Cache outperforms the standard posting lists cache in most of the cases, taking advantage not only of the intersection cache but also the query resolution strategy.