Space-Limited Ranked Query Evaluation Using Adaptive Pruning

Lester, Nicholas; Moffat, Alistair; Webber, William; Zobel, Justin

doi:10.1007/11581062_37

Cited by 24 publications

(21 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e docids in a posting list can be sorted in increasing order enabling the use of e cient compression algorithms and query processing [31]; or the posting lists can be frequency-sorted [39] or impact-sorted [2], allowing for good compression rates, but also presenting practical disadvantages such as their di culty of use for phrasal queries [22,37]. As such, in this paper, we focus on the more common search scenario of docid-sorted index layouts [15].…”

Section: Preliminariesmentioning

confidence: 99%

Efficient & Effective Selective Query Rewriting with Efficiency Predictions

Macdonald

Tonellotto

Ounis

2017

Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

To enhance e ectiveness, a user's query can be rewri en internally by the search engine in many ways, for example by applying proximity, or by expanding the query with related terms. However, approaches that bene t e ectiveness o en have a negative impact on e ciency, which has impacts upon the user satisfaction, if the query is excessively slow. In this paper, we propose a novel framework for using the predicted execution time of various query rewritings to select between alternatives on a per-query basis, in a manner that ensures both e ectiveness and e ciency. In particular, we propose the prediction of the execution time of ephemeral (e.g., proximity) posting lists generated from uni-gram inverted index posting lists, which are used in establishing the permissible query rewriting alternatives that may execute in the allowed time. Experiments examining both the e ectiveness and e ciency of the proposed approach demonstrate that a 49% decrease in mean response time (and 62% decrease in 95th-percentile response time) can be a ained without signi cantly hindering the e ectiveness of the search engine.

show abstract

Section: Preliminariesmentioning

confidence: 99%

Efficient & Effective Selective Query Rewriting with Efficiency Predictions

Macdonald

Tonellotto

Ounis

2017

Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

show abstract

“…Our algorithm combines the advantage of both previous descriptions. Finally, we have presented a skipping version of the space-limited pruning algorithm by Lester et al [7], which is the method used in the original description of pipelined query processing.…”

Section: Related Workmentioning

confidence: 99%

“…We use Algorithm 1 to describe the following processing step performed by the query processor. The space-limited pruning method itself has been originally presented by Lester et al [7] and the skipping optimization for a non-distributed index has already been presented in our previous work [5]. Therefore, the goal of this section is to describe the improvements to pipelined query processing.…”

Section: Improving Pipelined Query Processing With Skippingmentioning

confidence: 99%

See 1 more Smart Citation

Improving the performance of pipelined query processing with skipping—and its comparison to document-wise partitioning

Jonassen

Bratsberg

2013

World Wide Web

View full text Add to dashboard Cite

Abstract. Web search engines need to provide high throughput and short query latency. Recent results show that pipelined query processing over a term-wise partitioned inverted index may have superior throughput. However, the query processing latency and scalability with respect to the collections size are the main challenges associated with this method. In this paper, we evaluate the effect of inverted index skipping on the performance of pipelined query processing. Further, we introduce a novel idea of using Max-Score pruning within pipelined query processing and a new term assignment heuristic, partitioning by Max-Score. Our current results indicate a significant improvement over the state-of-the-art approach and lead to several further optimizations, which include dynamic load balancing, intra-query concurrent processing and a hybrid combination between pipelined and non-pipelined execution.

show abstract

“…• We use the Zettair in its default mode, which employs an early pruning strategy that dynamically limits the number of accumulators used for a query (see [9] for details). This is a crucial choice for the practicality of our proposal, since no real WSE would make a full evaluation and the CPU execution time clearly depends on the partial evaluation strategy employed in the system.…”

Section: The Setup For the Cost Measurementmentioning

confidence: 99%

Cost-Aware Strategies for Query Result Caching in Web Search Engines

2011

View full text Add to dashboard Cite

Abstract. Search engines and large scale IR systems need to cache query results for efficiency and scalability purposes. In this study, we propose to explicitly incorporate the query costs in the static caching policy. To this end, a query's cost is represented by its execution time, which involves CPU time to decompress the postings and compute the query-document similarities to obtain the final top-N answers. Simulation results using a large Web crawl data and a real query log reveal that the proposed strategy improves overall system performance in terms of the total query execution time.

show abstract

Space-Limited Ranked Query Evaluation Using Adaptive Pruning

Cited by 24 publications

References 9 publications

Efficient & Effective Selective Query Rewriting with Efficiency Predictions

Efficient & Effective Selective Query Rewriting with Efficiency Predictions

Improving the performance of pipelined query processing with skipping—and its comparison to document-wise partitioning

Cost-Aware Strategies for Query Result Caching in Web Search Engines

Contact Info

Product

Resources

About