Exploiting Global Impact Ordering for Higher Throughput in Selective Search

Siedlaczek, Michał; Rodriguez, Juan; Suel, Torsten

doi:10.1007/978-3-030-15719-7_2

Cited by 3 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent work, Mohammad et al [187] showed that the ideas presented in this chapter (and in the work of Culpepper et al [75]) can effectively solve the problem of dynamic shard cutoff prediction, where a predictor is built to decide how many index shards must be examined to achieve effective results in distributed multi-stage retrieval architectures. It would be interesting to see if this idea could be deployed in tandem with predicting k, especially in the selective search architectures explored by Siedlaczek et al [227] and Hafizoglu et al [107]. More recently, similar ideas were used by Petri et al [197] to train a model which can predict the final threshold of a dynamic pruning algorithm to accelerate query processing without requiring relevance judgments.…”

Section: Discussionmentioning

confidence: 99%

Managing tail latency in large scale information retrieval systems

Mackenzie

2020

SIGIR Forum

View full text Add to dashboard Cite

As both the availability of internet access and the prominence of smart devices continue to increase, data is being generated at a rate faster than ever before. This massive increase in data production comes with many challenges, including efficiency concerns for the storage and retrieval of such large-scale data. However, users have grown to expect the sub-second response times that are common in most modern search engines, creating a problem --- how can such large amounts of data continue to be served efficiently enough to satisfy end users? This dissertation investigates several issues regarding tail latency in large-scale information retrieval systems. Tail latency corresponds to the high percentile latency that is observed from a system --- in the case of search, this latency typically corresponds to how long it takes for a query to be processed. In particular, keeping tail latency as low as possible translates to a good experience for all users, as tail latency is directly related to the worst-case latency and hence, the worst possible user experience. The key idea in targeting tail latency is to move from questions such as "what is the median latency of our search engine?" to questions which more accurately capture user experience such as "how many queries take more than 200 ms to return answers?" or "what is the worst case latency that a user may be subject to, and how often might it occur?" While various strategies exist for efficiently processing queries over large textual corpora, prior research has focused almost entirely on improvements to the average processing time or cost of search systems. As a first contribution, we examine some state-of-the-art retrieval algorithms for two popular index organizations, and discuss the trade-offs between them, paying special attention to the notion of tail latency. This research uncovers a number of observations that are subsequently leveraged for improved search efficiency and effectiveness. We then propose and solve a new problem, which involves processing a number of related query variations together, known as multi-queries , to yield higher quality search results. We experiment with a number of algorithmic approaches to efficiently process these multi-queries, and report on the cost, efficiency, and effectiveness trade-offs present with each. Finally, we examine how predictive models can be used to improve the tail latency and end-to-end cost of a commonly used multi-stage retrieval architecture without impacting result effectiveness. By combining ideas from numerous areas of information retrieval, we propose a prediction framework which can be used for training and evaluating several efficiency/effectiveness trade-off parameters, resulting in improved trade-offs between cost, result quality, and tail latency.

show abstract

Section: Discussionmentioning

confidence: 99%

Managing tail latency in large scale information retrieval systems

Mackenzie

2020

SIGIR Forum

View full text Add to dashboard Cite

show abstract

“…Altingovde et al [4] examine the efficiency of cluster-based search systems with term-at-a-time retrieval, including exploration of a cluster-based document identifier reassignment technique, which assigns identifiers to documents within each cluster consecutively, and orders the clusters according to their creation order. A similar idea was recently investigated by Siedlaczek et al [61], who reorder documents by hit count, using a query log to determine the documents that appear most regularly in the top-𝑘 lists. A resource allocator task estimates the clusters to search, and how deep to search within each cluster, by providing a global ordering cutoff.…”

Section: Topical Shards and Selective Searchmentioning

confidence: 99%

“…The approaches used in selective search often make use of a central shard index (CSI), a sampling of a small percentage of documents from each of the shards, employed as an indicator as to which shards might contain high-scoring documents and should thus be handed the query [35]. Alternatives include combining a small number of features stored on a per-term per-shard basis [5], or using learned models over a dozen or more per-term per-shard features [18,54,61]. In selective search, it is necessary for both a shard ordering and a numeric shard count to generated at the time the broker is routing each query [34,36,54].…”

Section: Concatenatementioning

confidence: 99%

Anytime Ranking on Document-Ordered Indexes

Mackenzie¹,

Petri²,

Moffat³

2021

Preprint

View full text Add to dashboard Cite

Inverted indexes continue to be a mainstay of text search engines, allowing efficient querying of large document collections. While there are a number of possible organizations, document-ordered indexes are the most common, since they are amenable to various query types, support index updates, and allow for efficient dynamic pruning operations. One disadvantage with document-ordered indexes is that high-scoring documents can be distributed across the document identifier space, meaning that index traversal algorithms that terminate early might put search effectiveness at risk. The alternative is impact-ordered indexes, which primarily support top-𝑘 disjunctions, but also allow for anytime query processing, where the search can be terminated at any time, with search quality improving as processing latency increases. Anytime query processing can be used to effectively reduce high-percentile tail latency which is essential for operational scenarios in which a service level agreement (SLA) imposes response time requirements. In this work, we show how document-ordered indexes can be organized such that they can be queried in an anytime fashion, enabling strict latency control with effective early termination. Our experiments show that processing document-ordered topical segments selected by a simple score estimator outperforms existing anytime algorithms, and allows query runtimes to be accurately limited in order to comply with SLA requirements.

show abstract

Faster Dynamic Pruning via Reordering of Documents in Inverted Indexes

Yafay

Altıngövde

2023

Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

Exploiting Global Impact Ordering for Higher Throughput in Selective Search

Cited by 3 publications

References 24 publications

Managing tail latency in large scale information retrieval systems

Managing tail latency in large scale information retrieval systems

Anytime Ranking on Document-Ordered Indexes

Faster Dynamic Pruning via Reordering of Documents in Inverted Indexes

Contact Info

Product

Resources

About