Maguro, a system for indexing and searching over very large text collections

Risvik, Knut Magne; Chilimbi, Trishul; Tan, Hua; Kalyanaraman, Karthik; Anderson, Chris

doi:10.1145/2433396.2433486

Cited by 16 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, various document fields can be used to rank documents more effectively. To support advanced rankers across multiple fields, each field may be represented by field-based postings lists [210]. Each field is typically given a weight, and the document score can be computed as a weighted sum across the given fields.…”

Section: Improving Effectivenessmentioning

confidence: 99%

“…Most large-scale IR systems are backed by thousands of commodity servers which allows the processing workload to be distributed among many peers in a fault-tolerant manner, as evidenced by some recent outlines of large-scale search architectures from companies such as Google, Microsoft, eBay, and Twitter [20,39,210,245]. To achieve rapid response times, large document corpora are divided into a set of smaller partitions, which is known as index sharding.…”

Section: Search Engine Infrastructurementioning

confidence: 99%

“…This study concluded that users tend to becomes less engaged with the search engine if delays are consistently observed, and that revenue is higher when pages are served faster, thus confirming the previous observations from other studies [16] -balancing both efficiency and effectiveness is key in providing a positive user experience. [210] which powers the Bing search engine. At the top level, the corpus broker manages the load balancing of query processing across various partitions of the overall index.…”

Section: Efficiency Effectiveness and Revenuementioning

confidence: 99%

“…During this presentation, Pederson outlined the matching and ranking pipeline used within Bing's production search system, explaining that Bing used "a sequence of select, rank, and prune steps" during retrieval. 1 A deeper overview of this architecture was presented in a later paper from Risvik et al [210], who outline Maguro, a core retrieval system used within the Bing search engine. Wang et al [252] provided the first formal overview of the cascade ranking model (otherwise known as multi-stage retrieval), and showed that cascade rankers can achieve a similar effectiveness to monolithic rankers at a fraction of the cost.…”

Section: Candidate Generationmentioning

confidence: 99%

See 3 more Smart Citations

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: Improving Effectivenessmentioning

confidence: 99%

Section: Search Engine Infrastructurementioning

confidence: 99%

Section: Efficiency Effectiveness and Revenuementioning

confidence: 99%

Section: Candidate Generationmentioning

confidence: 99%

See 2 more Smart Citations

Managing tail latency in large scale information retrieval systems

Mackenzie

2020

SIGIR Forum

View full text Add to dashboard Cite

show abstract

“…In fact, the matching phase in a commercial search engine [15] returns a superset of the results and uses the ranking phase for further filtering. The ranking phase follows the matching phase [25], is typically performed in successive steps for pruning, and is extensively accelerated with custom hardware such as FPGAs [24] and ASICs [18] in commercial engines. Thus custom hardware reduces the ranking time, and subsequently introduces changes to the matching phase.…”

Section: Related Workmentioning

confidence: 99%

List intersection for web search

Kim

Lee²,

Hwang

et al. 2018

Proc. VLDB Endow.

View full text Add to dashboard Cite

This paper studies the optimization of list intersection, especially in the context of the matching phase of search engines. Given a user query, we intersect the postings lists corresponding to the query keywords to generate the list of documents matching all keywords. Since the speed of list intersection depends the algorithm, hardware, and list lengths and their correlations, none the existing intersection algorithms outperforms the others in every scenario. Therefore, we develop a cost-based approach in which we identify a search space, spanning existing algorithms and their combinations. We propose a cost model to estimate the cost of the algorithms with their combinations, and use the cost model to search for the lowest-cost algorithm. The resulting plan is usually a combination of 2-way algorithms, outperforming conventional 2-way and k-way algorithms. The proposed approach is more general than designing a specific algorithm, as the cost models can be adapted to different hardware. We validate the cost model experimentally on two different CPUs, and show that the cost model closely estimates the actual cost. Using both real and synthetic datasets, we show that the proposed cost-based optimizer outperforms the state-of-the-art alternatives.

show abstract