2011
DOI: 10.1109/mmul.2010.70
|View full text |Cite
|
Sign up to set email alerts
|

RanKloud: Scalable Multimedia Data Processing in Server Clusters

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
18
0

Year Published

2011
2011
2020
2020

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 28 publications
(18 citation statements)
references
References 12 publications
0
18
0
Order By: Relevance
“…The algorithm proposed for k-NN joins [77] uses z-ordering to map data to one-dimensional values and partition the one-dimensional space into ranges. In the case of top-k joins, both RanKloud [24] and the approach described in [38] propose partitioning schemes tailored to topk processing. RanKloud employs a utility-aware partitioning that splits the space in non-equal ranges, such that the useful work produced by each partition is roughly the same.…”
Section: Discussionmentioning
confidence: 99%
See 3 more Smart Citations
“…The algorithm proposed for k-NN joins [77] uses z-ordering to map data to one-dimensional values and partition the one-dimensional space into ranges. In the case of top-k joins, both RanKloud [24] and the approach described in [38] propose partitioning schemes tailored to topk processing. RanKloud employs a utility-aware partitioning that splits the space in non-equal ranges, such that the useful work produced by each partition is roughly the same.…”
Section: Discussionmentioning
confidence: 99%
“…RanKloud [24] has been proposed for top-k retrieval in the cloud. RanKloud computes statistics (at runtime) during scanning of records and uses these statistics to compute a threshold (the lowest score for top-k results) for early termination.…”
Section: Early Terminationmentioning
confidence: 99%
See 2 more Smart Citations
“…All these approaches focus on to improve future query executions using feedback, but not addressed improving query performance in mid execution which is addressed in the system proposed by Bruno N et al, [4] in a distributed environment. Rankloud computes run time statistics that is used to derive the lowest score for top-k results [48]. Starfish system consists of two components, Profiler and What-if Engine.…”
Section: Continuous or Iterative Processingmentioning
confidence: 99%