Visual Search Optimization

Prashanth Thinakaran; Diana Guttman; Mahmut Taylan Kandemir; Meenakshi Arunachalam; Rahul Khanna; Praveen Yedlapalli; Narayan Ranganathan

doi:10.1016/B978-0-12-803819-2.00021-5

Visual Search Optimization

Prashanth Thinakaran, Diana Guttman, Mahmut Taylan Kandemir, Meenakshi Arunachalam, Rahul Khanna, Praveen Yedlapalli, Narayan Ranganathan

Research output: Chapter in Book/Report/Conference proceeding › Chapter

1 Scopus citations

Abstract

This chapter presents an image-matching application that can take advantage of many-core architectures. Different parallelization strategies are explored that can take advantage of inter- and intraimage parallelism. The two main metrics that determine the application performance, tree creation time and search time, were studied in the context of scalability. Important insights obtained from a profiler-based analysis help identify the challenges in scalability of DB threads. The scalability with respect to increasing DBThreads with optimal KD-trees is shown to lead to 5.8× speedup in create time and 2.8× speedup in search time in the case of 120 threads when compared to single-threaded Xeon Phi performance.

Original language	English (US)
Title of host publication	High Performance Parallelism Pearls
Subtitle of host publication	Multicore and Many-core Programming Approaches
Publisher	Elsevier Inc.
Pages	191-209
Number of pages	19
Volume	2
ISBN (Electronic)	9780128038901
ISBN (Print)	9780128038192
DOIs	https://doi.org/10.1016/B978-0-12-803819-2.00021-5
State	Published - Jul 23 2015

All Science Journal Classification (ASJC) codes

General Computer Science

Access to Document

10.1016/B978-0-12-803819-2.00021-5

Cite this

@inbook{0ef954ef4a2445f7bf9569fc7d738539,

title = "Visual Search Optimization",

abstract = "This chapter presents an image-matching application that can take advantage of many-core architectures. Different parallelization strategies are explored that can take advantage of inter- and intraimage parallelism. The two main metrics that determine the application performance, tree creation time and search time, were studied in the context of scalability. Important insights obtained from a profiler-based analysis help identify the challenges in scalability of DB threads. The scalability with respect to increasing DBThreads with optimal KD-trees is shown to lead to 5.8× speedup in create time and 2.8× speedup in search time in the case of 120 threads when compared to single-threaded Xeon Phi performance.",

author = "Prashanth Thinakaran and Diana Guttman and Kandemir, \{Mahmut Taylan\} and Meenakshi Arunachalam and Rahul Khanna and Praveen Yedlapalli and Narayan Ranganathan",

year = "2015",

month = jul,

day = "23",

doi = "10.1016/B978-0-12-803819-2.00021-5",

language = "English (US)",

isbn = "9780128038192",

volume = "2",

pages = "191--209",

booktitle = "High Performance Parallelism Pearls",

publisher = "Elsevier Inc.",

address = "United States",

}

TY - CHAP

T1 - Visual Search Optimization

AU - Thinakaran, Prashanth

AU - Guttman, Diana

AU - Kandemir, Mahmut Taylan

AU - Arunachalam, Meenakshi

AU - Khanna, Rahul

AU - Yedlapalli, Praveen

AU - Ranganathan, Narayan

PY - 2015/7/23

Y1 - 2015/7/23

N2 - This chapter presents an image-matching application that can take advantage of many-core architectures. Different parallelization strategies are explored that can take advantage of inter- and intraimage parallelism. The two main metrics that determine the application performance, tree creation time and search time, were studied in the context of scalability. Important insights obtained from a profiler-based analysis help identify the challenges in scalability of DB threads. The scalability with respect to increasing DBThreads with optimal KD-trees is shown to lead to 5.8× speedup in create time and 2.8× speedup in search time in the case of 120 threads when compared to single-threaded Xeon Phi performance.

AB - This chapter presents an image-matching application that can take advantage of many-core architectures. Different parallelization strategies are explored that can take advantage of inter- and intraimage parallelism. The two main metrics that determine the application performance, tree creation time and search time, were studied in the context of scalability. Important insights obtained from a profiler-based analysis help identify the challenges in scalability of DB threads. The scalability with respect to increasing DBThreads with optimal KD-trees is shown to lead to 5.8× speedup in create time and 2.8× speedup in search time in the case of 120 threads when compared to single-threaded Xeon Phi performance.

UR - https://www.scopus.com/pages/publications/84982899042

UR - https://www.scopus.com/inward/citedby.url?scp=84982899042&partnerID=8YFLogxK

U2 - 10.1016/B978-0-12-803819-2.00021-5

DO - 10.1016/B978-0-12-803819-2.00021-5

M3 - Chapter

AN - SCOPUS:84982899042

SN - 9780128038192

VL - 2

SP - 191

EP - 209

BT - High Performance Parallelism Pearls

PB - Elsevier Inc.

ER -

Visual Search Optimization

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this