DOCSLIB.ORG

Multimedia Big Data Processing Using Hpcc Systems

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Multimedia Big Data Processing Using Hpcc Systems MULTIMEDIA BIG DATA PROCESSING USING HPCC SYSTEMS by Vishnu Chinta A Thesis Submitted to the Faculty of The College of Engineering & Computer Science In Partial Fulfillment of the Requirements for the Degree of Master of Science Florida Atlantic University Boca Raton, FL May 2017 Copyright by Vishnu Chinta 2017 ii ACKNOWLEDGEMENTS I would like to express gratitude to my advisor, Dr. Hari Kalva, for his constant guidance and support in helping me complete this thesis successfully. His enthusiasm, inspiration and his efforts to explain all aspects clearly and simply helped me throughout. My sincere thanks also go to my committee members Dr. Borko Furht and Dr. Xingquan Zhu for their valuable comments, suggestions and inputs to the thesis. I would also like to thank the LexisNexis for making this study possible by providing its funding, tools and all the technical support I’ve received. I would also like to thank all my professors during my time at FAU for all the invaluable lessons I’ve learnt in their classes. This work has been funded by the NSF Award No. 1464537, Industry/University Cooperative Research Center, Phase II under NSF 13-542. I would like to thank the NSF for this. Many thanks to my friends at FAU and my fellow lab mates in the Multimedia lab for the enthusiastic support and interesting times we have had during the past two years. Last but not the least I would like to thank my family for being a constant source of support and encouragement during my Masters. iv ABSTRACT Author: Vishnu Chinta Title: Multimedia Big Data Processing Using Hpcc Systems Institution: Florida Atlantic University Thesis Advisor: Dr. Hari Kalva Degree: Master of Science Year: 2017 There is now more data being created than ever before and this data can be any form of data, textual, multimedia, spatial etc. To process this data, several big data processing platforms have been developed including Hadoop, based on the MapReduce model and LexisNexis’ HPCC systems. In this thesis we evaluate the HPCC Systems framework with a special interest in multimedia data analysis and propose a framework for multimedia data processing. It is important to note that multimedia data encompasses a wide variety of data including but not limited to image data, video data, audio data and even textual data. While developing a unified framework for such wide variety of data, we have to consider computational complexity in dealing with the data. Preliminary results show that HPCC can potentially reduce the computational complexity significantly. v MULTIMEDIA BIG DATA PROCESSING USING HPCC SYSTEMS LIST OF FIGURES ......................................................................................................... viii INTRODUCTION .............................................................................................................. 1 1.1 Motivation ................................................................................................................. 1 1.2 Main Contributions ................................................................................................... 2 1.3 Overview of Thesis ................................................................................................... 3 1.4 Glossary of Terms ..................................................................................................... 3 CHAPTER 2 ....................................................................................................................... 4 BACKGROUND AND RELATED WORK ...................................................................... 4 2.1 Hadoop Ecosystem .................................................................................................... 4 2.2 HPCC Systems .......................................................................................................... 5 2.3 Similar Studies .......................................................................................................... 8 2.3.1 HIPI - Hadoop Image Processing Interface ....................................................... 8 2.3.2 CURT MapReduce: Caching and Utilizing Results of Tasks for MapReduce on Cloud Computing .......................................................................................... 9 2.3.3 Optimized Multiple Platforms for Big Data Analysis ..................................... 10 2.3.4 MapReduce Vocabulary Tree: An Approach for Large Scale Image Indexing and Search in the Cloud .................................................................................... 11 2.3.4 Extensible Video Processing Framework in Apache Hadoop ......................... 12 CHAPTER 3 ..................................................................................................................... 13 PROBLEM DESCRIPTION ............................................................................................. 13 vi 3.1 Multimedia Data Processing ................................................................................... 13 3.2 Multimedia Big Data in HPCC ............................................................................... 13 3.3 Proposed System Architecture ................................................................................ 14 CHAPTER 4 ..................................................................................................................... 16 PERFORMANCE EVALUATION .................................................................................. 16 4.1 Experimental Setup ................................................................................................. 16 4.2 Implementation ....................................................................................................... 16 4.3 Dataset ..................................................................................................................... 17 4.4 Dataset Processing .................................................................................................. 18 4.5 Decoding Image Data ............................................................................................. 21 4.6 Query Processing .................................................................................................... 21 4.6.1 Metadata Processing ........................................................................................ 22 4.6.2 Feature Extraction ............................................................................................ 28 4.7 Data Visualization ................................................................................................... 28 CHAPTER 5 ..................................................................................................................... 30 CONCLUSION AND FUTURE WORK ......................................................................... 30 5.1 Conclusion .............................................................................................................. 30 5.2 Other Applications .................................................................................................. 30 5.3 Future Work ............................................................................................................ 31 BIBLIOGRAPHY ............................................................................................................. 32 LIST OF FIGURES Figure 2.1 Hadoop Distributed File System Architecture .................................................. 5 Figure 2.2 HPCC THOR Cluster ........................................................................................ 8 Figure 2.3 How HIPI parallelizes image processing .......................................................... 9 Figure 3.14Proposed System Architecture ....................................................................... 14 Figure 4.15Sample images from dataset ........................................................................... 18 Figure 4.26Spray Timings ................................................................................................ 19 Figure 4.37Execution graph generated in ECL Watch ..................................................... 24 Figure 4.48Impact of dataset size on query processing .................................................... 26 Figure 4.59Execution times for finding images with occurrences of 'Cat' in the United States ......................................................................................................................... 27 Figure 4.6 Sample Images returned from query ………………………………………...28 Figure 4.7 Sample Visualization ………………………………………………………...29 LIST OF TABLES Table 4.1 Spray Timings ................................................................................................... 19 Table 4.2 Sample of Query Results .................................................................................. 23 Table 4.3 Breakdown of query execution times for 100m records ................................... 24 Table 4.4 Execution times for different input record sizes ............................................... 25 Table 4.5 Execution times for query finding images with occurrences of 'Cat' in United States ................................................................................................................ 26 ix CHAPTER 1 INTRODUCTION Big data analytics is the use of advanced analytic techniques against very large, diverse data sets that include different types such as structured/unstructured and streaming/batch, and different sizes from terabytes to petabytes. Big data is a term applied to datasets whose
Load more

Recommended publications
  • Deliver Performance and Scalability with Hitachi Vantara's Pentaho
    Deliver Performance and Scalability With Hitachi Vantara’s Pentaho Business Analytics Platform By Hitachi Vantara November 2018 Contents Executive Summary ........................................................................................... 2 Meet Enterprise Scalability and High-Performance Requirements With Pentaho Business Analytics Platform ............................................................................... 3 Pentaho Business Analytics Server................................................................................................................. 3 Deployment on 64-Bit Operating Systems ........................................................................................................ 4 Optimize Configuration of the Reporting and Analysis Engines .............................. 5 Pentaho Reporting .............................................................................................................................................. 5 Pentaho Analysis ................................................................................................................................................. 5 Pentaho Data Integration ..................................................................................... 7 1 Executive Summary Business analytics solutions are only valuable when they can be accessed and used by anyone, from anywhere and at any time. When selecting a business analytics platform, it is critical to assess the underlying architecture of the platform. This consideration ensures that it not
    [Show full text]
  • The HPCC Cluster Computing Paradigm and an Efficient Data-Centric Programming Language Are Key Factors in Our Company's Success
    Welcome HPCC Systems Architecture Overview & Pentaho Spoon/Kettle Integration Monday, March 12, 2012 7:00 pm, EST Logistics Twitter hashtags: #cfbi & #hpccmeetup Agenda 1. High level architecture of the HPCC platform: 20 minutes 2. Pentaho plugin integration demo: 20 minutes 3. Q&A (raise your hand) 20 minutes Presenters Flavio Villanustre & Arjuna Chala http://hpccsystems.com Risk Solutions 1 What is HPCC Systems? HPCC is a massive parallel-processing computing platform E S P Risk Solutions 2 The HPCC cluster computing paradigm and an efficient data-centric programming language are key factors in our company’s success “Grid” Computing Data-centric language (ECL) Integrated Delivery System Splits problems into pieces to be “Big Data” language brings the Consistent approach across data worked in parallel by commodity computing to the data ingestion, processing, and servers delivery systems + = Risk Solutions 3 The Three main HPCC components . Massively Parallel Extract Transform and Load (ETL) engine 1 HPCC Data Refinery (Thor) – Built from the ground up as a parallel data environment. Leverages inexpensive locally attached storage. Doesn’t require a SAN infrastructure. • Enables data integration on a scale not previously available: – Current LexisNexis person data build process generates 350 Billion intermediate results at peak • Suitable for: – Massive joins/merges – Massive sorts & transformations – Programmable using ECL . A massively parallel, high throughput, structured query response engine 2 HPCC Data Delivery Engine (Roxie) . Ultra fast low latency and highly available due to its read-only nature. Allows indices to be built onto data for efficient multi-user retrieval of data . Suitable for Volumes of structured queries Full text ranked Boolean search .
    [Show full text]
  • HPCC Systems Open Source, Big Data Processing and Analytics See
    See Through Patterns, Hidden Relationships and Networks to Find Opportunities in Big Data. HPCC Systems Open Source, Big Data Processing and Analytics LexisNexis LexisNexis leverages data about people, businesses, and assets to assess risk and opportunity associated with industry-specific problems Contributory Public Credit Phone Assess underwriting risk and verify applicant data; prevent/ Data Records Headers Listings Insurance investigate fraudulent claims; optimize policy administration Financial Services Prevent/investigate money laundering and comply with laws Prevent/investigate identity fraud HPCC Systems Open Source and commercial offerings of our leading Big Data technology platform Receivables Assist collections by locating delinquent debtors Management Assess collectability of debts and prioritize collection efforts Locate/vet witnesses, assess assets/associations of parties in Legal legal actions; Perform diligence on prospective clients (KYC) Locate missing children/suspects; research/document cases; Government reduce entitlement fraud; accelerate revenue collection Health Care Verify patient identity, eligibility, and ability to pay Validate provider credentials; prevent/investigate fraud 3 LexisNexis Risk Solutions works with Fortune 1000 and mid-market clients across all industries and both federal and state governments. • Customers in over 139 countries • 6 of the world’s top 10 banks • 100% of the top 50 U.S. banks • 80% of the Fortune 500 companies • 100% of U.S. P&C insurance carriers • All 50 U.S. states • 70% of local
    [Show full text]
  • Big Data Analytics Tool Hive
    Science, Technology and Development ISSN : 0950-0707 Big Data Analytics Tool Hive Shikhar Guglia 1, Anu Sharma 2 M.Tech Student, Assistant Professor Teerthanker Mahaveer University,Moradabad Abstract : We are surrounded by data. people upload videos, take pictures,text each other, reading mail, sending mails , leaves comment around the web , updates their LinkedIn , Facebook and Whatsapp status , click on ads and so forth. Machines too are generating and keeping more and more data . Even you can take reference a whole book Digitally or virtually and purchasing a book also can be recorded as data. The exponential growth of data has been biggest challenge to the tech giants like Google, Amazon , Micro-soft and Yahoo. Analyzingand reduce the manual work of getting the required results from a huge dataset of around gigabytes and terabytes which is quite tedious if performed on Relational Database Management Systems as it unable to tackle data. Key-Words : GigaBytes,TeraBytes , Hive Querylanguage , Relational Database , Dataset , Data Warehousing ,Hadoop ,Map-Reduce. 1. Introduction : ApacheHive is a data warehousing Tool that is built on the top of Hadoop to summarizing Big Data for Providing analysis and data query using Structured as well as Semi-Structured Data. It Uses HQL (Hive Query language) similar to SQL (Structured Query language) to process data using HQL Queries. It is familiar, fast, scalable and extensible . 1.1 Prerequisites: Before proceeding with Hive, we need a basic level knowledge of SQL , core Java, Hadoop File System and any Linux Operating System. Hadoop : Hadoop is an open-Source framework to store and process large amount of data in a distributed environment.
    [Show full text]
  • HPCC Benchmarking
    Final Project Report HPCC Benchmarking MAY 29, 2018 RUSHIKESH GHATPANDE SUPERVISOR :VINCENT WFREEH COMPUTER SCIENCE DEPARTMENT NC STATE UNIVERSITY ABSTRACT igh-performance Computing Cluster (HPCC) Thor is a big data analytics engine. It is designed to execute big data workflows including extraction, loading, cleansing, trans- Hformations, linking, and indexing. A Thor cluster is similar in its function, execution environment, filesystem, and capabilities to the Google and Hadoop MapReduce platforms. It can be optimized for its parallel data processing purpose. HPCC’s declarative and data flow-oriented language - ECL defines the processing result desired; the specific processing steps required to perform the processing are left to the language compiler. These salient features make it a viable alternative to established big data analytics engines like Spark and Hadoop. In this report, we have benchmarked HPCC alongside Hadoop. We present a detailed study of scalability, time for execution, and system-level metrics like CPU utilization, memory utilization and disk I/O patterns. We try to identify general trends in performance and which engine is better suited for a particular workload. i TABLE OF CONTENTS Page List of Tables iv List of Figures v 1 Introduction 1 2 Requirements 3 2.1 Functional Requirements .................................. 3 2.2 Non-Functional Requirements ............................... 3 3 System Environment5 3.1 Hardware Specifications................................... 5 3.2 Software Specifications.................................... 5 4 Design, Implementation and Results7 4.1 Design.............................................. 7 4.2 Implementation ........................................ 8 4.3 Results ............................................. 9 5 Observations and Challenges 13 5.1 Observation........................................... 13 5.1.1 CPU Utilization.................................... 13 5.1.2 Code Optimization.................................. 15 5.2 Challenges ..........................................
    [Show full text]
  • Beyond Batch Processing: Towards Real-Time and Streaming Big Data
    Beyond Batch Processing: Towards Real-Time and Streaming Big Data Saeed Shahrivari, and Saeed Jalili Computer Engineering Department, Tarbiat Modares University (TMU), Tehran, Iran [email protected], [email protected] Abstract: Today, big data is generated from many sources and there is a huge demand for storing, managing, processing, and querying on big data. The MapReduce model and its counterpart open source implementation Hadoop, has proven itself as the de facto solution to big data processing. Hadoop is inherently designed for batch and high throughput processing jobs. Although Hadoop is very suitable for batch jobs but there is an increasing demand for non-batch processes on big data like: interactive jobs, real-time queries, and big data streams. Since Hadoop is not proper for these non-batch workloads, new solutions are proposed to these new challenges. In this article, we discuss two categories of these solutions: real-time processing, and stream processing for big data. For each category, we discuss paradigms, strengths and differences to Hadoop. We also introduce some practical systems and frameworks for each category. Finally, some simple experiments are done to show effectiveness of some solutions compared to available Hadoop-based solutions. Keywords: Big data, MapReduce, Real-time processing, Stream processing 1. Introduction The “Big Data” paradigm is getting an expanding popularity recently. The “Big Data” term is generally used for datasets which are so huge that cannot be processed and managed using classical solutions like Relational Data Base Systems (RDBMS). Besides volume, large velocity and variety are other challenges of big data. Numerous sources generate big data.
    [Show full text]
  • HPCC System and Its Future Aspects in Maintaining Big Data
    Special Conference Issue: National Conference on Cloud Computing & Big Data 191 HPCC System and its Future Aspects in Maintaining Big Data Tarun Goya Dept. of Computer Science, Govt. Engineering College Bikaner, Bikaner, Rajasthan, India [email protected] Ved Prakash Upadhyay Dept. of Information Technology Govt. Engineering College Bikaner, Bikaner, Rajasthan, India [email protected] Bhartendu Shrimali Dept. of Information Technology Govt. Engineering College Bikaner, Bikaner, Rajasthan, India [email protected] -------------------------------------------------------------------ABSTRACT----------------------------------------------------------------- This research paper will make a complete overview of HPCC system and its future aspects in the field of maintaining big data. HPCC (High Performance Computing Cluster) system is developed and designed by LexisNexis. The main benefit of HPCC system is it have its own language ECL which make it faster and better than its competitors. Definitely HPCC will be a future player in the field of big data. As big data is the data in huge amount and maintenance such huge data is the main problem of the today’s world. Keywords—HPCC, big data, clusters, cloud computing, Hadoop, ECL and ROXIE. ----------------------------------------------------------------------------------------------------------------------------- ----------------------- hardware. But now we have HPCC (High- I. INTRODUCTION Performance Computing Cluster), which is also known as DAS (Data Analytics Supercomputer) [1], oday we live in the era of data, where we can find T is an open source, data-intensive computing system all the information in the form of data. As a result of platform developed by lexis nexis. information explosion, many organization have large amount of data, which they have collected from The HPCC system incorporates a software different sources and are stored in massive datasets.
    [Show full text]
  • Cloud Based Big Data Infrastructure: Architectural Components and Automated Provisioning
    Cloud Based Big Data Infrastructure: Architectural Components and Automated Provisioning The 2016 International Conference on High Performance Computing & Simulation (HPCS2016) 18-22 July 2016, Innsbruck, Austria Yuri Demchenko, University of Amsterdam HPCS2016 Cloud based Big Data Infrastructure 1 Outline • Big Data and new concepts – Big Data definition and Big Data Architecture Framework (BDAF) – Data driven vs data intensive vs data centric and Data Science • Cloud Computing as a platform of choice for Big Data applications – Big Data Stack and Cloud platforms for Big Data – Big Data Infrastructure provisioning automation • CYCLONE project and use cases for cloud based scientific applications automation • Slipstream and cloud automation tools – Slipstream recipe example • Discussion The research leading to these results has received funding from the Horizon2020 project CYCLONE HPCS2016 Cloud based Big Data Infrastructure 2 Big Data definition revisited: 6 V’s of Big Data Volume • Terabytes Generic Big Data • Records/Arch Properties Variety • Tables, Files Velocity • Distributed • Structured • Volume • Unstructured • Batch • Multi-factor • Real/near-time • Variety • Probabilistic • Processes • Linked • Streams • Velocity • Dynamic 6 Vs of Big Data Acquired Properties (after entering system) • Changing data • CorrelationsValue • Changing model • Statistical • Value • Linkage • Events • Hypothetical • Veracity Variability • Trustworthiness • Authenticity • Variability • Origin, Reputation • Availability • Accountability Commonly accepted
    [Show full text]
  • A Guide in the Big Data Jungle Thesis, Bachelor of Science
    A guide in the Big Data jungle Thesis, Bachelor of Science Anna Ohlsson, Dan Öman Faculty of Computing Blekinge Institute of Technology SE-371 79 Karlskrona Sweden Contact Information: Author(s): Anna Ohlsson, E-mail: [email protected], Dan Öman E-mail: [email protected] University advisor: Nina Dzamashvili Fogelström Department of Software Engineering Faculty of Computing Internet : www.bth.se/com Blekinge Institute of Technology Phone : +46 0455 38 50 00 SE-371 79 Karlskrona Sweden Fax : +46 0455 38 50 57 1 Abstract This bachelor thesis looks at the functionality of different frameworks for data analysis at large scale and the purpose of it is to serve as a guide among available tools. The amount of data that is generated every day keep growing and for companies to take advantage of the data they collect they need to know how to analyze it to gain maximal use out of it. The choice of platform for this analysis plays an important role and you need to look in to the functionality of the different alternatives that are available. We have created a guide to make this research easier and less time consuming. To evaluate our work we created a summary and a survey which we asked a number of IT-students, current and previous, to take part in. After analyzing their answers we could see that most of them find our thesis interesting and informative. 2 Content Introduction 5 1.1 Overview 5 Background and related work 7 2.1 Background 7 Figure 1 Overview of Platforms and Frameworks 8 2.1.1 What is Big Data? 8 2.1.2 Platforms and frameworks 9
    [Show full text]
  • The ECL Programming Paradigm
    Welcome http://hpccsystems.com BIG Data & Analytics Solving Big Data Problems with the Open Source HPCC Systems Platform John Holt, PhD, Senior Architect - LexisNexis Agenda 7:20-7:45pm: Presentation & Demo 7:45-7:55pm: Q&A / Open discussion 7:55-8:00pm: Kindle Fire give-away Twitter event hashtag: #hpccmeetup hpccsystems.com Risk Solutions 1 What is HPCC Systems? http://hpccsystems.com Discovery Collection Ingestion & Integration Analysis Delivery Cleansing E S P Risk Solutions 2 http://hpccsystems.com The HPCC cluster computing paradigm and an efficient data-centric programming language are key factors in our company’s success “Grid” Computing Data-centric language (ECL) Integrated Delivery System Splits problems into pieces to be “Big Data” language brings the Consistent approach across data worked in parallel by commodity computing to the data ingestion, processing, and servers delivery systems + = Risk Solutions 3 http://hpccsystems.com The ECL programming paradigm Paraphrasing Paul Graham (from the introduction to “On Lisp”): In ECL, you don't just write your program down toward the language, you also build the language up toward your program. As you're writing a program you may think "I wish ECL had such-and-such an operator”. So you go and write it. Afterward you realize that using the new operator would simplify the design of another part of the program, and so on. Language and program evolve together. Like the border between two warring states, the boundary between language and program is drawn and redrawn, until eventually it comes to rest along the mountains and rivers, the natural frontiers of your problem.
    [Show full text]
  • HPCC Systems: Introduction to HPCC (High-Performance Computing Cluster)
    White Paper HPCC Systems: Introduction to HPCC (High-Performance Computing Cluster) Author: Anthony M. Middleton, Ph.D. LexisNexis Risk Solutions Date: May 24, 2011 Executive Summary As a result of the continuing information explosion, many organizations now have the need to process and analyze massive volumes of data. These data-intensive computing requirements can be addressed by scalable systems based on hardware clusters of commodity servers coupled with system software to provide a distributed file storage system, job execution environment, online query capability, parallel application processing, and parallel programming development tools. The LexisNexis HPCC platform provides all of these capabilities in an integrated, easy to implement and use, commercially-available high-performance computing environment. This paper provides an introduction to the LexisNexis HPCC system architecture also referred to (in the government space) as the LexisNexis Data Analytics Supercomputer (DAS). LexisNexis Risk Solutions, an industry leader in data content, data aggregation, and information services, independently developed and implemented the HPCC platform as a solution for its own data-intensive computing requirements. In a similar manner to Hadoop (the open source implementation of MapReduce), the LexisNexis approach also uses commodity clusters of hardware running the Linux operating system and includes additional system software and middleware components to provide a complete and comprehensive job execution environment and distributed query and filesystem support needed for data-intensive computing. The HPCC platform includes a powerful high-level, heavily-optimized, data-centric declarative language for parallel data processing called ECL (Enterprise Data Control Language) which is also described in this paper. The power, flexibility, advanced capabilities, speed of development, maturity, and ease of use of the ECL programming language is a primary distinguishing factor between the LexisNexis HPCC platform and other data-intensive computing solutions.
    [Show full text]
  • The Pigmix Benchmark on Pig, Mapreduce, and HPCC Systems
    2015 IEEE International Congress on Big Data The PigMix Benchmark on Pig, MapReduce, and HPCC Systems Keren Ouaknine1, Michael Carey2, Scott Kirkpatrick1 {ouaknine}@cs.huji.ac.il 1School of Engineering and Computer Science, The Hebrew University of Jerusalem 2Donald Bren School of Information and Computer Sciences, University of California, Irvine Abstract—Soon after Google published MapReduce, QUERY JAVA PIG LATIN ECL their paradigm for processing large amounts of data, the LANGUAGE open-source world followed with the Hadoop ecosystem. Later on, LexisNexis, the company behind the world’s ECL CC COMPILER PIG largest database of legal documents, open-sourced its Big SERVER Data processing platform, called the High-Performance HAND-WRITTEN GENERATED JAVA GENERATED Computing Cluster (HPCC). This paper makes three MAPREDUCE JOB MAPREDUCE JOB C++ JOB contributions. First, we describe our additions and im- provements to the PigMix benchmark, the set of queries BIG DATA HADOOP HPCC originally written for Apache Pig, and the porting of PLATFORM PigMix to HPCC. Second, we compare the performance of queries written in Pig, Java MapReduce, and ECL. Last, we draw conclusions and issue recommendations for Fig. 1. Comparison of handwritten Java MapReduce (left), Pig scripts future system benchmarks and large-scale data-processing (middle), and ECL scripts compiled into C++ jobs for HPCC (right). platforms. Index Terms—PigMix, HPCC Systems, MapReduce, Hadoop, and ECLCCServer for HPCC. But which plat- Benchmark, Big Data, Performance form performs best? For which tasks? Can they be com- pared rigorously? PigMix [2] is a benchmark released I. INTRODUCTION by Yahoo! and Hortonworks, the main contributors to Pig.
    [Show full text]