Fine-grained Partitioning for Aggressive Data Skipping Liwen Sun, Michael J. Franklin, Sanjay Krishnan, Reynold S. Xin† UC Berkeley and †Databricks Inc.

Slides:



Advertisements
Similar presentations
Copyright © 2007 Ramez Elmasri and Shamkant B. Navathe Slide
Advertisements

Module 13: Performance Tuning. Overview Performance tuning methodologies Instance level Database level Application level Overview of tools and techniques.
The HV-tree: a Memory Hierarchy Aware Version Index Rui Zhang University of Melbourne Martin Stradling University of Melbourne.
Efficient Event-based Resource Discovery Wei Yan*, Songlin Hu*, Vinod Muthusamy +, Hans-Arno Jacobsen +, Li Zha* * Chinese Academy of Sciences, Beijing.
Overcoming Limitations of Sampling for Agrregation Queries Surajit ChaudhuriMicrosoft Research Gautam DasMicrosoft Research Mayur DatarStanford University.
LIBRA: Lightweight Data Skew Mitigation in MapReduce
SkewTune: Mitigating Skew in MapReduce Applications
A Scalable, Predictable Join Operator for Highly Concurrent Data Warehouses George Candea (EPFL & Aster Data) Neoklis Polyzotis (UC Santa Cruz) Radek Vingralek.
Spark Performance Patrick Wendell Databricks.
UC Berkeley Spark Cluster Computing with Working Sets Matei Zaharia, Mosharaf Chowdhury, Michael Franklin, Scott Shenker, Ion Stoica.
Spark: Cluster Computing with Working Sets
1 Large-Scale Machine Learning at Twitter Jimmy Lin and Alek Kolcz Twitter, Inc. Presented by: Yishuang Geng and Kexin Liu.
Shark Cliff Engle, Antonio Lupher, Reynold Xin, Matei Zaharia, Michael Franklin, Ion Stoica, Scott Shenker Hive on Spark.
Low-Cost Data Deduplication for Virtual Machine Backup in Cloud Storage Wei Zhang, Tao Yang, Gautham Narayanasamy University of California at Santa Barbara.
1 HYRISE – A Main Memory Hybrid Storage Engine By: Martin Grund, Jens Krüger, Hasso Plattner, Alexander Zeier, Philippe Cudre-Mauroux, Samuel Madden, VLDB.
Mesos A Platform for Fine-Grained Resource Sharing in Data Centers Benjamin Hindman, Andy Konwinski, Matei Zaharia, Ali Ghodsi, Anthony D. Joseph, Randy.
FAWN: A Fast Array of Wimpy Nodes Presented by: Aditi Bose & Hyma Chilukuri.
Presented by Cathrin Weiss, Panagiotis Karras, Abraham Bernstein Department of Informatics, University of Zurich Summarized by: Arpit Gagneja.
1 04/18/2005 Flux Flux: An Adaptive Partitioning Operator for Continuous Query Systems M.A. Shah, J.M. Hellerstein, S. Chandrasekaran, M.J. Franklin UC.
CPS216: Advanced Database Systems (Data-intensive Computing Systems) How MapReduce Works (in Hadoop) Shivnath Babu.
MapReduce Simplified Data Processing On large Clusters Jeffery Dean and Sanjay Ghemawat.
Outline | Motivation| Design | Results| Status| Future
Introduction to Column-Oriented Databases Seminar: Columnar Databases, Nov 2012, Univ. Helsinki.
Efficient Parallel Set-Similarity Joins Using Hadoop Chen Li Joint work with Michael Carey and Rares Vernica.
Design Patterns for Efficient Graph Algorithms in MapReduce Jimmy Lin and Michael Schatz University of Maryland MLG, January, 2014 Jaehwan Lee.
Virtualization and Cloud Computing Research at Vasabilab Kasidit Chanchio Vasabilab Dept of Computer Science, Faculty of Science and Technology, Thammasat.
RuleML-2007, Orlando, Florida1 Towards Knowledge Extraction from Weblogs and Rule-based Semantic Querying Xi Bai, Jigui Sun, Haiyan Che, Jin.
A Comparison of Join Algorithms for Log Processing in MapReduce Spyros Blanas, Jignesh M. Patel(University of Wisconsin-Madison) Eugene J. Shekita, Yuanyuan.
Task Scheduling for Highly Concurrent Analytical and Transactional Main-Memory Workloads Iraklis Psaroudakis (EPFL), Tobias Scheuer (SAP AG), Norman May.
Hexastore: Sextuple Indexing for Semantic Web Data Management
Zois Vasileios Α. Μ :4183 University of Patras Department of Computer Engineering & Informatics Diploma Thesis.
Software Engineering for Business Information Systems (sebis) Department of Informatics Technische Universität München, Germany wwwmatthes.in.tum.de Data-Parallel.
Storage in Big Data Systems
Parallelizing Security Checks on Commodity Hardware E.B. Nightingale, D. Peek, P.M. Chen and J. Flinn U Michigan.
Performance Issues in Parallelizing Data-Intensive applications on a Multi-core Cluster Vignesh Ravi and Gagan Agrawal
Introduction to Hadoop and HDFS
VLDB2012 Hoang Tam Vo #1, Sheng Wang #2, Divyakant Agrawal †3, Gang Chen §4, Beng Chin Ooi #5 #National University of Singapore, †University of California,
« Performance of Compressed Inverted List Caching in Search Engines » Proceedings of the International World Wide Web Conference Commitee, Beijing 2008)
CCGrid 2014 Improving I/O Throughput of Scientific Applications using Transparent Parallel Compression Tekin Bicer, Jian Yin and Gagan Agrawal Ohio State.
Map-Reduce-Merge: Simplified Relational Data Processing on Large Clusters Hung-chih Yang(Yahoo!), Ali Dasdan(Yahoo!), Ruey-Lung Hsiao(UCLA), D. Stott Parker(UCLA)
Performance Prediction for Random Write Reductions: A Case Study in Modelling Shared Memory Programs Ruoming Jin Gagan Agrawal Department of Computer and.
Indexing HDFS Data in PDW: Splitting the data from the index VLDB2014 WSIC、Microsoft Calvin
Fine-grained Partitioning for Aggressive Data Skipping Calvin SIGMOD 2014 UC Berkeley.
MC 2 : Map Concurrency Characterization for MapReduce on the Cloud Mohammad Hammoud and Majd Sakr 1.
Computer Science and Engineering Parallelizing Defect Detection and Categorization Using FREERIDE Leonid Glimcher P. 1 ipdps’05 Scaling and Parallelizing.
CCGrid 2014 Improving I/O Throughput of Scientific Applications using Transparent Parallel Compression Tekin Bicer, Jian Yin and Gagan Agrawal Ohio State.
SAGA: Array Storage as a DB with Support for Structural Aggregations SSDBM 2014 June 30 th, Aalborg, Denmark 1 Yi Wang, Arnab Nandi, Gagan Agrawal The.
Resilient Distributed Datasets: A Fault- Tolerant Abstraction for In-Memory Cluster Computing Matei Zaharia, Mosharaf Chowdhury, Tathagata Das, Ankur Dave,
HADOOP DISTRIBUTED FILE SYSTEM HDFS Reliability Based on “The Hadoop Distributed File System” K. Shvachko et al., MSST 2010 Michael Tsitrin 26/05/13.
Matei Zaharia, in collaboration with Mosharaf Chowdhury, Tathagata Das, Ankur Dave, Haoyuan Li, Justin Ma, Murphy McCauley, Joshua Rosen, Reynold Xin,
Page 1 A Platform for Scalable One-pass Analytics using MapReduce Boduo Li, E. Mazur, Y. Diao, A. McGregor, P. Shenoy SIGMOD 2011 IDS Fall Seminar 2011.
MapReduce: Simplified Data Processing on Large Clusters By Dinesh Dharme.
Rethinking Choices for Multi-dimensional Point Indexing You Jung Kim and Jignesh M. Patel University of Michigan.
Bigtable: A Distributed Storage System for Structured Data Google Inc. OSDI 2006.
Succinct: Enabling Queries on Compressed Data Presented by: Bhargav Mangipudi Rachit Agarwal, Anurag Khandelwal, and Ion Stoica, University of California,
Configuring SQL Server for a successful SharePoint Server Deployment Haaron Gonzalez Solution Architect & Consultant Microsoft MVP SharePoint Server
Only Aggressive Elephants are Fast Elephants Nov 11 th 2013 Database Lab. Wonseok Choi.
PACMan: Coordinated Memory Caching for Parallel Jobs Ganesh Ananthanarayanan, Ali Ghodsi, Andrew Wang, Dhruba Borthakur, Srikanth Kandula, Scott Shenker,
Presented by: Omar Alqahtani Fall 2016
Practical Database Design and Tuning
Seth Pugsley, Jeffrey Jestes,
Distributed Network Traffic Feature Extraction for a Real-time IDS
Parallel Databases.
Succinct: Enabling Queries on Compressed Data
Selectivity Estimation of Big Spatial Data
Sameh Shohdy, Yu Su, and Gagan Agrawal
KISS-Tree: Smart Latch-Free In-Memory Indexing on Modern Architectures
Practical Database Design and Tuning
Declarative Transfer Learning from Deep CNNs at Scale
Fast, Interactive, Language-Integrated Cluster Computing
Presentation transcript:

Fine-grained Partitioning for Aggressive Data Skipping Liwen Sun, Michael J. Franklin, Sanjay Krishnan, Reynold S. Xin† UC Berkeley and †Databricks Inc. VLDB 2014 March 17, 2015 Heymo Kou

 Introduction  Overview  Workload Analysis  The Partitioning Problem  Feature-based Data Skipping  Discussion  Experimental Evaluation  Related Work & Conclusion Contents 2 / 18

 Several ways tom improve data scan throughput ‒ Memory caching ‒ Parallelization ‒ Data compression ‒ Reduce the data access (Data skipping)  Increasing interest in reducing data access Introduction 3 / 18

Recall Google’s PowerDrill 4 / 18

 Traditionally, ranged partitioning  PowerDrill ‒ Composite range partitioning Logic difference Skew Inevitable 5 / 18

 Feature Selection ‒ Analyze frequent query features  Optimal Partitioning ‒ Formulate Balanced MaxSkip partitioning problem  Scalability Contributions 6 / 18

 Filter Commonality ‒ Only small set of filters are commonly used  Filter Stability ‒ Future queries have occurred before Overview Workload Assumptions 7 / 18

 Workload Analyzer ‒ Extract features  Featurization ‒ Evaluate filters ‒ tuple  (vector, tuple)  Reduction ‒ Group by (vector, tuple)  Partitioner ‒ Split data  Shuffle ‒ Augment partitioned data  Catalog Update ‒ Union vectors for each block Overview Blocking Workflow 8 / 18

 Goal : extract freatures from the query traces ‒ Given ‒ Predicate Augmentation ‒ Reduce Redundancy Workload Analysis 9 / 18

 Set of m features  Collection of m-dimensional bit vectors  Partitioning over V  Union vector of all vectors in P i  Cost Function(sum of tuples that can be skipped) Partitioning Problem Problem Definition 10 / 18

 Cost Function over a partitioning  Problem 1 (Balanced MaxSkip Partitioning)  NP-hard using hypergraph bisection Partitioning Problem Balanced MaxSkip Partitioning 11 / 18

Partitioning Problem Example of Blocking 12 / 18

 Query Execution Feature-Based Data Skipping 13 / 18

 Data Update ‒ Infrequent ad-hoc updates, batch-inserted, batch-deleted ‒ Still fine-grained blocking partitions separately  Parameter Selection ‒ Two key parameters in blocking process ‒ numFeat : number of features ‒ minSize : minimum number of tuples per block  Default Parameter ‒ numFeat : < 50 ‒ MinSize : 64 – 128MB (which fits in HDFS block) Discussion 14 / 18

 Environment ‒ Amazon Spark EC2 cluster ‒ 25 m2.4xlarge instances ‒ 8 x 2.66 GHz CPU cores ‒ 64.8 GB RAM ‒ 2 x 840 GB disk storage ‒ HDFS  Datasets ‒ TPC-H benchmark data ‒ TPC-H Skewed ‒ Conviva  Anonymized user access log of video streams Experiment [1/3] 15 / 18

 FullScan : disable data skipping  Range1 : Shark’s data skipping  Range2 : Composite range partitioning (PowerDrill) Experiment [2/3] 16 / 18

 Effect of numFeat  Breakdown of blocking time Experiment [3/3] 17 / 18

 Fine-grained data blocking techniques ‒ Partition data tuples into blocks  Data skipping reduce 5-7x less data access  2-5x improvement in query response time ‒ Compared to range-based blocking techniques Conclusion 18 / 18