Low memory Map-Reduce

Size: px
Start display at page:

Download "Low memory Map-Reduce"

Transcription

1 Low memory -Reduce Hrishikesh Amur, Karsten Schwan, Georgia Tech. Wolf Richter, Athula Balachandran, Erik Zawadzki, Dave Andersen, CMU Michael Kaminsky, Intel 1

2 Datasets are growing People see value (even a little) in storing data rather than throwing it away 2

3 -Reduce Red. Red. 3

4 -Reduce Red. Red. 3

5 -Reduce Red. Red. 3

6 -Reduce Red. Red. 3

7 -Reduce Red. Red. 3

8 M R M R M 4

9 Data transmitted over network can be reduced! M R M R M 4

10 Aggregation is critical... 5

11 Aggregation is critical... Useful data is small (selection problems) 5

12 Aggregation is critical... Useful data is small (selection problems) Aggregate smaller than sum of parts (aggregation problems) 5

13 Aggregation is critical... Useful data is small (selection problems) Aggregate smaller than sum of parts (aggregation problems) Networks usually oversubscribed 5

14 ... as others have said Parallel databases allow aggregation, but queries become complex Dryad, Reduce and Hadoop. 6

15 Pre-aggregation in Hadoop C Red. C Red. C 7

16 Pre-aggregation in Hadoop C 7

17 Pre-aggregation in Hadoop C Can aggregation be performed in memory-constrained environments? 7

18 Why memory-constrained? 8

19 Why memory-constrained? Energy 8

20 Why memory-constrained? Energy Decreasing memory per core 8

21 Why memory-constrained? Energy Decreasing memory per core Fun :) 8

22 Pre-aggregation in Hadoop Sort Add 9

23 Pre-aggregation in Hadoop Sort Add 9

24 Pre-aggregation in Hadoop Sort Add 9

25 Pre-aggregation in Hadoop Sort Add 9

26 Pre-aggregation in Hadoop Sort Add In-memory sort limits aggregation 9

27 Minni: Low-memory -Reduce 10

28 Minni: Low-memory -Reduce Memory-efficient 10

29 Minni: Low-memory -Reduce Memory-efficient Performance scales with available memory 10

30 Minni: Low-memory -Reduce Memory-efficient Performance scales with available memory External aggregation using SSDs 10

31 Partial Aggregation Object (PAO) Key, Value User-defined create(key, value) destroy() merge(pao) serialize() deserialize() Distributed Aggregation for Data-Parallel Computing:Interfaces and Implementations, Yu et. al., SOSP 09 11

32 Grouping by Hashing Sort Add 12

33 Grouping by Hashing Hash 12

34 Grouping by Hashing Hash 12

35 Grouping by Hashing Hash Aggregate as you hash 12

36 Grouping by Hashing Hash Aggregate But the hash table as you hash might not fit in memory 12

37 External Aggregators Bucketing External Sort External Hash 13

38 Bucketing Files on SSD Hash Part. Hash 14

39 Bucketing Files on SSD Hash Part. Cap: 10 keys Hash 14

40 Bucketing Files on SSD 100 keys Hash Part. Cap: 10 keys Hash 14

41 Bucketing Files on SSD 100 keys Hash Part. Cap: 10 keys 12 buckets Hash 14

42 Bucketing Files on SSD 100 keys Hash Part. Has <10 keys 12 buckets Hash 14

43 Bucketing Files on SSD 100 keys Hash Part. Has <10 keys 12 buckets Hash Can aggregate in memory! 14

44 Bucketing Technique: SSDs can support writes to many files 15

45 Bucketing Technique: SSDs can support But, how many? writes to many files 15

46 Bucketing Technique: SSDs can support But, how many? writes to many files Files on SSD 100 keys Hash Part. Cap: 10 keys 12 buckets 15

47 Bucketing Technique: SSDs can support writes to many files 15

48 External Sort Hash Overflow File Overflow File Ext. Sort Add 16

49 External Sort Technique: Trade-off memory consumption for extra CPU work 17

50 External Hash Hash Ext. Hash 18

51 External Hash Hash Ext. Hash Use random read capabilities of SSDs 18

52 Pipelining Aggregators implemented as pipelines in Intel Threading Building Blocks (TBB) 19

53 Effects of token size (bucketing) Wordcount: 8G dataset 7 B/key 1 mil keys 20

54 Comparisons Wordcount: 8G dataset 7 B/key 1 mil keys 21

55 Recap of Techniques 22

56 Recap of Techniques Use SSD capabilities Parallel writes to multiple files High random read capabilities 22

57 Recap of Techniques Use SSD capabilities Parallel writes to multiple files High random read capabilities Trade-off latency for low memory consumption 22

58 Recap of Techniques Use SSD capabilities Parallel writes to multiple files High random read capabilities Trade-off latency for low memory consumption Trade-off CPU work for low memory consumption 22

59 Questions & Suggestions 23

Interruptible Tasks: Treating Memory Pressure as Interrupts for Highly Scalable Data-Parallel Programs

Interruptible Tasks: Treating Memory Pressure as Interrupts for Highly Scalable Data-Parallel Programs Interruptible s: Treating Pressure as Interrupts for Highly Scalable Data-Parallel Programs Lu Fang 1, Khanh Nguyen 1, Guoqing(Harry) Xu 1, Brian Demsky 1, Shan Lu 2 1 University of California, Irvine

More information

Camdoop Exploiting In-network Aggregation for Big Data Applications Paolo Costa

Camdoop Exploiting In-network Aggregation for Big Data Applications Paolo Costa Camdoop Exploiting In-network Aggregation for Big Data Applications costa@imperial.ac.uk joint work with Austin Donnelly, Antony Rowstron, and Greg O Shea (MSR Cambridge) MapReduce Overview Input file

More information

FAWN: A Fast Array of Wimpy Nodes

FAWN: A Fast Array of Wimpy Nodes FAWN: A Fast Array of Wimpy Nodes David G. Andersen, Jason Franklin, Michael Kaminsky *, Amar Phanishayee, Lawrence Tan, Vijay Vasudevan Carnegie Mellon University, * Intel Labs SOSP 09 CAS ICT Storage

More information

Module 3: Hashing Lecture 9: Static and Dynamic Hashing. The Lecture Contains: Static hashing. Hashing. Dynamic hashing. Extendible hashing.

Module 3: Hashing Lecture 9: Static and Dynamic Hashing. The Lecture Contains: Static hashing. Hashing. Dynamic hashing. Extendible hashing. The Lecture Contains: Hashing Dynamic hashing Extendible hashing Insertion file:///c /Documents%20and%20Settings/iitkrana1/My%20Documents/Google%20Talk%20Received%20Files/ist_data/lecture9/9_1.htm[6/14/2012

More information

MemC3: MemCache with CLOCK and Concurrent Cuckoo Hashing

MemC3: MemCache with CLOCK and Concurrent Cuckoo Hashing MemC3: MemCache with CLOCK and Concurrent Cuckoo Hashing Bin Fan (CMU), Dave Andersen (CMU), Michael Kaminsky (Intel Labs) NSDI 2013 http://www.pdl.cmu.edu/ 1 Goal: Improve Memcached 1. Reduce space overhead

More information

Chapter 12: Query Processing. Chapter 12: Query Processing

Chapter 12: Query Processing. Chapter 12: Query Processing Chapter 12: Query Processing Database System Concepts, 6 th Ed. See www.db-book.com for conditions on re-use Chapter 12: Query Processing Overview Measures of Query Cost Selection Operation Sorting Join

More information

Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li

Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li Be Fast, Cheap and in Control with SwitchKV Xiaozhou Li Raghav Sethi Michael Kaminsky David G. Andersen Michael J. Freedman Goal: fast and cost-effective key-value store Target: cluster-level storage for

More information

Indexing: Overview & Hashing. CS 377: Database Systems

Indexing: Overview & Hashing. CS 377: Database Systems Indexing: Overview & Hashing CS 377: Database Systems Recap: Data Storage Data items Records Memory DBMS Blocks blocks Files Different ways to organize files for better performance Disk Motivation for

More information

CS122 Lecture 3 Winter Term,

CS122 Lecture 3 Winter Term, CS122 Lecture 3 Winter Term, 2017-2018 2 Record-Level File Organization Last time, finished discussing block-level organization Can also organize data files at the record-level Heap file organization A

More information

Chapter 12: Query Processing

Chapter 12: Query Processing Chapter 12: Query Processing Database System Concepts, 6 th Ed. See www.db-book.com for conditions on re-use Overview Chapter 12: Query Processing Measures of Query Cost Selection Operation Sorting Join

More information

From architecture to algorithms: Lessons from the FAWN project

From architecture to algorithms: Lessons from the FAWN project From architecture to algorithms: Lessons from the FAWN project David Andersen, Vijay Vasudevan, Michael Kaminsky*, Michael A. Kozuch*, Amar Phanishayee, Lawrence Tan, Jason Franklin, Iulian Moraru, Sang

More information

DRYAD: DISTRIBUTED DATA- PARALLEL PROGRAMS FROM SEQUENTIAL BUILDING BLOCKS

DRYAD: DISTRIBUTED DATA- PARALLEL PROGRAMS FROM SEQUENTIAL BUILDING BLOCKS DRYAD: DISTRIBUTED DATA- PARALLEL PROGRAMS FROM SEQUENTIAL BUILDING BLOCKS Authors: Michael Isard, Mihai Budiu, Yuan Yu, Andrew Birrell, Dennis Fetterly Presenter: Zelin Dai WHAT IS DRYAD Combines computational

More information

FAWN as a Service. 1 Introduction. Jintian Liang CS244B December 13, 2017

FAWN as a Service. 1 Introduction. Jintian Liang CS244B December 13, 2017 Liang 1 Jintian Liang CS244B December 13, 2017 1 Introduction FAWN as a Service FAWN, an acronym for Fast Array of Wimpy Nodes, is a distributed cluster of inexpensive nodes designed to give users a view

More information

Developing MapReduce Programs

Developing MapReduce Programs Cloud Computing Developing MapReduce Programs Dell Zhang Birkbeck, University of London 2017/18 MapReduce Algorithm Design MapReduce: Recap Programmers must specify two functions: map (k, v) * Takes

More information

GLADE: A Scalable Framework for Efficient Analytics. Florin Rusu (University of California, Merced) Alin Dobra (University of Florida)

GLADE: A Scalable Framework for Efficient Analytics. Florin Rusu (University of California, Merced) Alin Dobra (University of Florida) DE: A Scalable Framework for Efficient Analytics Florin Rusu (University of California, Merced) Alin Dobra (University of Florida) Big Data Analytics Big Data Storage is cheap ($100 for 1TB disk) Everything

More information

FAWN. A Fast Array of Wimpy Nodes. David Andersen, Jason Franklin, Michael Kaminsky*, Amar Phanishayee, Lawrence Tan, Vijay Vasudevan

FAWN. A Fast Array of Wimpy Nodes. David Andersen, Jason Franklin, Michael Kaminsky*, Amar Phanishayee, Lawrence Tan, Vijay Vasudevan FAWN A Fast Array of Wimpy Nodes David Andersen, Jason Franklin, Michael Kaminsky*, Amar Phanishayee, Lawrence Tan, Vijay Vasudevan Carnegie Mellon University *Intel Labs Pittsburgh Energy in computing

More information

Big Data Management and NoSQL Databases

Big Data Management and NoSQL Databases NDBI040 Big Data Management and NoSQL Databases Lecture 2. MapReduce Doc. RNDr. Irena Holubova, Ph.D. holubova@ksi.mff.cuni.cz http://www.ksi.mff.cuni.cz/~holubova/ndbi040/ Framework A programming model

More information

Query Processing. Debapriyo Majumdar Indian Sta4s4cal Ins4tute Kolkata DBMS PGDBA 2016

Query Processing. Debapriyo Majumdar Indian Sta4s4cal Ins4tute Kolkata DBMS PGDBA 2016 Query Processing Debapriyo Majumdar Indian Sta4s4cal Ins4tute Kolkata DBMS PGDBA 2016 Slides re-used with some modification from www.db-book.com Reference: Database System Concepts, 6 th Ed. By Silberschatz,

More information

HiTune. Dataflow-Based Performance Analysis for Big Data Cloud

HiTune. Dataflow-Based Performance Analysis for Big Data Cloud HiTune Dataflow-Based Performance Analysis for Big Data Cloud Jinquan (Jason) Dai, Jie Huang, Shengsheng Huang, Bo Huang, Yan Liu Intel Asia-Pacific Research and Development Ltd Shanghai, China, 200241

More information

Chapter 13: Query Processing

Chapter 13: Query Processing Chapter 13: Query Processing! Overview! Measures of Query Cost! Selection Operation! Sorting! Join Operation! Other Operations! Evaluation of Expressions 13.1 Basic Steps in Query Processing 1. Parsing

More information

! A relational algebra expression may have many equivalent. ! Cost is generally measured as total elapsed time for

! A relational algebra expression may have many equivalent. ! Cost is generally measured as total elapsed time for Chapter 13: Query Processing Basic Steps in Query Processing! Overview! Measures of Query Cost! Selection Operation! Sorting! Join Operation! Other Operations! Evaluation of Expressions 1. Parsing and

More information

Chapter 13: Query Processing Basic Steps in Query Processing

Chapter 13: Query Processing Basic Steps in Query Processing Chapter 13: Query Processing Basic Steps in Query Processing! Overview! Measures of Query Cost! Selection Operation! Sorting! Join Operation! Other Operations! Evaluation of Expressions 1. Parsing and

More information

Building a Scalable Recommender System with Apache Spark, Apache Kafka and Elasticsearch

Building a Scalable Recommender System with Apache Spark, Apache Kafka and Elasticsearch Nick Pentreath Nov / 14 / 16 Building a Scalable Recommender System with Apache Spark, Apache Kafka and Elasticsearch About @MLnick Principal Engineer, IBM Apache Spark PMC Focused on machine learning

More information

Hash-Based Indexes. Chapter 11

Hash-Based Indexes. Chapter 11 Hash-Based Indexes Chapter 11 1 Introduction : Hash-based Indexes Best for equality selections. Cannot support range searches. Static and dynamic hashing techniques exist: Trade-offs similar to ISAM vs.

More information

The Road to a Complete Tweet Index

The Road to a Complete Tweet Index The Road to a Complete Tweet Index Yi Zhuang Staff Software Engineer @ Twitter Outline 1. Current Scale of Twitter Search 2. The History of Twitter Search Infra 3. Complete Tweet Index 4. Search Engine

More information

Ordered Indices To gain fast random access to records in a file, we can use an index structure. Each index structure is associated with a particular search key. Just like index of a book, library catalog,

More information

StreamBox: Modern Stream Processing on a Multicore Machine

StreamBox: Modern Stream Processing on a Multicore Machine StreamBox: Modern Stream Processing on a Multicore Machine Hongyu Miao and Heejin Park, Purdue ECE; Myeongjae Jeon and Gennady Pekhimenko, Microsoft Research; Kathryn S. McKinley, Google; Felix Xiaozhu

More information

STA141C: Big Data & High Performance Statistical Computing

STA141C: Big Data & High Performance Statistical Computing STA141C: Big Data & High Performance Statistical Computing Lecture 7: Parallel Computing Cho-Jui Hsieh UC Davis May 3, 2018 Outline Multi-core computing, distributed computing Multi-core computing tools

More information

Hash Joins for Multi-core CPUs. Benjamin Wagner

Hash Joins for Multi-core CPUs. Benjamin Wagner Hash Joins for Multi-core CPUs Benjamin Wagner Joins fundamental operator in query processing variety of different algorithms many papers publishing different results main question: is tuning to modern

More information

RESILIENT DISTRIBUTED DATASETS: A FAULT-TOLERANT ABSTRACTION FOR IN-MEMORY CLUSTER COMPUTING

RESILIENT DISTRIBUTED DATASETS: A FAULT-TOLERANT ABSTRACTION FOR IN-MEMORY CLUSTER COMPUTING RESILIENT DISTRIBUTED DATASETS: A FAULT-TOLERANT ABSTRACTION FOR IN-MEMORY CLUSTER COMPUTING Matei Zaharia, Mosharaf Chowdhury, Tathagata Das, Ankur Dave, Justin Ma, Murphy McCauley, Michael J. Franklin,

More information

CISC 7610 Lecture 2b The beginnings of NoSQL

CISC 7610 Lecture 2b The beginnings of NoSQL CISC 7610 Lecture 2b The beginnings of NoSQL Topics: Big Data Google s infrastructure Hadoop: open google infrastructure Scaling through sharding CAP theorem Amazon s Dynamo 5 V s of big data Everyone

More information

Gearing Hadoop towards HPC sytems

Gearing Hadoop towards HPC sytems Gearing Hadoop towards HPC sytems Xuanhua Shi http://grid.hust.edu.cn/xhshi Huazhong University of Science and Technology HPC Advisory Council China Workshop 2014, Guangzhou 2014/11/5 2 Internet Internet

More information

Clustering Documents. Case Study 2: Document Retrieval

Clustering Documents. Case Study 2: Document Retrieval Case Study 2: Document Retrieval Clustering Documents Machine Learning for Big Data CSE547/STAT548, University of Washington Sham Kakade April 21 th, 2015 Sham Kakade 2016 1 Document Retrieval Goal: Retrieve

More information

CIS 601 Graduate Seminar. Dr. Sunnie S. Chung Dhruv Patel ( ) Kalpesh Sharma ( )

CIS 601 Graduate Seminar. Dr. Sunnie S. Chung Dhruv Patel ( ) Kalpesh Sharma ( ) Guide: CIS 601 Graduate Seminar Presented By: Dr. Sunnie S. Chung Dhruv Patel (2652790) Kalpesh Sharma (2660576) Introduction Background Parallel Data Warehouse (PDW) Hive MongoDB Client-side Shared SQL

More information

SSS: An Implementation of Key-value Store based MapReduce Framework. Hirotaka Ogawa (AIST, Japan) Hidemoto Nakada Ryousei Takano Tomohiro Kudoh

SSS: An Implementation of Key-value Store based MapReduce Framework. Hirotaka Ogawa (AIST, Japan) Hidemoto Nakada Ryousei Takano Tomohiro Kudoh SSS: An Implementation of Key-value Store based MapReduce Framework Hirotaka Ogawa (AIST, Japan) Hidemoto Nakada Ryousei Takano Tomohiro Kudoh MapReduce A promising programming tool for implementing largescale

More information

Join Processing for Flash SSDs: Remembering Past Lessons

Join Processing for Flash SSDs: Remembering Past Lessons Join Processing for Flash SSDs: Remembering Past Lessons Jaeyoung Do, Jignesh M. Patel Department of Computer Sciences University of Wisconsin-Madison $/MB GB Flash Solid State Drives (SSDs) Benefits of

More information

Adaptive Query Processing on Prefix Trees Wolfgang Lehner

Adaptive Query Processing on Prefix Trees Wolfgang Lehner Adaptive Query Processing on Prefix Trees Wolfgang Lehner Fachgruppentreffen, 22.11.2012 TU München Prof. Dr.-Ing. Wolfgang Lehner > Challenges for Database Systems Three things are important in the database

More information

Time Series Storage with Apache Kudu (incubating)

Time Series Storage with Apache Kudu (incubating) Time Series Storage with Apache Kudu (incubating) Dan Burkert (Committer) dan@cloudera.com @danburkert Tweet about this talk: @getkudu or #kudu 1 Time Series machine metrics event logs sensor telemetry

More information

Parallel SAH k-d Tree Construction. Byn Choi, Rakesh Komuravelli, Victor Lu, Hyojin Sung, Robert L. Bocchino, Sarita V. Adve, John C.

Parallel SAH k-d Tree Construction. Byn Choi, Rakesh Komuravelli, Victor Lu, Hyojin Sung, Robert L. Bocchino, Sarita V. Adve, John C. Parallel SAH k-d Tree Construction Byn Choi, Rakesh Komuravelli, Victor Lu, Hyojin Sung, Robert L. Bocchino, Sarita V. Adve, John C. Hart Motivation Real-time Dynamic Ray Tracing Efficient rendering with

More information

Add/Drop Index (Concurrently) in Peloton

Add/Drop Index (Concurrently) in Peloton Add/Drop Index (Concurrently) in Peloton Rong Huang Xingyu Jin Ziheng Liao rhuang@andrew.cmu.edu xingyuj1@andrew.cmu.edu zihengl@andrew.cmu.edu Advanced Database Systems 15-721 Final Presentation 2 Recap:

More information

Advanced Database Systems

Advanced Database Systems Lecture IV Query Processing Kyumars Sheykh Esmaili Basic Steps in Query Processing 2 Query Optimization Many equivalent execution plans Choosing the best one Based on Heuristics, Cost Will be discussed

More information

Clustering Documents. Document Retrieval. Case Study 2: Document Retrieval

Clustering Documents. Document Retrieval. Case Study 2: Document Retrieval Case Study 2: Document Retrieval Clustering Documents Machine Learning for Big Data CSE547/STAT548, University of Washington Sham Kakade April, 2017 Sham Kakade 2017 1 Document Retrieval n Goal: Retrieve

More information

Routing Domains in Data Centre Networks. Morteza Kheirkhah. Informatics Department University of Sussex. Multi-Service Networks July 2011

Routing Domains in Data Centre Networks. Morteza Kheirkhah. Informatics Department University of Sussex. Multi-Service Networks July 2011 Routing Domains in Data Centre Networks Morteza Kheirkhah Informatics Department University of Sussex Multi-Service Networks July 2011 What is a Data Centre? Large-scale Data Centres (DC) consist of tens

More information

Apache Lucene 4. Robert Muir

Apache Lucene 4. Robert Muir Apache Lucene 4 Robert Muir Agenda Overview of Lucene Conclusion Resources Q & A Download of Lucene: core/ analysis/ queryparser/ highlighter/ suggest/ expressions/ join/ memory/ codecs/... core/ Lucene

More information

Database Applications (15-415)

Database Applications (15-415) Database Applications (15-415) DBMS Internals- Part VI Lecture 17, March 24, 2015 Mohammad Hammoud Today Last Two Sessions: DBMS Internals- Part V External Sorting How to Start a Company in Five (maybe

More information

GLADE: A Scalable Framework for Efficient Analytics. Florin Rusu University of California, Merced

GLADE: A Scalable Framework for Efficient Analytics. Florin Rusu University of California, Merced GLADE: A Scalable Framework for Efficient Analytics Florin Rusu University of California, Merced Motivation and Objective Large scale data processing Map-Reduce is standard technique Targeted to distributed

More information

Big Data Infrastructure CS 489/698 Big Data Infrastructure (Winter 2016)

Big Data Infrastructure CS 489/698 Big Data Infrastructure (Winter 2016) Big Data Infrastructure CS 489/698 Big Data Infrastructure (Winter 2016) Week 2: MapReduce Algorithm Design (2/2) January 14, 2016 Jimmy Lin David R. Cheriton School of Computer Science University of Waterloo

More information

Mnemosyne Lightweight Persistent Memory

Mnemosyne Lightweight Persistent Memory Mnemosyne Lightweight Persistent Memory Haris Volos Andres Jaan Tack, Michael M. Swift University of Wisconsin Madison Executive Summary Storage-Class Memory (SCM) enables memory-like storage Persistent

More information

Cuckoo Linear Algebra

Cuckoo Linear Algebra Cuckoo Linear Algebra Li Zhou, CMU Dave Andersen, CMU and Mu Li, CMU and Alexander Smola, CMU and select advertisement p(click user, query) = logist (hw, (user, query)i) select advertisement find weight

More information

Impala Intro. MingLi xunzhang

Impala Intro. MingLi xunzhang Impala Intro MingLi xunzhang Overview MPP SQL Query Engine for Hadoop Environment Designed for great performance BI Connected(ODBC/JDBC, Kerberos, LDAP, ANSI SQL) Hadoop Components HDFS, HBase, Metastore,

More information

CSIT5300: Advanced Database Systems

CSIT5300: Advanced Database Systems CSIT5300: Advanced Database Systems E10: Exercises on Query Processing Dr. Kenneth LEUNG Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong SAR,

More information

Indexing: B + -Tree. CS 377: Database Systems

Indexing: B + -Tree. CS 377: Database Systems Indexing: B + -Tree CS 377: Database Systems Recap: Indexes Data structures that organize records via trees or hashing Speed up search for a subset of records based on values in a certain field (search

More information

A Light-weight Compaction Tree to Reduce I/O Amplification toward Efficient Key-Value Stores

A Light-weight Compaction Tree to Reduce I/O Amplification toward Efficient Key-Value Stores A Light-weight Compaction Tree to Reduce I/O Amplification toward Efficient Key-Value Stores T i n g Y a o 1, J i g u a n g W a n 1, P i n g H u a n g 2, X u b i n He 2, Q i n g x i n G u i 1, F e i W

More information

Parallelism: The Real Y2K Crisis. Darek Mihocka August 14, 2008

Parallelism: The Real Y2K Crisis. Darek Mihocka August 14, 2008 Parallelism: The Real Y2K Crisis Darek Mihocka August 14, 2008 The Free Ride For decades, Moore's Law allowed CPU vendors to rely on steady clock speed increases: late 1970's: 1 MHz (6502) mid 1980's:

More information

Fundamentals of Database Systems Prof. Arnab Bhattacharya Department of Computer Science and Engineering Indian Institute of Technology, Kanpur

Fundamentals of Database Systems Prof. Arnab Bhattacharya Department of Computer Science and Engineering Indian Institute of Technology, Kanpur Fundamentals of Database Systems Prof. Arnab Bhattacharya Department of Computer Science and Engineering Indian Institute of Technology, Kanpur Lecture - 18 Database Indexing: Hashing We will start on

More information

YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores

YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores YCSB++ Benchmarking Tool Performance Debugging Advanced Features of Scalable Table Stores Swapnil Patil Milo Polte, Wittawat Tantisiriroj, Kai Ren, Lin Xiao, Julio Lopez, Garth Gibson, Adam Fuchs *, Billie

More information

COMP 430 Intro. to Database Systems. Indexing

COMP 430 Intro. to Database Systems. Indexing COMP 430 Intro. to Database Systems Indexing How does DB find records quickly? Various forms of indexing An index is automatically created for primary key. SQL gives us some control, so we should understand

More information

Data Modeling and Databases Ch 10: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich

Data Modeling and Databases Ch 10: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Data Modeling and Databases Ch 10: Query Processing - Algorithms Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Transactions (Locking, Logging) Metadata Mgmt (Schema, Stats) Application

More information

V Conclusions. V.1 Related work

V Conclusions. V.1 Related work V Conclusions V.1 Related work Even though MapReduce appears to be constructed specifically for performing group-by aggregations, there are also many interesting research work being done on studying critical

More information

Optimizing Testing Performance With Data Validation Option

Optimizing Testing Performance With Data Validation Option Optimizing Testing Performance With Data Validation Option 1993-2016 Informatica LLC. No part of this document may be reproduced or transmitted in any form, by any means (electronic, photocopying, recording

More information

Data Modeling and Databases Ch 9: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich

Data Modeling and Databases Ch 9: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Data Modeling and Databases Ch 9: Query Processing - Algorithms Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Transactions (Locking, Logging) Metadata Mgmt (Schema, Stats) Application

More information

Query Processing with Indexes. Announcements (February 24) Review. CPS 216 Advanced Database Systems

Query Processing with Indexes. Announcements (February 24) Review. CPS 216 Advanced Database Systems Query Processing with Indexes CPS 216 Advanced Database Systems Announcements (February 24) 2 More reading assignment for next week Buffer management (due next Wednesday) Homework #2 due next Thursday

More information

The Gap Between the Virtual Machine and the Real Machine. Charles Forgy Production Systems Tech

The Gap Between the Virtual Machine and the Real Machine. Charles Forgy Production Systems Tech The Gap Between the Virtual Machine and the Real Machine Charles Forgy Production Systems Tech How to Improve Performance Use better algorithms. Use parallelism. Make better use of the hardware. Argument

More information

MapReduce programming model

MapReduce programming model MapReduce programming model technology basics for data scientists Spring - 2014 Jordi Torres, UPC - BSC www.jorditorres.eu @JordiTorresBCN Warning! Slides are only for presenta8on guide We will discuss+debate

More information

An Introduction to Big Data Analysis using Spark

An Introduction to Big Data Analysis using Spark An Introduction to Big Data Analysis using Spark Mohamad Jaber American University of Beirut - Faculty of Arts & Sciences - Department of Computer Science May 17, 2017 Mohamad Jaber (AUB) Spark May 17,

More information

Introducing Hashing. Chapter 21. Copyright 2012 by Pearson Education, Inc. All rights reserved

Introducing Hashing. Chapter 21. Copyright 2012 by Pearson Education, Inc. All rights reserved Introducing Hashing Chapter 21 Contents What Is Hashing? Hash Functions Computing Hash Codes Compressing a Hash Code into an Index for the Hash Table A demo of hashing (after) ARRAY insert hash index =

More information

Cuckoo Filter: Practically Better Than Bloom

Cuckoo Filter: Practically Better Than Bloom Cuckoo Filter: Practically Better Than Bloom Bin Fan (CMU/Google) David Andersen (CMU) Michael Kaminsky (Intel Labs) Michael Mitzenmacher (Harvard) 1 What is Bloom Filter? A Compact Data Structure Storing

More information

CSIT5300: Advanced Database Systems

CSIT5300: Advanced Database Systems CSIT5300: Advanced Database Systems L08: B + -trees and Dynamic Hashing Dr. Kenneth LEUNG Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong SAR,

More information

MapReduce Spark. Some slides are adapted from those of Jeff Dean and Matei Zaharia

MapReduce Spark. Some slides are adapted from those of Jeff Dean and Matei Zaharia MapReduce Spark Some slides are adapted from those of Jeff Dean and Matei Zaharia What have we learnt so far? Distributed storage systems consistency semantics protocols for fault tolerance Paxos, Raft,

More information

CMSC424: Database Design. Instructor: Amol Deshpande

CMSC424: Database Design. Instructor: Amol Deshpande CMSC424: Database Design Instructor: Amol Deshpande amol@cs.umd.edu Databases Data Models Conceptual representa1on of the data Data Retrieval How to ask ques1ons of the database How to answer those ques1ons

More information

Shark: SQL and Rich Analytics at Scale. Michael Xueyuan Han Ronny Hajoon Ko

Shark: SQL and Rich Analytics at Scale. Michael Xueyuan Han Ronny Hajoon Ko Shark: SQL and Rich Analytics at Scale Michael Xueyuan Han Ronny Hajoon Ko What Are The Problems? Data volumes are expanding dramatically Why Is It Hard? Needs to scale out Managing hundreds of machines

More information

ZFS The Future Of File Systems. C Sanjeev Kumar Charly V. Joseph Mewan Peter D Almeida Srinidhi K.

ZFS The Future Of File Systems. C Sanjeev Kumar Charly V. Joseph Mewan Peter D Almeida Srinidhi K. ZFS The Future Of File Systems C Sanjeev Kumar Charly V. Joseph Mewan Peter D Almeida Srinidhi K. Introduction What is a File System? File systems are an integral part of any operating systems with the

More information

Realtime Recommendations

Realtime Recommendations Realtime Recommendations with Redis Torben Brodt plista GmbH April 25th, 2013 NoSQL Search Roadshow http://nosqlroadshow.com/nosql-berlin-2013/ Introduction Torben Brodt, Head of Data Engineering computer

More information

Selection Queries. to answer a selection query (ssn=10) needs to traverse a full path.

Selection Queries. to answer a selection query (ssn=10) needs to traverse a full path. Hashing B+-tree is perfect, but... Selection Queries to answer a selection query (ssn=) needs to traverse a full path. In practice, 3-4 block accesses (depending on the height of the tree, buffering) Any

More information

DryadLINQ. by Yuan Yu et al., OSDI 08. Ilias Giechaskiel. January 28, Cambridge University, R212

DryadLINQ. by Yuan Yu et al., OSDI 08. Ilias Giechaskiel. January 28, Cambridge University, R212 DryadLINQ by Yuan Yu et al., OSDI 08 Ilias Giechaskiel Cambridge University, R212 ig305@cam.ac.uk January 28, 2014 Conclusions Takeaway Messages SQL cannot express iteration Unsuitable for machine learning,

More information

Track Join. Distributed Joins with Minimal Network Traffic. Orestis Polychroniou! Rajkumar Sen! Kenneth A. Ross

Track Join. Distributed Joins with Minimal Network Traffic. Orestis Polychroniou! Rajkumar Sen! Kenneth A. Ross Track Join Distributed Joins with Minimal Network Traffic Orestis Polychroniou Rajkumar Sen Kenneth A. Ross Local Joins Algorithms Hash Join Sort Merge Join Index Join Nested Loop Join Spilling to disk

More information

Database System Concepts

Database System Concepts Chapter 13: Query Processing s Departamento de Engenharia Informática Instituto Superior Técnico 1 st Semester 2008/2009 Slides (fortemente) baseados nos slides oficiais do livro c Silberschatz, Korth

More information

Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies

Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies Shadowfax: Scaling in Heterogeneous Cluster Systems via GPGPU Assemblies Alexander Merritt, Vishakha Gupta, Abhishek Verma, Ada Gavrilovska, Karsten Schwan {merritt.alex,abhishek.verma}@gatech.edu {vishakha,ada,schwan}@cc.gtaech.edu

More information

Apache Flink- A System for Batch and Realtime Stream Processing

Apache Flink- A System for Batch and Realtime Stream Processing Apache Flink- A System for Batch and Realtime Stream Processing Lecture Notes Winter semester 2016 / 2017 Ludwig-Maximilians-University Munich Prof Dr. Matthias Schubert 2016 Introduction to Apache Flink

More information

Optimized Data Integration for the MSO Market

Optimized Data Integration for the MSO Market Optimized Data Integration for the MSO Market Actions at the speed of data For Real-time Decisioning and Big Data Problems VelociData for FinTech and the Enterprise VelociData s technology has been providing

More information

Most of the slides in this lecture are either from or adapted from slides provided by the authors of the textbook Computer Systems: A Programmer s

Most of the slides in this lecture are either from or adapted from slides provided by the authors of the textbook Computer Systems: A Programmer s Most of the slides in this lecture are either from or adapted from slides provided by the authors of the textbook Computer Systems: A Programmer s Perspective, 2 nd Edition and are provided from the website

More information

Detecting and Fixing Memory-Related Performance Problems in Managed Languages

Detecting and Fixing Memory-Related Performance Problems in Managed Languages Detecting and Fixing -Related Performance Problems in Managed Languages Lu Fang Committee: Prof. Guoqing Xu (Chair), Prof. Alex Nicolau, Prof. Brian Demsky University of California, Irvine May 26, 2017,

More information

An In-Depth Analysis of Data Aggregation Cost Factors in a Columnar In-Memory Database

An In-Depth Analysis of Data Aggregation Cost Factors in a Columnar In-Memory Database An In-Depth Analysis of Data Aggregation Cost Factors in a Columnar In-Memory Database Stephan Müller, Hasso Plattner Enterprise Platform and Integration Concepts Hasso Plattner Institute, Potsdam (Germany)

More information

More B-trees, Hash Tables, etc. CS157B Chris Pollett Feb 21, 2005.

More B-trees, Hash Tables, etc. CS157B Chris Pollett Feb 21, 2005. More B-trees, Hash Tables, etc. CS157B Chris Pollett Feb 21, 2005. Outline B-tree Domain of Application B-tree Operations Hash Tables on Disk Hash Table Operations Extensible Hash Tables Multidimensional

More information

TrafficDB: HERE s High Performance Shared-Memory Data Store Ricardo Fernandes, Piotr Zaczkowski, Bernd Göttler, Conor Ettinoffe, and Anis Moussa

TrafficDB: HERE s High Performance Shared-Memory Data Store Ricardo Fernandes, Piotr Zaczkowski, Bernd Göttler, Conor Ettinoffe, and Anis Moussa TrafficDB: HERE s High Performance Shared-Memory Data Store Ricardo Fernandes, Piotr Zaczkowski, Bernd Göttler, Conor Ettinoffe, and Anis Moussa EPL646: Advanced Topics in Databases Christos Hadjistyllis

More information

Chapter 17. Disk Storage, Basic File Structures, and Hashing. Records. Blocking

Chapter 17. Disk Storage, Basic File Structures, and Hashing. Records. Blocking Chapter 17 Disk Storage, Basic File Structures, and Hashing Records Fixed and variable length records Records contain fields which have values of a particular type (e.g., amount, date, time, age) Fields

More information

MapReduce Algorithm Design

MapReduce Algorithm Design MapReduce Algorithm Design Contents Combiner and in mapper combining Complex keys and values Secondary Sorting Combiner and in mapper combining Purpose Carry out local aggregation before shuffle and sort

More information

Recall use of logical clocks

Recall use of logical clocks Causal Consistency Consistency models Linearizability Causal Eventual COS 418: Distributed Systems Lecture 16 Sequential Michael Freedman 2 Recall use of logical clocks Lamport clocks: C(a) < C(z) Conclusion:

More information

Performance Benefits of Running RocksDB on Samsung NVMe SSDs

Performance Benefits of Running RocksDB on Samsung NVMe SSDs Performance Benefits of Running RocksDB on Samsung NVMe SSDs A Detailed Analysis 25 Samsung Semiconductor Inc. Executive Summary The industry has been experiencing an exponential data explosion over the

More information

An Adaptive Query Execution Engine for Data Integration

An Adaptive Query Execution Engine for Data Integration An Adaptive Query Execution Engine for Data Integration Zachary Ives, Daniela Florescu, Marc Friedman, Alon Levy, Daniel S. Weld University of Washington Presented by Peng Li@CS.UBC 1 Outline The Background

More information

Hashing. Data organization in main memory or disk

Hashing. Data organization in main memory or disk Hashing Data organization in main memory or disk sequential, indexed sequential, binary trees, the location of a record depends on other keys unnecessary key comparisons to find a key The goal of hashing

More information

ENERGY-EFFICIENT VISUALIZATION PIPELINES A CASE STUDY IN CLIMATE SIMULATION

ENERGY-EFFICIENT VISUALIZATION PIPELINES A CASE STUDY IN CLIMATE SIMULATION ENERGY-EFFICIENT VISUALIZATION PIPELINES A CASE STUDY IN CLIMATE SIMULATION Vignesh Adhinarayanan Ph.D. (CS) Student Synergy Lab, Virginia Tech INTRODUCTION Supercomputers are constrained by power Power

More information

Voldemort. Smruti R. Sarangi. Department of Computer Science Indian Institute of Technology New Delhi, India. Overview Design Evaluation

Voldemort. Smruti R. Sarangi. Department of Computer Science Indian Institute of Technology New Delhi, India. Overview Design Evaluation Voldemort Smruti R. Sarangi Department of Computer Science Indian Institute of Technology New Delhi, India Smruti R. Sarangi Leader Election 1/29 Outline 1 2 3 Smruti R. Sarangi Leader Election 2/29 Data

More information

Ch 5 : Query Processing & Optimization

Ch 5 : Query Processing & Optimization Ch 5 : Query Processing & Optimization Basic Steps in Query Processing 1. Parsing and translation 2. Optimization 3. Evaluation Basic Steps in Query Processing (Cont.) Parsing and translation translate

More information

SEMem: deployment of MPI-based in-memory storage for Hadoop on supercomputers

SEMem: deployment of MPI-based in-memory storage for Hadoop on supercomputers 1 SEMem: deployment of MPI-based in-memory storage for Hadoop on supercomputers and Shigeru Chiba The University of Tokyo, Japan 2 Running Hadoop on modern supercomputers Hadoop assumes every compute node

More information

C 1. Recap. CSE 486/586 Distributed Systems Distributed File Systems. Traditional Distributed File Systems. Local File Systems.

C 1. Recap. CSE 486/586 Distributed Systems Distributed File Systems. Traditional Distributed File Systems. Local File Systems. Recap CSE 486/586 Distributed Systems Distributed File Systems Optimistic quorum Distributed transactions with replication One copy serializability Primary copy replication Read-one/write-all replication

More information

Dataflow Architectures. Karin Strauss

Dataflow Architectures. Karin Strauss Dataflow Architectures Karin Strauss Introduction Dataflow machines: programmable computers with hardware optimized for fine grain data-driven parallel computation fine grain: at the instruction granularity

More information

The physical database. Contents - physical database design DATABASE DESIGN I - 1DL300. Introduction to Physical Database Design

The physical database. Contents - physical database design DATABASE DESIGN I - 1DL300. Introduction to Physical Database Design DATABASE DESIGN I - 1DL300 Fall 2011 Introduction to Physical Database Design Elmasri/Navathe ch 16 and 17 Padron-McCarthy/Risch ch 21 and 22 An introductory course on database systems http://www.it.uu.se/edu/course/homepage/dbastekn/ht11

More information

Database Applications (15-415)

Database Applications (15-415) Database Applications (15-415) DBMS Internals- Part V Lecture 15, March 15, 2015 Mohammad Hammoud Today Last Session: DBMS Internals- Part IV Tree-based (i.e., B+ Tree) and Hash-based (i.e., Extendible

More information

Chronos Latency - Pole Position Performance

Chronos Latency - Pole Position Performance WHITE PAPER Chronos Latency - Pole Position Performance By G. Rinaldi and M. T. Moreira, Chronos Tech 1 Introduction Modern SoC performance is often limited by the capability to exchange information at

More information