SlideShare a Scribd company logo
1 of 28
Download to read offline
Google Bigtable
(Bigtable: A Distributed Storage System for Structured Data)

           Komadinovic Vanja, Vast Platform team
Presentation overview

- introduction
- design
- basic implementation
- GFS - HDFS introduction
- MapReduce introduction
- implementation
- HBase - Apache Bigtable solution
- performances and usage case
- some thoughts for discussion
Introduction

- what is bigtable?
- why need for something more than SQL?
- machines number increase - scaling
- fault tolerance
- row, column, cell
- versioning
- storage on distributed file system
- read, write operations on row or set of rows
- scanner - iterator
- no need for indexing
Design

- column family basic unit of access control

- column family must be created before data insertion

- column labels number is unlimited

- row key is arbitrary string, max. size 64Kb

- every cell has copies in time - configurable
Design
Design

- table is divided in tablets
- tablets are distributed on machines
- each tablet contains N rows
- tablet size 100 - 200 Mb
- tablet server splits tablet when it grows, only operation
initiated by tablet server
- tablets are stored on GFS
Design
Basic implementation
GFS - HDFS introduction

- distributed file system
- files are divided in relatively large blocks, 64Mb
- blocks of one file are stored on multiple machines
- every block is replicated in configurable number of copies
- fault tolerance guaranteed
- Master - Slaves architecture
GFS - HDFS introduction: overview
MapReduce introduction

- framework build on top of (GFS) HDFS
- used for distributed data processing
- Master - Slaves architecture
- between data elements there is no dependency
- everything is executed in parallel
- two parts: Mapper and Reducer
- Mapper reads input data and emits ( key, value pairs ) to
Reducer
- Data on Reducer is grouped by key
- Reducer produces output data
- Combiner and Partitioner - who are they, and what they want?
MapReduce introduction: overview
Implementation

- Chubby - distributed lock service, synchronization point, start
point for master, tablet server and even clients
- Master server, only one, guaranteed by Chubby
- in case of Master tragical death tablet servers continue to
serve client
- Master decides who serves each tablet ( including ROOT and
METATABLE )
- tablets are saved on GFS
- all trafic between clients and Bigtable is done directly within
client and tablet server ( Master is not required )
- client caches METATABLE, on first fault recache is done
Implementation: overview
Implementation: oragnization
- two special tables: ROOT and METADATA
- ROOT never gets splitted
- location of ROOT is in Chubby file
- estimated ROOT size 128Mb => 2^34 tablets
- METADATA contains all user tablets, row key is an encoding
of the tablet’s table identifier and its end row
Implementation

- SSTable - file representation of tablet
   - consist of blocks
   - column family labels are stored side by side
   - index of blocks at end of file

- compaction
   - minor
   - major

- locality groups

- caching
HBase - Apache Bigtable solution 
HBase - Bigtable synonyms

- Bigtable ~ HBase
- tablet ~ region
- Master server ~ HBase master
- Tablet server ~ HBase Region server
- Chubby ~ Zookeeper ???
- GFS ~ HDFS
- MapReduce ~ MapReduce :)
- SSTable file ~ MapFile file
HBase differences

- MapFiles index is stored in separate file instead at end of file
as in SSTable

- Unlike Bigtable which identifies a row range by the table name
and end-key, HBase identifies a row range by the table name
and start-key

- when the HBase master dies, the cluster will shut down
Performance and usage case
- number of 1000-byte values read/written per second
- table shows the rate per tablet server

           Experiment   1 TS    50 TS   250 TS   500 TS

            random      1212     593     479      241
             reads
             random     10811   8511    8000     6250
           reads mem

            random      8850    3745    3425     2000
             writes
           sequential   4425    2463    2625     2469
             reads
           sequential   8547    3623    2451     1905
             writes
             scans      15385   10526   9524     7843
Performances and usage case

Scaling:

- performance of random reads from memory increases by
almost a factor of 300 as the number of tablet server increases
by a factor of 500

- significant drop in per-server throughput when going from 1 to
50 tablet servers

- the random read benchmark shows the worst scaling (an
increase in aggregate throughput by only a factor of 100 for a
500-fold increase in number of servers)
Performances and usage case

Real users:

- Google Analytics:
   - It provides aggregate statistics, such as the number of
unique visitors per day and the page views per URL per day
   - two tables - raw user clicks (~200Tb, 14% compression)
and summary (~20Tb, 29% compression)

- Google Earth:
   - Google operates a collection of services that provide users
with access to high-resolution satellite imagery of the world’s
surface
   - two tables - (~70Tb, imagery, no compression), (~500Gb,
cache, served on hundreds of tablet servers
Cooked by Vast: RecordStack

Introduction:

- long-term archival storage for vertical data to HDFS
- optimized for later analytical processing.
- stores:
   - listings data
   - transactional log data
   - user clicks
   - leads
   - other reporting stuff
Cooked by Vast: RecordStack

Goals:

- Efficient archival of records to HDFS - conserving space
- Checksuming records and tracking change
- Optmization of HDFS - stored structure by periodical data
compacting
- Simple access from Hadoop tasks for analytical processing
- Simple access via client application - Remote console
- generation of reports
Some thoughts for discussion

- doing Bigtable with some number of "classic" databases?

- expose of Chubby to client, is it required?

- why not open source Bigtable?

- run of tablet servers on data nodes - closer to data?
Useful links

- Scalability: http://scalability.rs/

- NoSQL Summer: http://nosqlsummer.org/

- Google Bigtable paper: http://tinyurl.com/37rlevv

- HBase Architecture: http://tinyurl.com/3y3fhbk

- HBase related blog: http://www.larsgeorge.com/

- Hadoop: The Definitive guide: http://tinyurl.com/36ponz3
If you, maybe, want to contact me?



e-mail: vanjakom@gmail.com, vanja@vast.com

skype: komadinovic.vanja

mobile: =381 (64) 296 03 43

twitter: vanjakom
Thanks !!!

More Related Content

What's hot

ClickHouse Materialized Views: The Magic Continues
ClickHouse Materialized Views: The Magic ContinuesClickHouse Materialized Views: The Magic Continues
ClickHouse Materialized Views: The Magic ContinuesAltinity Ltd
 
Apache Kudu: Technical Deep Dive


Apache Kudu: Technical Deep Dive

Apache Kudu: Technical Deep Dive


Apache Kudu: Technical Deep Dive

Cloudera, Inc.
 
High Performance, High Reliability Data Loading on ClickHouse
High Performance, High Reliability Data Loading on ClickHouseHigh Performance, High Reliability Data Loading on ClickHouse
High Performance, High Reliability Data Loading on ClickHouseAltinity Ltd
 
Dynamo and BigTable in light of the CAP theorem
Dynamo and BigTable in light of the CAP theoremDynamo and BigTable in light of the CAP theorem
Dynamo and BigTable in light of the CAP theoremGrisha Weintraub
 
A Fast Intro to Fast Query with ClickHouse, by Robert Hodges
A Fast Intro to Fast Query with ClickHouse, by Robert HodgesA Fast Intro to Fast Query with ClickHouse, by Robert Hodges
A Fast Intro to Fast Query with ClickHouse, by Robert HodgesAltinity Ltd
 
ClickHouse Introduction by Alexander Zaitsev, Altinity CTO
ClickHouse Introduction by Alexander Zaitsev, Altinity CTOClickHouse Introduction by Alexander Zaitsev, Altinity CTO
ClickHouse Introduction by Alexander Zaitsev, Altinity CTOAltinity Ltd
 
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...DataWorks Summit/Hadoop Summit
 
ClickHouse Deep Dive, by Aleksei Milovidov
ClickHouse Deep Dive, by Aleksei MilovidovClickHouse Deep Dive, by Aleksei Milovidov
ClickHouse Deep Dive, by Aleksei MilovidovAltinity Ltd
 
Introduction to Apache Hadoop Eco-System
Introduction to Apache Hadoop Eco-SystemIntroduction to Apache Hadoop Eco-System
Introduction to Apache Hadoop Eco-SystemMd. Hasan Basri (Angel)
 
Overview - IBM Big Data Platform
Overview - IBM Big Data PlatformOverview - IBM Big Data Platform
Overview - IBM Big Data PlatformVikas Manoria
 
Making Apache Spark Better with Delta Lake
Making Apache Spark Better with Delta LakeMaking Apache Spark Better with Delta Lake
Making Apache Spark Better with Delta LakeDatabricks
 
Data Engineer's Lunch #83: Strategies for Migration to Apache Iceberg
Data Engineer's Lunch #83: Strategies for Migration to Apache IcebergData Engineer's Lunch #83: Strategies for Migration to Apache Iceberg
Data Engineer's Lunch #83: Strategies for Migration to Apache IcebergAnant Corporation
 
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEO
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEOClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEO
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEOAltinity Ltd
 
Google file system GFS
Google file system GFSGoogle file system GFS
Google file system GFSzihad164
 
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlare
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlareClickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlare
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlareAltinity Ltd
 
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...Spark Summit
 
Getting Started with Amazon Redshift
Getting Started with Amazon RedshiftGetting Started with Amazon Redshift
Getting Started with Amazon RedshiftAmazon Web Services
 

What's hot (20)

Intro to HBase
Intro to HBaseIntro to HBase
Intro to HBase
 
ClickHouse Materialized Views: The Magic Continues
ClickHouse Materialized Views: The Magic ContinuesClickHouse Materialized Views: The Magic Continues
ClickHouse Materialized Views: The Magic Continues
 
Apache Kudu: Technical Deep Dive


Apache Kudu: Technical Deep Dive

Apache Kudu: Technical Deep Dive


Apache Kudu: Technical Deep Dive


 
High Performance, High Reliability Data Loading on ClickHouse
High Performance, High Reliability Data Loading on ClickHouseHigh Performance, High Reliability Data Loading on ClickHouse
High Performance, High Reliability Data Loading on ClickHouse
 
Dynamo and BigTable in light of the CAP theorem
Dynamo and BigTable in light of the CAP theoremDynamo and BigTable in light of the CAP theorem
Dynamo and BigTable in light of the CAP theorem
 
A Fast Intro to Fast Query with ClickHouse, by Robert Hodges
A Fast Intro to Fast Query with ClickHouse, by Robert HodgesA Fast Intro to Fast Query with ClickHouse, by Robert Hodges
A Fast Intro to Fast Query with ClickHouse, by Robert Hodges
 
ClickHouse Introduction by Alexander Zaitsev, Altinity CTO
ClickHouse Introduction by Alexander Zaitsev, Altinity CTOClickHouse Introduction by Alexander Zaitsev, Altinity CTO
ClickHouse Introduction by Alexander Zaitsev, Altinity CTO
 
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...
The Columnar Era: Leveraging Parquet, Arrow and Kudu for High-Performance Ana...
 
Cassandra
CassandraCassandra
Cassandra
 
The delta architecture
The delta architectureThe delta architecture
The delta architecture
 
ClickHouse Deep Dive, by Aleksei Milovidov
ClickHouse Deep Dive, by Aleksei MilovidovClickHouse Deep Dive, by Aleksei Milovidov
ClickHouse Deep Dive, by Aleksei Milovidov
 
Introduction to Apache Hadoop Eco-System
Introduction to Apache Hadoop Eco-SystemIntroduction to Apache Hadoop Eco-System
Introduction to Apache Hadoop Eco-System
 
Overview - IBM Big Data Platform
Overview - IBM Big Data PlatformOverview - IBM Big Data Platform
Overview - IBM Big Data Platform
 
Making Apache Spark Better with Delta Lake
Making Apache Spark Better with Delta LakeMaking Apache Spark Better with Delta Lake
Making Apache Spark Better with Delta Lake
 
Data Engineer's Lunch #83: Strategies for Migration to Apache Iceberg
Data Engineer's Lunch #83: Strategies for Migration to Apache IcebergData Engineer's Lunch #83: Strategies for Migration to Apache Iceberg
Data Engineer's Lunch #83: Strategies for Migration to Apache Iceberg
 
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEO
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEOClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEO
ClickHouse tips and tricks. Webinar slides. By Robert Hodges, Altinity CEO
 
Google file system GFS
Google file system GFSGoogle file system GFS
Google file system GFS
 
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlare
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlareClickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlare
Clickhouse Capacity Planning for OLAP Workloads, Mik Kocikowski of CloudFlare
 
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...
Building Real-Time BI Systems with Kafka, Spark, and Kudu: Spark Summit East ...
 
Getting Started with Amazon Redshift
Getting Started with Amazon RedshiftGetting Started with Amazon Redshift
Getting Started with Amazon Redshift
 

Viewers also liked

Big table
Big tableBig table
Big tablePSIT
 
Google Bigtable paper presentation
Google Bigtable paper presentationGoogle Bigtable paper presentation
Google Bigtable paper presentationvanjakom
 
The Google File System (GFS)
The Google File System (GFS)The Google File System (GFS)
The Google File System (GFS)Romain Jacotin
 
Google Architecture - Breaking it Open
Google Architecture - Breaking it OpenGoogle Architecture - Breaking it Open
Google Architecture - Breaking it OpenHARMAN Services
 
Bigtable: A Distributed Storage System for Structured Data
Bigtable: A Distributed Storage System for Structured DataBigtable: A Distributed Storage System for Structured Data
Bigtable: A Distributed Storage System for Structured Dataelliando dias
 
Big Data Berlin - Criteo
Big Data Berlin - CriteoBig Data Berlin - Criteo
Big Data Berlin - CriteoSofian Djamaa
 
30 billion requests per day with a NoSQL architecture (2013)
30 billion requests per day with a NoSQL architecture (2013)30 billion requests per day with a NoSQL architecture (2013)
30 billion requests per day with a NoSQL architecture (2013)Julien SIMON
 
Bigtable
BigtableBigtable
Bigtablenextlib
 
Introduction_OF_Hadoop_and_BigData
Introduction_OF_Hadoop_and_BigDataIntroduction_OF_Hadoop_and_BigData
Introduction_OF_Hadoop_and_BigDataNilay Mishra
 
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEY
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEYROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEY
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEYSupport Driven
 
I've (probably) been using Google App Engine for a week longer than you have
I've (probably) been using Google App Engine for a week longer than you haveI've (probably) been using Google App Engine for a week longer than you have
I've (probably) been using Google App Engine for a week longer than you haveSimon Willison
 
HFile: A Block-Indexed File Format to Store Sorted Key-Value Pairs
HFile: A Block-Indexed File Format to Store Sorted Key-Value PairsHFile: A Block-Indexed File Format to Store Sorted Key-Value Pairs
HFile: A Block-Indexed File Format to Store Sorted Key-Value PairsSchubert Zhang
 
Introduction to Python
Introduction to PythonIntroduction to Python
Introduction to PythonBogdan Sabău
 
Anti theftsystemforvechicles1 final
Anti theftsystemforvechicles1 finalAnti theftsystemforvechicles1 final
Anti theftsystemforvechicles1 finalAshu0711
 

Viewers also liked (20)

Big table
Big tableBig table
Big table
 
google Bigtable
google Bigtablegoogle Bigtable
google Bigtable
 
Google Bigtable paper presentation
Google Bigtable paper presentationGoogle Bigtable paper presentation
Google Bigtable paper presentation
 
The Google File System (GFS)
The Google File System (GFS)The Google File System (GFS)
The Google File System (GFS)
 
Bigtable
BigtableBigtable
Bigtable
 
Google Architecture - Breaking it Open
Google Architecture - Breaking it OpenGoogle Architecture - Breaking it Open
Google Architecture - Breaking it Open
 
Bigtable: A Distributed Storage System for Structured Data
Bigtable: A Distributed Storage System for Structured DataBigtable: A Distributed Storage System for Structured Data
Bigtable: A Distributed Storage System for Structured Data
 
Big Data Berlin - Criteo
Big Data Berlin - CriteoBig Data Berlin - Criteo
Big Data Berlin - Criteo
 
30 billion requests per day with a NoSQL architecture (2013)
30 billion requests per day with a NoSQL architecture (2013)30 billion requests per day with a NoSQL architecture (2013)
30 billion requests per day with a NoSQL architecture (2013)
 
Bigtable
BigtableBigtable
Bigtable
 
Introduction_OF_Hadoop_and_BigData
Introduction_OF_Hadoop_and_BigDataIntroduction_OF_Hadoop_and_BigData
Introduction_OF_Hadoop_and_BigData
 
Php.Mvc Presentation
Php.Mvc PresentationPhp.Mvc Presentation
Php.Mvc Presentation
 
Big table
Big tableBig table
Big table
 
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEY
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEYROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEY
ROCKING YOUR SEAT AT THE BIG TABLE - ROB BAILEY
 
I've (probably) been using Google App Engine for a week longer than you have
I've (probably) been using Google App Engine for a week longer than you haveI've (probably) been using Google App Engine for a week longer than you have
I've (probably) been using Google App Engine for a week longer than you have
 
HFile: A Block-Indexed File Format to Store Sorted Key-Value Pairs
HFile: A Block-Indexed File Format to Store Sorted Key-Value PairsHFile: A Block-Indexed File Format to Store Sorted Key-Value Pairs
HFile: A Block-Indexed File Format to Store Sorted Key-Value Pairs
 
Introduction to Python
Introduction to PythonIntroduction to Python
Introduction to Python
 
Google's BigTable
Google's BigTableGoogle's BigTable
Google's BigTable
 
Big table
Big tableBig table
Big table
 
Anti theftsystemforvechicles1 final
Anti theftsystemforvechicles1 finalAnti theftsystemforvechicles1 final
Anti theftsystemforvechicles1 final
 

Similar to Google Bigtable Paper Presentation

HBase, crazy dances on the elephant back.
HBase, crazy dances on the elephant back.HBase, crazy dances on the elephant back.
HBase, crazy dances on the elephant back.Roman Nikitchenko
 
GCP Data Engineer cheatsheet
GCP Data Engineer cheatsheetGCP Data Engineer cheatsheet
GCP Data Engineer cheatsheetGuang Xu
 
Apache hadoop, hdfs and map reduce Overview
Apache hadoop, hdfs and map reduce OverviewApache hadoop, hdfs and map reduce Overview
Apache hadoop, hdfs and map reduce OverviewNisanth Simon
 
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)Sascha Dittmann
 
Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017Vinoth Chandar
 
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...Yahoo Developer Network
 
MapR M7: Providing an enterprise quality Apache HBase API
MapR M7: Providing an enterprise quality Apache HBase APIMapR M7: Providing an enterprise quality Apache HBase API
MapR M7: Providing an enterprise quality Apache HBase APImcsrivas
 
Inside MapR's M7
Inside MapR's M7Inside MapR's M7
Inside MapR's M7Ted Dunning
 
Kudu - Fast Analytics on Fast Data
Kudu - Fast Analytics on Fast DataKudu - Fast Analytics on Fast Data
Kudu - Fast Analytics on Fast DataRyan Bosshart
 
02.28.13 WANdisco ApacheCon 2013
02.28.13 WANdisco ApacheCon 201302.28.13 WANdisco ApacheCon 2013
02.28.13 WANdisco ApacheCon 2013WANdisco Plc
 
Meethadoop
MeethadoopMeethadoop
MeethadoopIIIT-H
 
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...Cloudera, Inc.
 

Similar to Google Bigtable Paper Presentation (20)

Gcp data engineer
Gcp data engineerGcp data engineer
Gcp data engineer
 
HBase, crazy dances on the elephant back.
HBase, crazy dances on the elephant back.HBase, crazy dances on the elephant back.
HBase, crazy dances on the elephant back.
 
GCP Data Engineer cheatsheet
GCP Data Engineer cheatsheetGCP Data Engineer cheatsheet
GCP Data Engineer cheatsheet
 
Apache hadoop, hdfs and map reduce Overview
Apache hadoop, hdfs and map reduce OverviewApache hadoop, hdfs and map reduce Overview
Apache hadoop, hdfs and map reduce Overview
 
HBase introduction talk
HBase introduction talkHBase introduction talk
HBase introduction talk
 
Hadoop - Introduction to HDFS
Hadoop - Introduction to HDFSHadoop - Introduction to HDFS
Hadoop - Introduction to HDFS
 
Hbase: an introduction
Hbase: an introductionHbase: an introduction
Hbase: an introduction
 
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)
SQLSaturday #230 - Introduction to Microsoft Big Data (Part 2)
 
Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017Hoodie - DataEngConf 2017
Hoodie - DataEngConf 2017
 
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...
Apache Hadoop India Summit 2011 talk "Searching Information Inside Hadoop Pla...
 
Inside MapR's M7
Inside MapR's M7Inside MapR's M7
Inside MapR's M7
 
Training
TrainingTraining
Training
 
MapR M7: Providing an enterprise quality Apache HBase API
MapR M7: Providing an enterprise quality Apache HBase APIMapR M7: Providing an enterprise quality Apache HBase API
MapR M7: Providing an enterprise quality Apache HBase API
 
Inside MapR's M7
Inside MapR's M7Inside MapR's M7
Inside MapR's M7
 
Kudu - Fast Analytics on Fast Data
Kudu - Fast Analytics on Fast DataKudu - Fast Analytics on Fast Data
Kudu - Fast Analytics on Fast Data
 
02.28.13 WANdisco ApacheCon 2013
02.28.13 WANdisco ApacheCon 201302.28.13 WANdisco ApacheCon 2013
02.28.13 WANdisco ApacheCon 2013
 
Meethadoop
MeethadoopMeethadoop
Meethadoop
 
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
Hadoop World 2011: Hadoop Network and Compute Architecture Considerations - J...
 
Hbase 20141003
Hbase 20141003Hbase 20141003
Hbase 20141003
 
Hadoop Research
Hadoop Research Hadoop Research
Hadoop Research
 

Google Bigtable Paper Presentation

  • 1. Google Bigtable (Bigtable: A Distributed Storage System for Structured Data) Komadinovic Vanja, Vast Platform team
  • 2. Presentation overview - introduction - design - basic implementation - GFS - HDFS introduction - MapReduce introduction - implementation - HBase - Apache Bigtable solution - performances and usage case - some thoughts for discussion
  • 3. Introduction - what is bigtable? - why need for something more than SQL? - machines number increase - scaling - fault tolerance - row, column, cell - versioning - storage on distributed file system - read, write operations on row or set of rows - scanner - iterator - no need for indexing
  • 4. Design - column family basic unit of access control - column family must be created before data insertion - column labels number is unlimited - row key is arbitrary string, max. size 64Kb - every cell has copies in time - configurable
  • 6. Design - table is divided in tablets - tablets are distributed on machines - each tablet contains N rows - tablet size 100 - 200 Mb - tablet server splits tablet when it grows, only operation initiated by tablet server - tablets are stored on GFS
  • 9. GFS - HDFS introduction - distributed file system - files are divided in relatively large blocks, 64Mb - blocks of one file are stored on multiple machines - every block is replicated in configurable number of copies - fault tolerance guaranteed - Master - Slaves architecture
  • 10. GFS - HDFS introduction: overview
  • 11. MapReduce introduction - framework build on top of (GFS) HDFS - used for distributed data processing - Master - Slaves architecture - between data elements there is no dependency - everything is executed in parallel - two parts: Mapper and Reducer - Mapper reads input data and emits ( key, value pairs ) to Reducer - Data on Reducer is grouped by key - Reducer produces output data - Combiner and Partitioner - who are they, and what they want?
  • 13. Implementation - Chubby - distributed lock service, synchronization point, start point for master, tablet server and even clients - Master server, only one, guaranteed by Chubby - in case of Master tragical death tablet servers continue to serve client - Master decides who serves each tablet ( including ROOT and METATABLE ) - tablets are saved on GFS - all trafic between clients and Bigtable is done directly within client and tablet server ( Master is not required ) - client caches METATABLE, on first fault recache is done
  • 15. Implementation: oragnization - two special tables: ROOT and METADATA - ROOT never gets splitted - location of ROOT is in Chubby file - estimated ROOT size 128Mb => 2^34 tablets - METADATA contains all user tablets, row key is an encoding of the tablet’s table identifier and its end row
  • 16. Implementation - SSTable - file representation of tablet - consist of blocks - column family labels are stored side by side - index of blocks at end of file - compaction - minor - major - locality groups - caching
  • 17. HBase - Apache Bigtable solution 
  • 18. HBase - Bigtable synonyms - Bigtable ~ HBase - tablet ~ region - Master server ~ HBase master - Tablet server ~ HBase Region server - Chubby ~ Zookeeper ??? - GFS ~ HDFS - MapReduce ~ MapReduce :) - SSTable file ~ MapFile file
  • 19. HBase differences - MapFiles index is stored in separate file instead at end of file as in SSTable - Unlike Bigtable which identifies a row range by the table name and end-key, HBase identifies a row range by the table name and start-key - when the HBase master dies, the cluster will shut down
  • 20. Performance and usage case - number of 1000-byte values read/written per second - table shows the rate per tablet server Experiment 1 TS 50 TS 250 TS 500 TS random 1212 593 479 241 reads random 10811 8511 8000 6250 reads mem random 8850 3745 3425 2000 writes sequential 4425 2463 2625 2469 reads sequential 8547 3623 2451 1905 writes scans 15385 10526 9524 7843
  • 21. Performances and usage case Scaling: - performance of random reads from memory increases by almost a factor of 300 as the number of tablet server increases by a factor of 500 - significant drop in per-server throughput when going from 1 to 50 tablet servers - the random read benchmark shows the worst scaling (an increase in aggregate throughput by only a factor of 100 for a 500-fold increase in number of servers)
  • 22. Performances and usage case Real users: - Google Analytics: - It provides aggregate statistics, such as the number of unique visitors per day and the page views per URL per day - two tables - raw user clicks (~200Tb, 14% compression) and summary (~20Tb, 29% compression) - Google Earth: - Google operates a collection of services that provide users with access to high-resolution satellite imagery of the world’s surface - two tables - (~70Tb, imagery, no compression), (~500Gb, cache, served on hundreds of tablet servers
  • 23. Cooked by Vast: RecordStack Introduction: - long-term archival storage for vertical data to HDFS - optimized for later analytical processing. - stores: - listings data - transactional log data - user clicks - leads - other reporting stuff
  • 24. Cooked by Vast: RecordStack Goals: - Efficient archival of records to HDFS - conserving space - Checksuming records and tracking change - Optmization of HDFS - stored structure by periodical data compacting - Simple access from Hadoop tasks for analytical processing - Simple access via client application - Remote console - generation of reports
  • 25. Some thoughts for discussion - doing Bigtable with some number of "classic" databases? - expose of Chubby to client, is it required? - why not open source Bigtable? - run of tablet servers on data nodes - closer to data?
  • 26. Useful links - Scalability: http://scalability.rs/ - NoSQL Summer: http://nosqlsummer.org/ - Google Bigtable paper: http://tinyurl.com/37rlevv - HBase Architecture: http://tinyurl.com/3y3fhbk - HBase related blog: http://www.larsgeorge.com/ - Hadoop: The Definitive guide: http://tinyurl.com/36ponz3
  • 27. If you, maybe, want to contact me? e-mail: vanjakom@gmail.com, vanja@vast.com skype: komadinovic.vanja mobile: =381 (64) 296 03 43 twitter: vanjakom