SlideShare a Scribd company logo
1 of 20
Dremel
            Interactive Analysis
           of Web-Scale Datasets
Sergey Melnik, Andrey Gubarev, Jing Jing Long, Geoffrey Romer, Shiva
               Shivakumar, Matt Tolton, Theo Vassilakis




                 Presented by Maria Stylianou
                     marsty5@gmail.com
                        November 8th, 2012

                  KTH – Royal Institute of Technology
Outline
●   Motivation

●   Dremel – basic information
●   Dremel's Key Aspects
    –   Columnar Format
    –   Query Execution


●   Evaluation & Conclusions     2
Motivation

    Data                  Big Data
●   Web-scale Datasets → more frequent
●   Large-scale Data Analysis → essential!


                  NOT
                  FAST
            Speed Matters!                   3
Dremel to the rescue!
●   Interactive ad-hoc query system
    Scalable     Fault tolerant   Fast

                                      Access data
                                       'in place'
●   Analysis on in situ nested data

       Non
    relational
                                                4
MapReduce or Dremel
      or both



        ?

                      5
Key Aspects of Dremel
●   Storage Format
    –   Columnar storage representation for nested
        data

●   Query Language & Execution
    –   SQL & Multi-level serving tree



                                                     6
Storage Format
Columnar Storage Representation




                                  7
Data Model
     ●   Based on strongly-typed nested records
                                            schema




Repetition
  Level
          Definition
            Level            records
Query Language & Execution
          SQL & Multi-level Serving Tree
  Tablet
 Contains
N rows from
 the table




                                           9
Query Execution
                 Query Dispatcher

●   Schedules queries based on their priorities
●   Balances the load
                                           Servers
●   Provides fault tolerance               running
    –   Handles stragglers                  slow
    –   Tablets are three-way replicated


                                                     10
Experiments
Environment




              11
Experiments
Local Disk - Performance




                           12
Experiments
                 MapReduce and Dremel

Counts the average number
 of terms in a specific field

                                          3000 workers
                        hours
                                minutes

                                            seconds




                                                         13
Experiments
Impact of Stragglers




                       14
Experiments
                          Scalability

 Selects top-20 adverts and
Their number of occurrences
            In T4




                                        15
What's happening today?
●   Google BigQuery
    –   Web Service [pay-per-query]


●   Open Dremel → Apache Drill
    –   Open Source Implementation
        of Google BigQuery
    –   Flexibility: broader range of query languages

                                                        16
MapReduce or Dremel
                  or both
                                       ?
                      MR           Dremel
Data Processing      Record        Column
                     Oriented      Oriented
In-situ Processing     No            Yes!

Size of Queries       Large     Small/Medium


      MapReduce AND Dremel                    17
Conclusions
Multi-level             Columnar
Execution                 Data
  trees                  Layout




      Scalable & Efficient
      MapReduce benefits
      Near-linear scalability

                                   18
Dremel
            Interactive Analysis
           of Web-Scale Datasets
Sergey Melnik, Andrey Gubarev, Jing Jing Long, Geoffrey Romer, Shiva
               Shivakumar, Matt Tolton, Theo Vassilakis




                 Presented by Maria Stylianou
                     marsty5@gmail.com
                        November 8th, 2012

                  KTH – Royal Institute of Technology
References
●   S. Melnik et al. Dremel: Interactive Analysis of Web-
    Scale Datasets. PVLDB, 3(1):330–339, 2010
●
    G. Czajkowski. Sorting 1PB with MapReduce.
    http://googleblog.blogspot.se/2008/11/sorting-1pb-with-mapreduce.html

●   Apache Drill, http://wiki.apache.org/incubator/DrillProposal
●   Google BigQuery, https://developers.google.com/bigquery/

More Related Content

What's hot

Benchmarking Oracle I/O Performance with Orion by Alex Gorbachev
Benchmarking Oracle I/O Performance with Orion by Alex GorbachevBenchmarking Oracle I/O Performance with Orion by Alex Gorbachev
Benchmarking Oracle I/O Performance with Orion by Alex Gorbachev
Alex Gorbachev
 
The columnar roadmap: Apache Parquet and Apache Arrow
The columnar roadmap: Apache Parquet and Apache ArrowThe columnar roadmap: Apache Parquet and Apache Arrow
The columnar roadmap: Apache Parquet and Apache Arrow
DataWorks Summit
 
ORC File & Vectorization - Improving Hive Data Storage and Query Performance
ORC File & Vectorization - Improving Hive Data Storage and Query PerformanceORC File & Vectorization - Improving Hive Data Storage and Query Performance
ORC File & Vectorization - Improving Hive Data Storage and Query Performance
DataWorks Summit
 

What's hot (20)

Benchmarking Oracle I/O Performance with Orion by Alex Gorbachev
Benchmarking Oracle I/O Performance with Orion by Alex GorbachevBenchmarking Oracle I/O Performance with Orion by Alex Gorbachev
Benchmarking Oracle I/O Performance with Orion by Alex Gorbachev
 
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
 
Apache flink
Apache flinkApache flink
Apache flink
 
Linux tuning to improve PostgreSQL performance
Linux tuning to improve PostgreSQL performanceLinux tuning to improve PostgreSQL performance
Linux tuning to improve PostgreSQL performance
 
Deep Dive with Spark Streaming - Tathagata Das - Spark Meetup 2013-06-17
Deep Dive with Spark Streaming - Tathagata  Das - Spark Meetup 2013-06-17Deep Dive with Spark Streaming - Tathagata  Das - Spark Meetup 2013-06-17
Deep Dive with Spark Streaming - Tathagata Das - Spark Meetup 2013-06-17
 
Ceph scale testing with 10 Billion Objects
Ceph scale testing with 10 Billion ObjectsCeph scale testing with 10 Billion Objects
Ceph scale testing with 10 Billion Objects
 
Druid
DruidDruid
Druid
 
Accelerate Your Analytic Queries with Amazon Aurora Parallel Query (DAT362) -...
Accelerate Your Analytic Queries with Amazon Aurora Parallel Query (DAT362) -...Accelerate Your Analytic Queries with Amazon Aurora Parallel Query (DAT362) -...
Accelerate Your Analytic Queries with Amazon Aurora Parallel Query (DAT362) -...
 
The columnar roadmap: Apache Parquet and Apache Arrow
The columnar roadmap: Apache Parquet and Apache ArrowThe columnar roadmap: Apache Parquet and Apache Arrow
The columnar roadmap: Apache Parquet and Apache Arrow
 
ProxySQL - High Performance and HA Proxy for MySQL
ProxySQL - High Performance and HA Proxy for MySQLProxySQL - High Performance and HA Proxy for MySQL
ProxySQL - High Performance and HA Proxy for MySQL
 
Introducing BinarySortedMultiMap - A new Flink state primitive to boost your ...
Introducing BinarySortedMultiMap - A new Flink state primitive to boost your ...Introducing BinarySortedMultiMap - A new Flink state primitive to boost your ...
Introducing BinarySortedMultiMap - A new Flink state primitive to boost your ...
 
Polyglot persistence @ netflix (CDE Meetup)
Polyglot persistence @ netflix (CDE Meetup) Polyglot persistence @ netflix (CDE Meetup)
Polyglot persistence @ netflix (CDE Meetup)
 
Under the Hood of a Shard-per-Core Database Architecture
Under the Hood of a Shard-per-Core Database ArchitectureUnder the Hood of a Shard-per-Core Database Architecture
Under the Hood of a Shard-per-Core Database Architecture
 
Zeus: Uber’s Highly Scalable and Distributed Shuffle as a Service
Zeus: Uber’s Highly Scalable and Distributed Shuffle as a ServiceZeus: Uber’s Highly Scalable and Distributed Shuffle as a Service
Zeus: Uber’s Highly Scalable and Distributed Shuffle as a Service
 
Building a Scalable Web Crawler with Hadoop
Building a Scalable Web Crawler with HadoopBuilding a Scalable Web Crawler with Hadoop
Building a Scalable Web Crawler with Hadoop
 
Spark shuffle introduction
Spark shuffle introductionSpark shuffle introduction
Spark shuffle introduction
 
ORC File & Vectorization - Improving Hive Data Storage and Query Performance
ORC File & Vectorization - Improving Hive Data Storage and Query PerformanceORC File & Vectorization - Improving Hive Data Storage and Query Performance
ORC File & Vectorization - Improving Hive Data Storage and Query Performance
 
Introduction to Spark Streaming
Introduction to Spark StreamingIntroduction to Spark Streaming
Introduction to Spark Streaming
 
Introducing Apache Airflow and how we are using it
Introducing Apache Airflow and how we are using itIntroducing Apache Airflow and how we are using it
Introducing Apache Airflow and how we are using it
 
Apache Iceberg - A Table Format for Hige Analytic Datasets
Apache Iceberg - A Table Format for Hige Analytic DatasetsApache Iceberg - A Table Format for Hige Analytic Datasets
Apache Iceberg - A Table Format for Hige Analytic Datasets
 

Similar to Google's Dremel

The elephantintheroom bigdataanalyticsinthecloud
The elephantintheroom bigdataanalyticsinthecloudThe elephantintheroom bigdataanalyticsinthecloud
The elephantintheroom bigdataanalyticsinthecloud
Khazret Sapenov
 
NO SQL: What, Why, How
NO SQL: What, Why, HowNO SQL: What, Why, How
NO SQL: What, Why, How
Igor Moochnick
 

Similar to Google's Dremel (20)

Challenges in Large Scale Machine Learning
Challenges in Large Scale  Machine LearningChallenges in Large Scale  Machine Learning
Challenges in Large Scale Machine Learning
 
Dremel
DremelDremel
Dremel
 
Next generation analytics with yarn, spark and graph lab
Next generation analytics with yarn, spark and graph labNext generation analytics with yarn, spark and graph lab
Next generation analytics with yarn, spark and graph lab
 
Spark
SparkSpark
Spark
 
Spark Summit EU talk by Ahsan Javed Awan
Spark Summit EU talk by Ahsan Javed AwanSpark Summit EU talk by Ahsan Javed Awan
Spark Summit EU talk by Ahsan Javed Awan
 
MapR & Skytree:
MapR & Skytree: MapR & Skytree:
MapR & Skytree:
 
Distributed computing abstractions_data_science_6_june_2016_ver_0.4
Distributed computing abstractions_data_science_6_june_2016_ver_0.4Distributed computing abstractions_data_science_6_june_2016_ver_0.4
Distributed computing abstractions_data_science_6_june_2016_ver_0.4
 
The elephantintheroom bigdataanalyticsinthecloud
The elephantintheroom bigdataanalyticsinthecloudThe elephantintheroom bigdataanalyticsinthecloud
The elephantintheroom bigdataanalyticsinthecloud
 
Drill lightning-london-big-data-10-01-2012
Drill lightning-london-big-data-10-01-2012Drill lightning-london-big-data-10-01-2012
Drill lightning-london-big-data-10-01-2012
 
Is Spark the right choice for data analysis ?
Is Spark the right choice for data analysis ?Is Spark the right choice for data analysis ?
Is Spark the right choice for data analysis ?
 
Secrets of Spark's success - Deenar Toraskar, Think Reactive
Secrets of Spark's success - Deenar Toraskar, Think Reactive Secrets of Spark's success - Deenar Toraskar, Think Reactive
Secrets of Spark's success - Deenar Toraskar, Think Reactive
 
AI on Greenplum Using
 Apache MADlib and MADlib Flow - Greenplum Summit 2019
AI on Greenplum Using
 Apache MADlib and MADlib Flow - Greenplum Summit 2019AI on Greenplum Using
 Apache MADlib and MADlib Flow - Greenplum Summit 2019
AI on Greenplum Using
 Apache MADlib and MADlib Flow - Greenplum Summit 2019
 
NO SQL: What, Why, How
NO SQL: What, Why, HowNO SQL: What, Why, How
NO SQL: What, Why, How
 
Big Data Analytics with Storm, Spark and GraphLab
Big Data Analytics with Storm, Spark and GraphLabBig Data Analytics with Storm, Spark and GraphLab
Big Data Analytics with Storm, Spark and GraphLab
 
Notes on data-intensive processing with Hadoop Mapreduce
Notes on data-intensive processing with Hadoop MapreduceNotes on data-intensive processing with Hadoop Mapreduce
Notes on data-intensive processing with Hadoop Mapreduce
 
Hadoop.mapreduce
Hadoop.mapreduceHadoop.mapreduce
Hadoop.mapreduce
 
Productionizing Deep Learning From the Ground Up
Productionizing Deep Learning From the Ground UpProductionizing Deep Learning From the Ground Up
Productionizing Deep Learning From the Ground Up
 
IaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and ExperienceIaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and Experience
 
Dremel Paper Review
Dremel Paper ReviewDremel Paper Review
Dremel Paper Review
 
Introduction to map reduce
Introduction to map reduceIntroduction to map reduce
Introduction to map reduce
 

More from Maria Stylianou

More from Maria Stylianou (16)

SPARJA: a Distributed Social Graph Partitioning and Replication Middleware
SPARJA: a Distributed Social Graph Partitioning and Replication MiddlewareSPARJA: a Distributed Social Graph Partitioning and Replication Middleware
SPARJA: a Distributed Social Graph Partitioning and Replication Middleware
 
Quantum Cryptography and Possible Attacks
Quantum Cryptography and Possible AttacksQuantum Cryptography and Possible Attacks
Quantum Cryptography and Possible Attacks
 
Scaling Online Social Networks (OSNs)
Scaling Online Social Networks (OSNs)Scaling Online Social Networks (OSNs)
Scaling Online Social Networks (OSNs)
 
Erlang in 10 minutes
Erlang in 10 minutesErlang in 10 minutes
Erlang in 10 minutes
 
Pregel - Paper Review
Pregel - Paper ReviewPregel - Paper Review
Pregel - Paper Review
 
Green Optical Networks with Signal Quality Guarantee
Green Optical Networks with Signal Quality Guarantee Green Optical Networks with Signal Quality Guarantee
Green Optical Networks with Signal Quality Guarantee
 
Cano projectGreen Optical Networks with Signal Quality Guarantee
Cano projectGreen Optical Networks with Signal Quality Guarantee Cano projectGreen Optical Networks with Signal Quality Guarantee
Cano projectGreen Optical Networks with Signal Quality Guarantee
 
A Survey on Large-Scale Decentralized Storage Systems to be used by Volunteer...
A Survey on Large-Scale Decentralized Storage Systems to be used by Volunteer...A Survey on Large-Scale Decentralized Storage Systems to be used by Volunteer...
A Survey on Large-Scale Decentralized Storage Systems to be used by Volunteer...
 
Performance Analysis of multithreaded applications based on Hardware Simulati...
Performance Analysis of multithreaded applications based on Hardware Simulati...Performance Analysis of multithreaded applications based on Hardware Simulati...
Performance Analysis of multithreaded applications based on Hardware Simulati...
 
Automatic Energy-based Scheduling
Automatic Energy-based SchedulingAutomatic Energy-based Scheduling
Automatic Energy-based Scheduling
 
Intelligent Placement of Datacenters for Internet Services
Intelligent Placement of Datacenters for Internet ServicesIntelligent Placement of Datacenters for Internet Services
Intelligent Placement of Datacenters for Internet Services
 
Instrumenting the MG applicaiton of NAS Parallel Benchmark
Instrumenting the MG applicaiton of NAS Parallel BenchmarkInstrumenting the MG applicaiton of NAS Parallel Benchmark
Instrumenting the MG applicaiton of NAS Parallel Benchmark
 
Low-Latency Multi-Writer Atomic Registers
Low-Latency Multi-Writer Atomic RegistersLow-Latency Multi-Writer Atomic Registers
Low-Latency Multi-Writer Atomic Registers
 
How Companies Learn Your Secrets
How Companies Learn Your SecretsHow Companies Learn Your Secrets
How Companies Learn Your Secrets
 
EEDC - Why use of REST for Web Services
EEDC - Why use of REST for Web Services EEDC - Why use of REST for Web Services
EEDC - Why use of REST for Web Services
 
EEDC - Distributed Systems
EEDC - Distributed SystemsEEDC - Distributed Systems
EEDC - Distributed Systems
 

Recently uploaded

Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
Joaquim Jorge
 

Recently uploaded (20)

How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Tech Trends Report 2024 Future Today Institute.pdf
Tech Trends Report 2024 Future Today Institute.pdfTech Trends Report 2024 Future Today Institute.pdf
Tech Trends Report 2024 Future Today Institute.pdf
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 

Google's Dremel

  • 1. Dremel Interactive Analysis of Web-Scale Datasets Sergey Melnik, Andrey Gubarev, Jing Jing Long, Geoffrey Romer, Shiva Shivakumar, Matt Tolton, Theo Vassilakis Presented by Maria Stylianou marsty5@gmail.com November 8th, 2012 KTH – Royal Institute of Technology
  • 2. Outline ● Motivation ● Dremel – basic information ● Dremel's Key Aspects – Columnar Format – Query Execution ● Evaluation & Conclusions 2
  • 3. Motivation Data Big Data ● Web-scale Datasets → more frequent ● Large-scale Data Analysis → essential! NOT FAST Speed Matters! 3
  • 4. Dremel to the rescue! ● Interactive ad-hoc query system Scalable Fault tolerant Fast Access data 'in place' ● Analysis on in situ nested data Non relational 4
  • 5. MapReduce or Dremel or both ? 5
  • 6. Key Aspects of Dremel ● Storage Format – Columnar storage representation for nested data ● Query Language & Execution – SQL & Multi-level serving tree 6
  • 8. Data Model ● Based on strongly-typed nested records schema Repetition Level Definition Level records
  • 9. Query Language & Execution SQL & Multi-level Serving Tree Tablet Contains N rows from the table 9
  • 10. Query Execution Query Dispatcher ● Schedules queries based on their priorities ● Balances the load Servers ● Provides fault tolerance running – Handles stragglers slow – Tablets are three-way replicated 10
  • 12. Experiments Local Disk - Performance 12
  • 13. Experiments MapReduce and Dremel Counts the average number of terms in a specific field 3000 workers hours minutes seconds 13
  • 15. Experiments Scalability Selects top-20 adverts and Their number of occurrences In T4 15
  • 16. What's happening today? ● Google BigQuery – Web Service [pay-per-query] ● Open Dremel → Apache Drill – Open Source Implementation of Google BigQuery – Flexibility: broader range of query languages 16
  • 17. MapReduce or Dremel or both ? MR Dremel Data Processing Record Column Oriented Oriented In-situ Processing No Yes! Size of Queries Large Small/Medium MapReduce AND Dremel 17
  • 18. Conclusions Multi-level Columnar Execution Data trees Layout Scalable & Efficient MapReduce benefits Near-linear scalability 18
  • 19. Dremel Interactive Analysis of Web-Scale Datasets Sergey Melnik, Andrey Gubarev, Jing Jing Long, Geoffrey Romer, Shiva Shivakumar, Matt Tolton, Theo Vassilakis Presented by Maria Stylianou marsty5@gmail.com November 8th, 2012 KTH – Royal Institute of Technology
  • 20. References ● S. Melnik et al. Dremel: Interactive Analysis of Web- Scale Datasets. PVLDB, 3(1):330–339, 2010 ● G. Czajkowski. Sorting 1PB with MapReduce. http://googleblog.blogspot.se/2008/11/sorting-1pb-with-mapreduce.html ● Apache Drill, http://wiki.apache.org/incubator/DrillProposal ● Google BigQuery, https://developers.google.com/bigquery/

Editor's Notes

  1. - Hello everybody. I will present Dremel, a tool developed in Google, - It is being used at Google since 2006 - But the paper was published in 2010
  2. Let's briefly see the outline of the presentation. I will start with the motivation of the authors do develop Dremel Then I will explain what is Dremel and which are the key aspects that make Dremel to be novel I will continue with with the evaluation, showing some of the experiments the authors contacted to support their idea. And of course I will close my presentation with some observations and conclusions
  3. Their motivation begun with the observation that data are becoming BIG Web-scale Datasets are becoming more frequent Performing Data analysis at scale is essential As you may know Pig and Hive can perform ad-hoc queries into web-scale datasets BUT they are NOT FAST This is because they translate queries into MapReduce jobs, which makes the execution slower The thing is... Speed Matters! So, what the authors wanted to do is to develop a tool that would execute ad-hoc queries in large-scale datasets rapidly
  4. Dremel is an interactive ad-hoc query system It is scalable, fault tolerant and Fast It performs analysis on in situ nested data In situ means: it accesses data 'in place' Which means, it executes the computation in the place that the data are stored. In this case, BigTable of Google File System is used, so it does not take the data and take them into the tool, but the tool operates inside the dataset. Nested data, non relational data An Interoperation between the Dremel (query processor) and other data management tools
  5. There is a clear comparison between Dremel and MapReduce on the paper. For now, I'll leave this blank and come back when it's time :)
  6. So! Let's start with the main characteristics of Dremel! What makes Dremel so special is the use & combination of: Columnar storage format of the data Multi-level serving tree for query execution
  7. So far, data were stored as records. Let's imagine we have a database with information for each EMDC student. Each record (raw) consists of name, age, nationality and other data of the student What's done so far, was to store all information for each student gathered in a record Google, then, comes with this novel idea to store data in columns. That means, all names are stored together, all ages together, nationalities, etc. So if Sarunas wants to see the ages of his students, he can just query the age and only the column age will be read. That way, they improved retrieval efficiency → less data have to be read
  8. Dremel uses an SQL-like language And for executing queries, it uses multi-level serving trees We have many servers, and one of them is the root server. The root server receives the query from the client and: – determines all tablets of the table related to the query – rewrites the query and sends it to the next level servers → How it rewrites it? In a way that each intermediate server will be assigned some of the tablets – the intermediate servers do the same – rewrite the query they received – and send it to the next level. – when queries reach the leaf servers, they scan the tablets & execute the queries in parallel – by accessing the common storage (Google File System) and send the result back to their parent – each intermediate server receives more than one values and aggregates the results into one. – this is done in all servers, till we reach the root server. Each servers has an internal execution tree which includes evaluation of aggregation functions → for optimization purposes
  9. Dremel is a multi-user system → several queries are executed at the same time. Fault-tolerance and straggler detection also play positively in to execution time 3-way replication When a leaf server can not access a tablet replica, it falls over to another replica. Parameter specifies the minimum percentage of tablets that must be scanned before returning a result. → setting up this parameter low, it can speed up the execution significantly. Dremel allows for "99.9%" type results, that reflect almost all, but not quite all, of the data.
  10. Now let's move on to the experiments they conducted. I only present the most important ones – according to me :) The authors used 5 different tables in 2 different datasets, each one with different number of records, starting from 4 billion, up to more than 1 trillion. The compressed data vary from 13TB to 105 TB While The number of fields begin with 30 and reaches 1200
  11. In the first experiment they
  12. A team of Israeli engineers is building a clone they called OpenDremel, though one of these developers, David Gruzman, tells us that coding is only just beginning again after a long hiatus. Google now offers a Dremel web service it calls BigQuery. You can use the platform via an online API, or application programming interface. Basically, you upload your data to Google, and it lets you run queries on its internal infrastructure.
  13. There is a clear comparison between Dremel and MapReduce on the paper. Their intention is not to replace MapReduce But to complement MapReduce
  14. - Hello everybody. I will present Dremel, a tool developed in Google, - It is being used at Google since 2006 - But the paper was published in 2010