SlideShare ist ein Scribd-Unternehmen logo
1 von 33
Downloaden Sie, um offline zu lesen
Rela%onal	
  Cloud	
  
A	
  Database-­‐as-­‐a-­‐Service	
  for	
  the	
  Cloud	
  

            Paper	
  by	
  Carlo	
  Curino	
  et	
  al.	
  @mit.edu	
  
                                       	
  
              Presenta%on	
  by	
  Antonio	
  Severien	
  
                            severien@kth.se	
  	
  
Overview	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø 	
  Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                2	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                3	
  
Rela%onal	
  Databases	
  
Ø 1970	
  by	
  Edgar	
  Codd,	
  IBM	
  research	
  San	
  Jose	
  
Ø Tables	
  
    Ø Rows	
  à	
  Tuples	
  
    Ø Columns	
  à	
  AEributes	
  
Ø Rela%onal	
  Database	
  Management	
  Systems	
  
   (RDBMS)	
  




                                                                        4	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                5	
  
Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Cloud	
  
Ø Reduce	
  management,	
  opera%onal	
  	
  
   and	
  energy	
  costs	
  
Ø Elas%city	
  and	
  scale	
  economy	
        amazon	
  RDS	
  
Ø Pay-­‐per-­‐use	
  




                                                                     6	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                7	
  
Problems	
  AEacked	
  	
  


                      Efficient	
  	
  
                    Mul%-­‐tenancy	
  




Elas%c	
  Scalability	
                  Privacy	
  




                                                       8	
  
Efficient	
  	
  
Mul%-­‐tenancy	
  




                     9	
  
Efficient	
  Mul%-­‐tenancy	
  
Ø Reduce	
  costs	
  
Ø Efficient	
  usage	
  of	
  resources	
  
Ø Maximize	
  hardware	
  u%liza%on	
  
Ø Single	
  database	
  server	
  on	
  each	
  machine	
  
Ø Maintain	
  applica%on	
  query	
  performance	
  




                                                               10	
  
Efficient	
  Mul%-­‐tenancy	
  
Ø Reduce	
  costs	
  
Ø Efficient	
  usage	
  of	
  resources	
  
Ø Maximize	
  hardware	
  u%liza%on	
  
Ø Single	
  database	
  server	
  on	
  each	
  machine?	
  
Ø Maintain	
  applica%on	
  query	
  performance	
  
	
  


                       Virtual	
  Machine	
                     11	
  
Efficient	
  Mul%-­‐tenancy	
  
Ø Problems	
  
   Ø Monitoring	
  resource	
  requirements	
  for	
  workloads	
  
   Ø Predic%ng	
  the	
  load	
  generated	
  
   Ø Assigning	
  workloads	
  to	
  physical	
  machines	
  
   Ø Migra%ng	
  workloads	
  between	
  nodes	
  
   Ø Live	
  migra*on	
  




                                                                   12	
  
Efficient	
  Mul%-­‐tenancy	
  
Ø Kairos	
  (Monitoring	
  and	
  consolida%on	
  engine)	
  
   Ø Resource	
  Monitor	
  
       Disk	
  ac%vity	
  and	
  RAM	
  requirements	
  
   Ø Combined	
  Load	
  Predictor	
  
       CPU,	
  RAM,	
  Disk	
  model	
  that	
  predicts	
  the	
  combined	
  
       resource	
  requirements	
  
   Ø Consolida%on	
  Engine	
  
       Non-­‐linear	
  op%miza%on	
  techniques	
  to…	
  
         	
  …	
  minimize	
  the	
  number	
  of	
  machines	
  needed	
  
         	
  …	
  balance	
  load	
  between	
  back-­‐end	
  machines	
  

                                                                                  13	
  
Elas%c	
  Scalability	
  




                            14	
  
Elas%c	
  Scalability	
  
Ø Workload	
  exceeds	
  single	
  machine	
  capacity	
  
	
  

  Ø Scale	
  a	
  single	
  database	
  to	
  mul%ple	
  nodes	
  
  Ø Scale-­‐out	
  by	
  query	
  processing	
  par%%oning	
  
  Ø Granular	
  placement	
  and	
  load	
  balance	
  on	
  backend	
  




                                                                        15	
  
Elas%c	
  Scalability	
  
Ø Strategy	
  well	
  suited	
  for	
  OLTP	
  and	
  Web	
  
   workloads…	
  but	
  can	
  extend	
  to	
  OLAP	
  
Ø Minimize	
  cross-­‐node	
  distributed	
  transac%ons	
  
   	
  
Ø Workload-­‐aware	
  par**oner	
  
   Ø Par%%on	
  data	
  to	
  minimize	
  mul%-­‐node	
  transac%ons	
  
   Ø Front-­‐end	
  analyses	
  execu%on	
  traces	
  represented	
  
     as	
  a	
  graph	
  

                                                                       16	
  
Graph	
  Par%%oning	
  
                                                   we=2

                                                                               id	
     name	
       age	
     salary	
  


                    id	
     name	
     age	
     salary	
  



                                                                                                   we=1



                             we=10                     id	
     name	
     age	
        salary	
  




we :	
  weight	
  of	
  edge
                                                                                                                            17	
  
Graph	
  Par%%oning	
  
                                                   we=2

                                                                               id	
     name	
       age	
     salary	
  


                    id	
     name	
     age	
     salary	
  



                                                                                                   we=1



                             we=10                     id	
     name	
     age	
        salary	
  




we :	
  weight	
  of	
  edge
                                                                                                                            18	
  
Graph	
  Par%%oning	
  

                                                           id	
     name	
       age	
     salary	
  


id	
     name	
     age	
     salary	
  




                                   id	
     name	
     age	
        salary	
  




                                                                                                        19	
  
Privacy	
  




              20	
  
Privacy	
  
Ø Adjustable	
  security	
  
   Ø Onion	
  ring	
  encryp%on	
  design	
  
       2	
  onion	
  layer	
  and	
  1	
  homomorphic	
  encryp%on	
  of	
  integer	
  
   Ø SQL	
  query	
  on	
  encrypted	
  data	
  
   Ø Security	
  level	
  dynamically	
  adap%ve	
  
       Converge	
  to	
  an	
  overall	
  security	
  level	
  




                                                                                          21	
  
Onion	
  Layers	
  of	
  Encryp%on	
  
6.	
  RND:	
  no	
  func%onality	
               5.	
  RND:	
  no	
  func%onality	
        HOM:	
  addi%on	
  
                                                                                               int	
  value	
  
4.	
  DET:	
  equality	
  selec%on	
             3.	
  OPE:	
  inequality	
  select,	
  
                                                 min,	
  max,	
  sort,	
  group-­‐by	
  
  2.	
  DET:	
  equality	
  join	
                                                                  or	
  
                                                   1.	
  OPE:	
  inequality	
  join	
  
                                                                                            String	
  search	
  
               Value	
                                         Value	
  
                                                                                             string	
  value	
  


          Strong	
  
                           RND	
  =	
  Randomized	
  Encryp%on	
  (no	
  opera%ons	
  allowed)	
  
                           DET	
  =	
  Determinis%c	
  Encryp%on	
  	
  
                           OPE	
  =	
  Order-­‐preserving	
  Encryp%on	
  
                           HOM	
  =	
  Homomorphic	
  Encryp%on	
  (opera%ons	
  over	
  encrypted	
  data)	
  
         Weak	
  
                                                                                                                   22	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                23	
  
Rela%onal	
  Cloud	
  Architecture	
  
                                                                        Client	
  Nodes	
  
                               Users	
                                                  User	
  Applica%on	
  
                                                                            JDBC-­‐client	
  (CryptoDB	
  enabled)	
  
 Trusted	
  Pla,orm	
  (Private/Secured)	
                   Privacy-­‐preserving	
                     Privacy-­‐preserving	
  
 Untrusted	
  Pla,orm	
  (Public)	
                               Queries	
                                   Results	
  

    Admin	
  Nodes	
                                   Frontend	
  Nodes	
  
                                                         Router	
        Distributed	
  Transac%onal	
  Coordina%on	
  
         Par%%oning	
  Engine	
  

           Placement	
  and	
  
          Migra%on	
  Engine	
  
                                                       Backend	
  Nodes	
                                 Backend	
  Nodes	
  
                                   Database	
  	
          CryptoDB	
                                          CryptoDB	
  
Par**ons	
                         load	
  stats	
  
                                                       Encryp%on	
  Engine	
                               Encryp%on	
  Engine	
  
Placement	
  




                                                                                                                                   24	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                25	
  
Experiments	
  




                  26	
  
Experiments	
  




      Bad	
  results?	
  
Tradeoff	
  for	
  be=er	
  privacy	
  
                                         27	
  
Experiments	
  
  Scaling	
  TPC-­‐C	
  




                           28	
  
Rela%onal	
  Cloud	
  
Ø Rela%onal	
  Databases	
  
Ø Database-­‐as-­‐a-­‐Service	
  (DBaaS)	
  
Ø Problems	
  AEacked	
  
   Ø Efficient	
  Mul%-­‐tenancy	
  
   Ø Elas%c	
  Scalability	
  
   Ø Database	
  Privacy	
  
Ø Rela%onal	
  Cloud	
  
Ø Experiments	
  
Ø Conclusion	
  

                                                29	
  
Conclusion	
  
Ø Presented	
  Rela%onal	
  Cloud	
  
Ø Efficient	
  Mul%-­‐tenancy	
  
     Ø Novel	
  resource	
  es%ma%on	
  
     Ø Non-­‐linear	
  op%miza%on-­‐based	
  consolida%on	
  technique	
  
Ø Scalability	
  
     Ø Graph-­‐based	
  par%%oning	
  
Ø Privacy	
  	
  
     Ø Adjustable	
  privacy	
  
     Ø SQL	
  queries	
  on	
  encrypted	
  data	
  
Ø DBaaS	
  is	
  a	
  viable	
  cloud	
  service	
  
	
  
                                                                              30	
  
References	
  
Ø  "Rela%onal	
  Cloud:	
  a	
  Database	
  Service	
  for	
  the	
  cloud"	
  Carlo	
  
    Curino,	
  Evan	
  Jones,	
  Raluca	
  Popa,	
  Nirmesh	
  Malviya,	
  Eugene	
  
    Wu,	
  Sam	
  Madden,	
  Har	
  Balakrishnan,	
  Nickolai	
  Zeldovich	
  
Ø  hEp://rela%onalcloud.com	
  	
  




                                                                                            32	
  
Privacy	
  
CryptoDB	
  Example	
  

                                                                                            DET-­‐encrypted	
  	
  
 Return	
  to	
  JDBC	
  client	
  decrypted	
  	
                                          cyphertext	
  
 RND	
  cyphertexts	
  


  SELECT i_price, ... FROM item WHERE i_id = N



                                                       JDBC	
  client	
  decrypts	
  	
  
                                                       DET	
  level	
  4	
  


                                                                                                                      33	
  

Weitere ähnliche Inhalte

Andere mochten auch

Relational cloud, A Database-as-a-Service for the Cloud
Relational cloud, A Database-as-a-Service for the CloudRelational cloud, A Database-as-a-Service for the Cloud
Relational cloud, A Database-as-a-Service for the CloudHossein Riasati
 
Mycobacterium
MycobacteriumMycobacterium
Mycobacteriumnava007
 
Salmonella typhi
Salmonella typhiSalmonella typhi
Salmonella typhinava007
 
Introduction to Pig_cert
Introduction to Pig_certIntroduction to Pig_cert
Introduction to Pig_certCruise Zhong
 
G structure du tour et TBA
G structure du tour et TBAG structure du tour et TBA
G structure du tour et TBALoïc Hervier
 
J pile et capacités déclenchées
J pile et capacités déclenchéesJ pile et capacités déclenchées
J pile et capacités déclenchéesLoïc Hervier
 
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mounting
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mountingSOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mounting
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mountingSOLO Swiss SA
 
Spss print screen analisis regreasi
Spss print screen analisis regreasiSpss print screen analisis regreasi
Spss print screen analisis regreasiFatimah Mustaffa
 
Mari kita sambut sang raja
Mari kita sambut sang rajaMari kita sambut sang raja
Mari kita sambut sang rajaRonal Pasaribu
 

Andere mochten auch (14)

Relational cloud, A Database-as-a-Service for the Cloud
Relational cloud, A Database-as-a-Service for the CloudRelational cloud, A Database-as-a-Service for the Cloud
Relational cloud, A Database-as-a-Service for the Cloud
 
Driegeleding
DriegeledingDriegeleding
Driegeleding
 
Mycobacterium
MycobacteriumMycobacterium
Mycobacterium
 
Salmonella typhi
Salmonella typhiSalmonella typhi
Salmonella typhi
 
Introduction to Pig_cert
Introduction to Pig_certIntroduction to Pig_cert
Introduction to Pig_cert
 
Pratico 1
Pratico 1Pratico 1
Pratico 1
 
Bretagne
BretagneBretagne
Bretagne
 
G structure du tour et TBA
G structure du tour et TBAG structure du tour et TBA
G structure du tour et TBA
 
J pile et capacités déclenchées
J pile et capacités déclenchéesJ pile et capacités déclenchées
J pile et capacités déclenchées
 
Solar water pump by durak impex pvt ltd
Solar water pump by durak impex pvt ltdSolar water pump by durak impex pvt ltd
Solar water pump by durak impex pvt ltd
 
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mounting
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mountingSOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mounting
SOLO Swiss Profitherm Atmosphere Furnace Metal Heat Treatment Fast mounting
 
Spss print screen analisis regreasi
Spss print screen analisis regreasiSpss print screen analisis regreasi
Spss print screen analisis regreasi
 
Slug
SlugSlug
Slug
 
Mari kita sambut sang raja
Mari kita sambut sang rajaMari kita sambut sang raja
Mari kita sambut sang raja
 

Ähnlich wie Relational Cloud

Oop Overview
Oop OverviewOop Overview
Oop Overviewphananhvu
 
High-Performance Graph Analysis and Modeling
High-Performance Graph Analysis and ModelingHigh-Performance Graph Analysis and Modeling
High-Performance Graph Analysis and ModelingNesreen K. Ahmed
 
Neo4j -- or why graph dbs kick ass
Neo4j -- or why graph dbs kick assNeo4j -- or why graph dbs kick ass
Neo4j -- or why graph dbs kick assEmil Eifrem
 
Services Oriented Infrastructure in a Web2.0 World
Services Oriented Infrastructure in a Web2.0 WorldServices Oriented Infrastructure in a Web2.0 World
Services Oriented Infrastructure in a Web2.0 WorldLexumo
 
API's, Freebase, and the Collaborative Semantic web
API's, Freebase, and the Collaborative Semantic webAPI's, Freebase, and the Collaborative Semantic web
API's, Freebase, and the Collaborative Semantic webDan Delany
 
Neo4j - The Benefits of Graph Databases (OSCON 2009)
Neo4j - The Benefits of Graph Databases (OSCON 2009)Neo4j - The Benefits of Graph Databases (OSCON 2009)
Neo4j - The Benefits of Graph Databases (OSCON 2009)Emil Eifrem
 
NoSQL Data Stores: Introduzione alle Basi di Dati Non Relazionali
NoSQL Data Stores: Introduzione alle Basi di Dati Non RelazionaliNoSQL Data Stores: Introduzione alle Basi di Dati Non Relazionali
NoSQL Data Stores: Introduzione alle Basi di Dati Non RelazionaliSteve Maraspin
 
Tuning Skype’s Redundancy Control Algorithm for User Satisfaction
Tuning Skype’s Redundancy Control Algorithm for User SatisfactionTuning Skype’s Redundancy Control Algorithm for User Satisfaction
Tuning Skype’s Redundancy Control Algorithm for User SatisfactionAcademia Sinica
 
HP Storage Works -Clemes Esser
HP Storage Works -Clemes EsserHP Storage Works -Clemes Esser
HP Storage Works -Clemes EsserHPDutchWorld
 
Memory Management for High-Performance Applications
Memory Management for High-Performance ApplicationsMemory Management for High-Performance Applications
Memory Management for High-Performance ApplicationsEmery Berger
 
Financial Networks VI - Correlation Networks
Financial Networks VI - Correlation NetworksFinancial Networks VI - Correlation Networks
Financial Networks VI - Correlation NetworksKimmo Soramaki
 
Rails Conf Europe 2007 Notes
Rails Conf  Europe 2007  NotesRails Conf  Europe 2007  Notes
Rails Conf Europe 2007 NotesRoss Lawley
 
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorialBuilding Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorialXavier Amatriain
 
Table29 Data Validation 95
Table29 Data Validation 95Table29 Data Validation 95
Table29 Data Validation 95Franky Lao
 
Dynamo Systems - QCon SF 2012 Presentation
Dynamo Systems - QCon SF 2012 PresentationDynamo Systems - QCon SF 2012 Presentation
Dynamo Systems - QCon SF 2012 PresentationShanley Kane
 
Linked In Lessons Learned And Growth And Scalability
Linked In Lessons Learned And Growth And ScalabilityLinked In Lessons Learned And Growth And Scalability
Linked In Lessons Learned And Growth And ScalabilityConSanFrancisco123
 
Massive MapReduce Matrix Computations & Multicore Graph Algorithms
Massive MapReduce Matrix Computations & Multicore Graph AlgorithmsMassive MapReduce Matrix Computations & Multicore Graph Algorithms
Massive MapReduce Matrix Computations & Multicore Graph AlgorithmsDavid Gleich
 
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory Course
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory CourseRuby on Rails 101 - Presentation Slides for a Five Day Introductory Course
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory Coursepeter_marklund
 

Ähnlich wie Relational Cloud (20)

Oop Overview
Oop OverviewOop Overview
Oop Overview
 
saurabh soni rac
saurabh soni racsaurabh soni rac
saurabh soni rac
 
High-Performance Graph Analysis and Modeling
High-Performance Graph Analysis and ModelingHigh-Performance Graph Analysis and Modeling
High-Performance Graph Analysis and Modeling
 
Neo4j -- or why graph dbs kick ass
Neo4j -- or why graph dbs kick assNeo4j -- or why graph dbs kick ass
Neo4j -- or why graph dbs kick ass
 
Services Oriented Infrastructure in a Web2.0 World
Services Oriented Infrastructure in a Web2.0 WorldServices Oriented Infrastructure in a Web2.0 World
Services Oriented Infrastructure in a Web2.0 World
 
API's, Freebase, and the Collaborative Semantic web
API's, Freebase, and the Collaborative Semantic webAPI's, Freebase, and the Collaborative Semantic web
API's, Freebase, and the Collaborative Semantic web
 
Neo4j - The Benefits of Graph Databases (OSCON 2009)
Neo4j - The Benefits of Graph Databases (OSCON 2009)Neo4j - The Benefits of Graph Databases (OSCON 2009)
Neo4j - The Benefits of Graph Databases (OSCON 2009)
 
NoSQL Data Stores: Introduzione alle Basi di Dati Non Relazionali
NoSQL Data Stores: Introduzione alle Basi di Dati Non RelazionaliNoSQL Data Stores: Introduzione alle Basi di Dati Non Relazionali
NoSQL Data Stores: Introduzione alle Basi di Dati Non Relazionali
 
Tuning Skype’s Redundancy Control Algorithm for User Satisfaction
Tuning Skype’s Redundancy Control Algorithm for User SatisfactionTuning Skype’s Redundancy Control Algorithm for User Satisfaction
Tuning Skype’s Redundancy Control Algorithm for User Satisfaction
 
HP Storage Works -Clemes Esser
HP Storage Works -Clemes EsserHP Storage Works -Clemes Esser
HP Storage Works -Clemes Esser
 
Data Aggregation System
Data Aggregation SystemData Aggregation System
Data Aggregation System
 
Memory Management for High-Performance Applications
Memory Management for High-Performance ApplicationsMemory Management for High-Performance Applications
Memory Management for High-Performance Applications
 
Financial Networks VI - Correlation Networks
Financial Networks VI - Correlation NetworksFinancial Networks VI - Correlation Networks
Financial Networks VI - Correlation Networks
 
Rails Conf Europe 2007 Notes
Rails Conf  Europe 2007  NotesRails Conf  Europe 2007  Notes
Rails Conf Europe 2007 Notes
 
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorialBuilding Large-scale Real-world Recommender Systems - Recsys2012 tutorial
Building Large-scale Real-world Recommender Systems - Recsys2012 tutorial
 
Table29 Data Validation 95
Table29 Data Validation 95Table29 Data Validation 95
Table29 Data Validation 95
 
Dynamo Systems - QCon SF 2012 Presentation
Dynamo Systems - QCon SF 2012 PresentationDynamo Systems - QCon SF 2012 Presentation
Dynamo Systems - QCon SF 2012 Presentation
 
Linked In Lessons Learned And Growth And Scalability
Linked In Lessons Learned And Growth And ScalabilityLinked In Lessons Learned And Growth And Scalability
Linked In Lessons Learned And Growth And Scalability
 
Massive MapReduce Matrix Computations & Multicore Graph Algorithms
Massive MapReduce Matrix Computations & Multicore Graph AlgorithmsMassive MapReduce Matrix Computations & Multicore Graph Algorithms
Massive MapReduce Matrix Computations & Multicore Graph Algorithms
 
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory Course
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory CourseRuby on Rails 101 - Presentation Slides for a Five Day Introductory Course
Ruby on Rails 101 - Presentation Slides for a Five Day Introductory Course
 

Mehr von Antonio Severien

Scalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsScalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsAntonio Severien
 
Scalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsScalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsAntonio Severien
 
On Pragmatism and Scientific Freedom
On Pragmatism and Scientific FreedomOn Pragmatism and Scientific Freedom
On Pragmatism and Scientific FreedomAntonio Severien
 
Community cloud antonioseverien
Community cloud antonioseverienCommunity cloud antonioseverien
Community cloud antonioseverienAntonio Severien
 

Mehr von Antonio Severien (6)

Scalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsScalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data Streams
 
Scalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data StreamsScalable Distributed Real-Time Clustering for Big Data Streams
Scalable Distributed Real-Time Clustering for Big Data Streams
 
NoSQL: Cassadra vs. HBase
NoSQL: Cassadra vs. HBaseNoSQL: Cassadra vs. HBase
NoSQL: Cassadra vs. HBase
 
On Pragmatism and Scientific Freedom
On Pragmatism and Scientific FreedomOn Pragmatism and Scientific Freedom
On Pragmatism and Scientific Freedom
 
Community cloud antonioseverien
Community cloud antonioseverienCommunity cloud antonioseverien
Community cloud antonioseverien
 
Soap vs rest
Soap vs restSoap vs rest
Soap vs rest
 

Kürzlich hochgeladen

Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
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.pptxHampshireHUG
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024The Digital Insurer
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEarley Information Science
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
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 textsMaria Levchenko
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsTop 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsRoshan Dwivedi
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
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 2024The Digital Insurer
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 

Kürzlich hochgeladen (20)

Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
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
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
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
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsTop 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
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
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 

Relational Cloud

  • 1. Rela%onal  Cloud   A  Database-­‐as-­‐a-­‐Service  for  the  Cloud   Paper  by  Carlo  Curino  et  al.  @mit.edu     Presenta%on  by  Antonio  Severien   severien@kth.se    
  • 2. Overview   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø   Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   2  
  • 3. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   3  
  • 4. Rela%onal  Databases   Ø 1970  by  Edgar  Codd,  IBM  research  San  Jose   Ø Tables   Ø Rows  à  Tuples   Ø Columns  à  AEributes   Ø Rela%onal  Database  Management  Systems   (RDBMS)   4  
  • 5. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   5  
  • 6. Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Cloud   Ø Reduce  management,  opera%onal     and  energy  costs   Ø Elas%city  and  scale  economy   amazon  RDS   Ø Pay-­‐per-­‐use   6  
  • 7. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   7  
  • 8. Problems  AEacked     Efficient     Mul%-­‐tenancy   Elas%c  Scalability   Privacy   8  
  • 10. Efficient  Mul%-­‐tenancy   Ø Reduce  costs   Ø Efficient  usage  of  resources   Ø Maximize  hardware  u%liza%on   Ø Single  database  server  on  each  machine   Ø Maintain  applica%on  query  performance   10  
  • 11. Efficient  Mul%-­‐tenancy   Ø Reduce  costs   Ø Efficient  usage  of  resources   Ø Maximize  hardware  u%liza%on   Ø Single  database  server  on  each  machine?   Ø Maintain  applica%on  query  performance     Virtual  Machine   11  
  • 12. Efficient  Mul%-­‐tenancy   Ø Problems   Ø Monitoring  resource  requirements  for  workloads   Ø Predic%ng  the  load  generated   Ø Assigning  workloads  to  physical  machines   Ø Migra%ng  workloads  between  nodes   Ø Live  migra*on   12  
  • 13. Efficient  Mul%-­‐tenancy   Ø Kairos  (Monitoring  and  consolida%on  engine)   Ø Resource  Monitor   Disk  ac%vity  and  RAM  requirements   Ø Combined  Load  Predictor   CPU,  RAM,  Disk  model  that  predicts  the  combined   resource  requirements   Ø Consolida%on  Engine   Non-­‐linear  op%miza%on  techniques  to…    …  minimize  the  number  of  machines  needed    …  balance  load  between  back-­‐end  machines   13  
  • 15. Elas%c  Scalability   Ø Workload  exceeds  single  machine  capacity     Ø Scale  a  single  database  to  mul%ple  nodes   Ø Scale-­‐out  by  query  processing  par%%oning   Ø Granular  placement  and  load  balance  on  backend   15  
  • 16. Elas%c  Scalability   Ø Strategy  well  suited  for  OLTP  and  Web   workloads…  but  can  extend  to  OLAP   Ø Minimize  cross-­‐node  distributed  transac%ons     Ø Workload-­‐aware  par**oner   Ø Par%%on  data  to  minimize  mul%-­‐node  transac%ons   Ø Front-­‐end  analyses  execu%on  traces  represented   as  a  graph   16  
  • 17. Graph  Par%%oning   we=2 id   name   age   salary   id   name   age   salary   we=1 we=10 id   name   age   salary   we :  weight  of  edge 17  
  • 18. Graph  Par%%oning   we=2 id   name   age   salary   id   name   age   salary   we=1 we=10 id   name   age   salary   we :  weight  of  edge 18  
  • 19. Graph  Par%%oning   id   name   age   salary   id   name   age   salary   id   name   age   salary   19  
  • 20. Privacy   20  
  • 21. Privacy   Ø Adjustable  security   Ø Onion  ring  encryp%on  design   2  onion  layer  and  1  homomorphic  encryp%on  of  integer   Ø SQL  query  on  encrypted  data   Ø Security  level  dynamically  adap%ve   Converge  to  an  overall  security  level   21  
  • 22. Onion  Layers  of  Encryp%on   6.  RND:  no  func%onality   5.  RND:  no  func%onality   HOM:  addi%on   int  value   4.  DET:  equality  selec%on   3.  OPE:  inequality  select,   min,  max,  sort,  group-­‐by   2.  DET:  equality  join   or   1.  OPE:  inequality  join   String  search   Value   Value   string  value   Strong   RND  =  Randomized  Encryp%on  (no  opera%ons  allowed)   DET  =  Determinis%c  Encryp%on     OPE  =  Order-­‐preserving  Encryp%on   HOM  =  Homomorphic  Encryp%on  (opera%ons  over  encrypted  data)   Weak   22  
  • 23. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   23  
  • 24. Rela%onal  Cloud  Architecture   Client  Nodes   Users   User  Applica%on   JDBC-­‐client  (CryptoDB  enabled)   Trusted  Pla,orm  (Private/Secured)   Privacy-­‐preserving   Privacy-­‐preserving   Untrusted  Pla,orm  (Public)   Queries   Results   Admin  Nodes   Frontend  Nodes   Router   Distributed  Transac%onal  Coordina%on   Par%%oning  Engine   Placement  and   Migra%on  Engine   Backend  Nodes   Backend  Nodes   Database     CryptoDB   CryptoDB   Par**ons   load  stats   Encryp%on  Engine   Encryp%on  Engine   Placement   24  
  • 25. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   25  
  • 26. Experiments   26  
  • 27. Experiments   Bad  results?   Tradeoff  for  be=er  privacy   27  
  • 28. Experiments   Scaling  TPC-­‐C   28  
  • 29. Rela%onal  Cloud   Ø Rela%onal  Databases   Ø Database-­‐as-­‐a-­‐Service  (DBaaS)   Ø Problems  AEacked   Ø Efficient  Mul%-­‐tenancy   Ø Elas%c  Scalability   Ø Database  Privacy   Ø Rela%onal  Cloud   Ø Experiments   Ø Conclusion   29  
  • 30. Conclusion   Ø Presented  Rela%onal  Cloud   Ø Efficient  Mul%-­‐tenancy   Ø Novel  resource  es%ma%on   Ø Non-­‐linear  op%miza%on-­‐based  consolida%on  technique   Ø Scalability   Ø Graph-­‐based  par%%oning   Ø Privacy     Ø Adjustable  privacy   Ø SQL  queries  on  encrypted  data   Ø DBaaS  is  a  viable  cloud  service     30  
  • 31.
  • 32. References   Ø  "Rela%onal  Cloud:  a  Database  Service  for  the  cloud"  Carlo   Curino,  Evan  Jones,  Raluca  Popa,  Nirmesh  Malviya,  Eugene   Wu,  Sam  Madden,  Har  Balakrishnan,  Nickolai  Zeldovich   Ø  hEp://rela%onalcloud.com     32  
  • 33. Privacy   CryptoDB  Example   DET-­‐encrypted     Return  to  JDBC  client  decrypted     cyphertext   RND  cyphertexts   SELECT i_price, ... FROM item WHERE i_id = N JDBC  client  decrypts     DET  level  4   33  

Hinweis der Redaktion

  1. Talk about the importance of relational databases and their legacy
  2. Talk about the market and the viability of relational databases as a service in the cloud
  3. Make this slide better
  4. Make this slide better
  5. Challenge: workload exceeds capacity of single machine
  6. - THE WAY TO SCALE THE WORKLOADS is to MINIMIZE # of MULTI-NODE TRANSACTIONS… why? OVERHEAD ON HOLDING LOCKS on the BACKEND
  7. Detail how the provacy works and follow to exemplify on the next sliideKnow well homomrphismUses symetric encryption
  8. Comparison between consolidated DBs in one machine versus DBs on Virtual Machines.Explained the difference between UNIFORM and SKEWED: uniform load and skewed (50% of the requests goes to one of the 20 DBs)Consolidated 20 databases to one physical machine
  9. Explain what is TPC-C (benchmarks for databases…. Etc)