Solution Spotlight Presents
Supermicro © 2009



                    Cloudera Ready - Infrastructure for Hadoop




                                                                         SuperBlade®



                GPU Solutions




                                               Storage

                        Universal I/O
                                                                        Twin Architecture
                                                         Double-Sided



                                  Datacenter
                                  Optimized




                                www.supermicro.com/hadoop
Building Block Platforms for Hadoop


l  1U 815 Compute Hadoop Nodes
     l  Dual Quad Core Xeon
     l  4, 3.5 HDD spindles per node
     l  Hot Swap HDD, PSUs
     l  Up to 8 TB per 1U
     l  Common Components
     l  On Board IPMI Remote Management
     l  93% Efficiency PMBus PSUs

l  2UTwin 827 Hadoop Nodes
     l  Dual Quad Core Xeon
     l  6, 3.5 HDD spindles per node
     l  2 Nodes in 2U = 6 HDD per U
     l  On Board IPMI Remote Management
     l  Hot Swap Nodes, HDD, PSUs
     l  94% Efficiency PMBus PSUs
Building Block Platforms for Hadoop

l  2U 825 Balanced Hadoop Nodes
       l  Dual Quad Core Xeon
       l  8, 3.5 HDD
       l  On Board IPMI Remote Management
       l  Common Components
       l  Hot Swap HDD, PSUs
       l  93% Efficiency PMBus PSUs

l  2U 826 Flex Hadoop Nodes
       l  Dual Quad Core Xeon
       l  12, 3.5 HDD
       l  On Board IPMI Remote Management
       l  Hot Swap HDD, PSUs
       l  Up to 24TB per 2U
       l  Common Components
       l  93% Efficiency PMBus PSUs
Optimized Hadoop Nodes - Cloudera Ready
Name Node / Job Tracker:


                  Name Node – 48GB
                  PIO-HNU0-15056248-HADP



Data Nodes:           1TB SATA Drives                  2TB SATA Drives


                  Data Node – 4 x 1TB 24GB        Data Node – 4 x 2TB 24GB
                  PIO-HDU0-15156224-HADP   1U     PIO-HDU0-15256224-HADP    1U




                  Data Node – 6 x 1TB 24GB        Data Node – 6 x 2TB 24GB
                  PIO-HDT0-27156224-HADP 2UTwin   PIO-HDT0-27256224-HADP 2UTwin




                  Data Node – 8 x 1TB 24GB        Data Node – 8 x 2TB 24GB
                  PIO-HDU0-25156224-HADP    2U    PIO-HDU0-25256224-HADP    2U




                  Data Node – 12 x 1TB 24GB       Data Node – 12 x 2TB 24GB
                  PIO-HDU0-26156224-HADP     2U   PIO-HDU0-26256224-HADP    2U
Turn Key Pilot Cluster Kits

 Pilot Kit - Configuration:


 Ethernet Switch:       1G Ethernet Switch
   (10G Uplink
    Capability)
                        SSE-G24-TG4
                        SSE-G48-TG4
                                      1U 24 port
                                      1U 48 port        10 Node Pilot Cluster
                        Name Node – 48GB                • 48 TB
   Name Node:           PIO-HNU0-15056248-HADP

   Secondary            Name Node – 48GB
                                                        • 3.5     SATA
Name/Job Tracker:       PIO-HNU0-15056248-HADP
                                                        • 64 Cores
  2 Data Nodes:         Data Node – 6 x 1TB 24GB
                        PIO-HDT0-27156224-HADP 2UTwin   • 48 Spindles
  2 Data Nodes:         Data Node – 6 x 1TB 24GB
                        PIO-HDT0-27156224-HADP 2UTwin   • 288 GB Memory
  2 Data Nodes:         Data Node – 6 x 1TB 24GB
                        PIO-HDT0-27156224-HADP 2UTwin   • Occupy 11U
  2 Data Nodes:         Data Node – 6 x 1TB 24GB
                        PIO-HDT0-27156224-HADP 2UTwin
SUPERMICRO – Cloudera Certified Technology
Hadoop Series Naming Convention


                   PIO-HDU0-26156224-HADP

H = Hadoop
                                            24 = GB Memory
  D = Data Node                             48 = GB Memory
  N = Name Node

    U = X8DTU-F                            562 = Xeon E5620
    T = X8DTT-HF
      0 = Intel
      1 = AMD                               0 = OS HDD x 4
                                            1 = 1TB HDD
       15 = 815 - 2U - 4 HDD                2 = 2TB HDD
       25 = 825 - 2U - 8 HDD
       26 = 826 - 2U - 12 HDD
       27 = 827 - 2UTwin - 2x6 HDD
Hadoop Series Naming Convention




Model/Part #             Description / Functionality
PIO-HNU0-15056248-HADP   1U Hadoop 815 Name E5620 48GB MEM

PIO-HDU0-15156224-HADP   1U Hadoop 815 4 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
PIO-HDU0-15256224-HADP   1U Hadoop 815 4 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM

PIO-HDT0-27156224-HADP   2U Hadoop 827 6 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
PIO-HDT0-27256224-HADP   2U Hadoop 827 6 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM

PIO-HDU0-25156224-HADP   2U Hadoop 825 8 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
PIO-HDU0-25256224-HADP   2U Hadoop 825 8 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM

PIO-HDU0-26156224-HADP   2U Hadoop 826 12 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
PIO-HDU0-26256224-HADP   2U Hadoop 826 12 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
Turn Key Hadoop Cluster Racks




+               +             +           =      ROB
                                              Return on Byte$




    Optimized
     Nodes      Integration       Delivered
1G Turn Key Hadoop Pilot Rack



                           l  1 x Management Node for
                                   Cloudera SCM Express
                           l  2 x Name/Job Tracker Nodes
                           l  34 x Data Nodes
          +           =    l  1Gbps Networking
                                 l  1 x 48 Port Switch
                                          l  Data Network
                                 l  1 x 48 Port Switch
                                          l  IPMI Management Network
                           l  204-408 TB Rack Capacity

Integration   Delivered
Focused on Improving TCO on all Slices
http://www.supermicro.com
http://www.supermicro.com/hadoop
http://www.cloudera.com/partners

More Related Content

PDF
Supermicro cloudera hadoop
PDF
AI橋渡しクラウド(ABCI)における高性能計算とAI/ビッグデータ処理の融合
PDF
20201006_PGconf_Online_Large_Data_Processing
PDF
GTC Japan 2014
PDF
ABCI: AI Bridging Cloud Infrastructure for Scalable AI/Big Data
PDF
20181212 - PGconfASIA - LT - English
PDF
GIST AI-X Computing Cluster
PDF
PG-Strom - GPGPU meets PostgreSQL, PGcon2015
Supermicro cloudera hadoop
AI橋渡しクラウド(ABCI)における高性能計算とAI/ビッグデータ処理の融合
20201006_PGconf_Online_Large_Data_Processing
GTC Japan 2014
ABCI: AI Bridging Cloud Infrastructure for Scalable AI/Big Data
20181212 - PGconfASIA - LT - English
GIST AI-X Computing Cluster
PG-Strom - GPGPU meets PostgreSQL, PGcon2015

What's hot (20)

PDF
dCUDA: Distributed GPU Computing with Hardware Overlap
PDF
SQL+GPU+SSD=∞ (English)
PDF
Let's turn your PostgreSQL into columnar store with cstore_fdw
PDF
20181016_pgconfeu_ssd2gpu_multi
PDF
GPGPU Accelerates PostgreSQL (English)
PDF
20201128_OSC_Fukuoka_Online_GPUPostGIS
PDF
20170602_OSSummit_an_intelligent_storage
PDF
GPU/SSD Accelerates PostgreSQL - challenge towards query processing throughpu...
PDF
PL/CUDA - Fusion of HPC Grade Power with In-Database Analytics
PDF
Building Software Ecosystems for AI Cloud using Singularity HPC Container
PDF
NSCC Training - Introductory Class
PPTX
NSCC Training Introductory Class
PDF
NSCC Training Introductory Class
PDF
PG-Strom
PDF
20180920_DBTS_PGStrom_EN
PDF
PG-Strom v2.0 Technical Brief (17-Apr-2018)
PDF
PDF
GPGPU Accelerates PostgreSQL ~Unlock the power of multi-thousand cores~
PDF
pgconfasia2016 plcuda en
PDF
20181116 Massive Log Processing using I/O optimized PostgreSQL
dCUDA: Distributed GPU Computing with Hardware Overlap
SQL+GPU+SSD=∞ (English)
Let's turn your PostgreSQL into columnar store with cstore_fdw
20181016_pgconfeu_ssd2gpu_multi
GPGPU Accelerates PostgreSQL (English)
20201128_OSC_Fukuoka_Online_GPUPostGIS
20170602_OSSummit_an_intelligent_storage
GPU/SSD Accelerates PostgreSQL - challenge towards query processing throughpu...
PL/CUDA - Fusion of HPC Grade Power with In-Database Analytics
Building Software Ecosystems for AI Cloud using Singularity HPC Container
NSCC Training - Introductory Class
NSCC Training Introductory Class
NSCC Training Introductory Class
PG-Strom
20180920_DBTS_PGStrom_EN
PG-Strom v2.0 Technical Brief (17-Apr-2018)
GPGPU Accelerates PostgreSQL ~Unlock the power of multi-thousand cores~
pgconfasia2016 plcuda en
20181116 Massive Log Processing using I/O optimized PostgreSQL
Ad

Viewers also liked (18)

PDF
Solving Big Data Challenges for Enterprise Application Performance Management
PDF
Data Decay: Why Your CRM Data SUCKS
PPT
Composite Information Server
PPT
Pervasive DataRush
PDF
RingLead: Accelerating Revenue Through Smart Data
PDF
How Intronis Reduced the Data Import Process With RingLead
PDF
How to create self-service analytics tool from activity logs garbage
PPTX
Business driven BI - Self-service Techniques
PDF
Datameer Analytics Solution
PDF
Learning-based Data Cleaning
PDF
Dimension Data – Enabling the Journey to the Cloud: Real Examples
PDF
Data Rehab Series: Automating Taxonomy
PDF
Run IT as Business Meetup self-service BI
PPT
Self Serve Training in SAP
PPTX
Tableau Administrators User Group - Data Governance
PDF
Best Practices for Big Data Analytics with Machine Learning by Datameer
PPTX
Enabling Self Service Business Intelligence using Excel
PPTX
Self-Service Analytics on Hadoop: Lessons Learned
Solving Big Data Challenges for Enterprise Application Performance Management
Data Decay: Why Your CRM Data SUCKS
Composite Information Server
Pervasive DataRush
RingLead: Accelerating Revenue Through Smart Data
How Intronis Reduced the Data Import Process With RingLead
How to create self-service analytics tool from activity logs garbage
Business driven BI - Self-service Techniques
Datameer Analytics Solution
Learning-based Data Cleaning
Dimension Data – Enabling the Journey to the Cloud: Real Examples
Data Rehab Series: Automating Taxonomy
Run IT as Business Meetup self-service BI
Self Serve Training in SAP
Tableau Administrators User Group - Data Governance
Best Practices for Big Data Analytics with Machine Learning by Datameer
Enabling Self Service Business Intelligence using Excel
Self-Service Analytics on Hadoop: Lessons Learned
Ad

Similar to Supermicro High Performance Enterprise Hadoop Infrastructure (20)

PDF
Hadoop on a personal supercomputer
PDF
Greenplum Analytics Workbench - What Can a Private Hadoop Cloud Do For You?
 
PDF
App Cap2956v2 121001194956 Phpapp01 (1)
PDF
Inside the Hadoop Machine @ VMworld
PDF
App cap2956v2-121001194956-phpapp01 (1)
PDF
Apache Hadoop on Virtual Machines
PDF
Introduction to National Supercomputer center in Tianjin TH-1A Supercomputer
PDF
Introducing Amazon RDS Using Oracle Database
PPTX
Hadoop World 2011: Hadoop as a Service in Cloud
PDF
Hadoop on VMware
PPT
IBM: Servery a datová úložiště vhodná pro virtualizaci a budování privátních ...
PDF
Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃
PPTX
Apache con 2012 taking the guesswork out of your hadoop infrastructure
PDF
Oracle on aws overview sep 2011
PDF
6dec2011 - DELL
PPTX
High-Density Top-Loading Storage for Cloud Scale Applications
PDF
Clustering Enhancements
PDF
Dell high density GPU solution
PPTX
Delivering Supermicro Software Defined Storage Solutions with OSNexus QuantaStor
PDF
CC-4009, "Optimizing Hadoop Deployments with SeaMicro SM15000" by Satheesh Na...
Hadoop on a personal supercomputer
Greenplum Analytics Workbench - What Can a Private Hadoop Cloud Do For You?
 
App Cap2956v2 121001194956 Phpapp01 (1)
Inside the Hadoop Machine @ VMworld
App cap2956v2-121001194956-phpapp01 (1)
Apache Hadoop on Virtual Machines
Introduction to National Supercomputer center in Tianjin TH-1A Supercomputer
Introducing Amazon RDS Using Oracle Database
Hadoop World 2011: Hadoop as a Service in Cloud
Hadoop on VMware
IBM: Servery a datová úložiště vhodná pro virtualizaci a budování privátních ...
Track B-3 解構大數據架構 - 大數據系統的伺服器與網路資源規劃
Apache con 2012 taking the guesswork out of your hadoop infrastructure
Oracle on aws overview sep 2011
6dec2011 - DELL
High-Density Top-Loading Storage for Cloud Scale Applications
Clustering Enhancements
Dell high density GPU solution
Delivering Supermicro Software Defined Storage Solutions with OSNexus QuantaStor
CC-4009, "Optimizing Hadoop Deployments with SeaMicro SM15000" by Satheesh Na...

More from templedf (7)

PPTX
JavaOne14 Hands-on Hadoop
PPTX
Java one14 handsonhadoop
PPTX
Couchbase Server
PPTX
Revolution Analytics
PPTX
Talend
PDF
Puppet Labs Puppet Enterprise
PPTX
Couchbase
JavaOne14 Hands-on Hadoop
Java one14 handsonhadoop
Couchbase Server
Revolution Analytics
Talend
Puppet Labs Puppet Enterprise
Couchbase

Recently uploaded (20)

PDF
DP Operators-handbook-extract for the Mautical Institute
PDF
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
PPT
Geologic Time for studying geology for geologist
PDF
NewMind AI Weekly Chronicles – August ’25 Week III
PDF
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
PDF
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
PDF
sustainability-14-14877-v2.pddhzftheheeeee
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PPTX
Group 1 Presentation -Planning and Decision Making .pptx
PDF
A novel scalable deep ensemble learning framework for big data classification...
PDF
WOOl fibre morphology and structure.pdf for textiles
PDF
Developing a website for English-speaking practice to English as a foreign la...
PDF
A comparative study of natural language inference in Swahili using monolingua...
PPTX
Tartificialntelligence_presentation.pptx
PDF
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
PPTX
Chapter 5: Probability Theory and Statistics
DOCX
search engine optimization ppt fir known well about this
PDF
August Patch Tuesday
PDF
Enhancing emotion recognition model for a student engagement use case through...
PPTX
Modernising the Digital Integration Hub
DP Operators-handbook-extract for the Mautical Institute
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
Geologic Time for studying geology for geologist
NewMind AI Weekly Chronicles – August ’25 Week III
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
sustainability-14-14877-v2.pddhzftheheeeee
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
Group 1 Presentation -Planning and Decision Making .pptx
A novel scalable deep ensemble learning framework for big data classification...
WOOl fibre morphology and structure.pdf for textiles
Developing a website for English-speaking practice to English as a foreign la...
A comparative study of natural language inference in Swahili using monolingua...
Tartificialntelligence_presentation.pptx
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
Chapter 5: Probability Theory and Statistics
search engine optimization ppt fir known well about this
August Patch Tuesday
Enhancing emotion recognition model for a student engagement use case through...
Modernising the Digital Integration Hub

Supermicro High Performance Enterprise Hadoop Infrastructure

  • 2. Supermicro © 2009 Cloudera Ready - Infrastructure for Hadoop SuperBlade® GPU Solutions Storage Universal I/O Twin Architecture Double-Sided Datacenter Optimized www.supermicro.com/hadoop
  • 3. Building Block Platforms for Hadoop l  1U 815 Compute Hadoop Nodes l  Dual Quad Core Xeon l  4, 3.5 HDD spindles per node l  Hot Swap HDD, PSUs l  Up to 8 TB per 1U l  Common Components l  On Board IPMI Remote Management l  93% Efficiency PMBus PSUs l  2UTwin 827 Hadoop Nodes l  Dual Quad Core Xeon l  6, 3.5 HDD spindles per node l  2 Nodes in 2U = 6 HDD per U l  On Board IPMI Remote Management l  Hot Swap Nodes, HDD, PSUs l  94% Efficiency PMBus PSUs
  • 4. Building Block Platforms for Hadoop l  2U 825 Balanced Hadoop Nodes l  Dual Quad Core Xeon l  8, 3.5 HDD l  On Board IPMI Remote Management l  Common Components l  Hot Swap HDD, PSUs l  93% Efficiency PMBus PSUs l  2U 826 Flex Hadoop Nodes l  Dual Quad Core Xeon l  12, 3.5 HDD l  On Board IPMI Remote Management l  Hot Swap HDD, PSUs l  Up to 24TB per 2U l  Common Components l  93% Efficiency PMBus PSUs
  • 5. Optimized Hadoop Nodes - Cloudera Ready Name Node / Job Tracker: Name Node – 48GB PIO-HNU0-15056248-HADP Data Nodes: 1TB SATA Drives 2TB SATA Drives Data Node – 4 x 1TB 24GB Data Node – 4 x 2TB 24GB PIO-HDU0-15156224-HADP 1U PIO-HDU0-15256224-HADP 1U Data Node – 6 x 1TB 24GB Data Node – 6 x 2TB 24GB PIO-HDT0-27156224-HADP 2UTwin PIO-HDT0-27256224-HADP 2UTwin Data Node – 8 x 1TB 24GB Data Node – 8 x 2TB 24GB PIO-HDU0-25156224-HADP 2U PIO-HDU0-25256224-HADP 2U Data Node – 12 x 1TB 24GB Data Node – 12 x 2TB 24GB PIO-HDU0-26156224-HADP 2U PIO-HDU0-26256224-HADP 2U
  • 6. Turn Key Pilot Cluster Kits Pilot Kit - Configuration: Ethernet Switch: 1G Ethernet Switch (10G Uplink Capability) SSE-G24-TG4 SSE-G48-TG4 1U 24 port 1U 48 port 10 Node Pilot Cluster Name Node – 48GB • 48 TB Name Node: PIO-HNU0-15056248-HADP Secondary Name Node – 48GB • 3.5 SATA Name/Job Tracker: PIO-HNU0-15056248-HADP • 64 Cores 2 Data Nodes: Data Node – 6 x 1TB 24GB PIO-HDT0-27156224-HADP 2UTwin • 48 Spindles 2 Data Nodes: Data Node – 6 x 1TB 24GB PIO-HDT0-27156224-HADP 2UTwin • 288 GB Memory 2 Data Nodes: Data Node – 6 x 1TB 24GB PIO-HDT0-27156224-HADP 2UTwin • Occupy 11U 2 Data Nodes: Data Node – 6 x 1TB 24GB PIO-HDT0-27156224-HADP 2UTwin
  • 7. SUPERMICRO – Cloudera Certified Technology
  • 8. Hadoop Series Naming Convention PIO-HDU0-26156224-HADP H = Hadoop 24 = GB Memory D = Data Node 48 = GB Memory N = Name Node U = X8DTU-F 562 = Xeon E5620 T = X8DTT-HF 0 = Intel 1 = AMD 0 = OS HDD x 4 1 = 1TB HDD 15 = 815 - 2U - 4 HDD 2 = 2TB HDD 25 = 825 - 2U - 8 HDD 26 = 826 - 2U - 12 HDD 27 = 827 - 2UTwin - 2x6 HDD
  • 9. Hadoop Series Naming Convention Model/Part # Description / Functionality PIO-HNU0-15056248-HADP 1U Hadoop 815 Name E5620 48GB MEM PIO-HDU0-15156224-HADP 1U Hadoop 815 4 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDU0-15256224-HADP 1U Hadoop 815 4 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDT0-27156224-HADP 2U Hadoop 827 6 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDT0-27256224-HADP 2U Hadoop 827 6 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDU0-25156224-HADP 2U Hadoop 825 8 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDU0-25256224-HADP 2U Hadoop 825 8 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDU0-26156224-HADP 2U Hadoop 826 12 1TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM PIO-HDU0-26256224-HADP 2U Hadoop 826 12 2TB 4C E5620 2.4G 12M 5.86GT 80W 24GB MEM
  • 10. Turn Key Hadoop Cluster Racks + + + = ROB Return on Byte$ Optimized Nodes Integration Delivered
  • 11. 1G Turn Key Hadoop Pilot Rack l  1 x Management Node for Cloudera SCM Express l  2 x Name/Job Tracker Nodes l  34 x Data Nodes + = l  1Gbps Networking l  1 x 48 Port Switch l  Data Network l  1 x 48 Port Switch l  IPMI Management Network l  204-408 TB Rack Capacity Integration Delivered
  • 12. Focused on Improving TCO on all Slices