SlideShare a Scribd company logo
By: Ashraful Hoda
Ashraful.hoda01@mail.com










Introduction
History
why cluster computing?
Architecture
Clustering Concept
Several application
Operating System
Companies that use it
High performance Clusters (HPC)

2
 A computer cluster consists of a set of loosely connected

computers that work together so that in many respects they
can be viewed as a single system.

Cluster consists of:
 Nodes(master+computing)
 Network
 OS
 Cluster middleware: It permits
compute clustering programs
to be portable to a wide variety
of clusters.

…

Cluster Middle ware

High Speed Local Network

CPU

CPU

…

CPU

Cluster
3
INTRODUCTION
 Consists of many of the same or similar type of machines.

 Tightly-coupled using dedicated network connections.
 The components of a cluster are usually connected to each other

through fast local area networks, each node running its own
instance on an operating system.
 All machines share resources.
 They must trust each other so that does not require a password,

otherwise you would need to do a manual start on each machine.
HISTORY
 The first commercial clustering product was ARCnet, developed

by Data point in 1977.
 Digital Equipment Corporation released their VAX cluster product

in 1984 for the VAX/VMS operating system.
 The ARCnet and VAX cluster products not only supported parallel

computing, but also shared file systems and peripheral devices.
 The idea was to provide the advantages of parallel processing,

while maintaining data reliability and uniqueness.
Through CLUSTERS
 Data sharing
 Message passing and communication
 Task scheduling
 Node failure management
 Parallel programming
 Debugging and monitoring
Logical view
ARCHITECTURE
 It consists of a collection of interconnected stand-alone computers

cooperatively working together a single, integrated computing
resource.
 A node:


a single or multiprocessor system with memory, I/O facilities, &OS



generally 2 or more computers (nodes) connected together in a single
cabinet, or physically separated & connected via a LAN appear as a
single system to users and applications provide a cost-effective way to
gain features and benefits
ARCHITECTURE
 Database Replication Clusters
The components required to the development
of low cost clusters are:





Processors
Memory
Networking components
Motherboards, busses, and other sub-systems
Beowulf cluster
Start from 1994
Donald Becker of NASA assembled this cluster.
Also called Beowulf cluster
Applications like data mining, simulations, parallel processing,
weather modeling, etc

 A Beowulf Cluster is a computer design that uses parallel processing

across multiple computers to create cheap and powerful
supercomputers. A Beowulf Cluster in practice is usually a collection of
generic computers connected through an internal network.
 A cluster has two types of computers, a master computer, and node

computers.
 When a large problem or set of data is given to a Beowulf cluster, the

master computer first runs a program that breaks the problem into small
discrete pieces; it then sends a piece to each node to compute. As nodes
finish their tasks, the master computer continually sends more pieces to
them until the entire problem has been computed.

12
( Ethernet,Myrinet….)
+ (MPI)

 Master: or service node or front node ( used to interact with users and manage the
cluster )
 Nodes : a group of computers (computing node
s)( keyboard, mouse, floppy, video…)
 Communications between nodes on an interconnect network platform
( Ethernet, Myrinet….)
 In order for the master and node computers to communicate, some sort message
passing control structure is required. MPI,(Message Passing Interface) is the most
commonly used such control.

13
Brief Technical Parameters:


OS:



Service node:



Computing nodes:



System Memory:






CentOS 5 managed by Rochs-cluster
1 (Intel P4 2.4 GHz)
32(Intel P4 2.4- 2.8 GHz)
1 GB per node

Network Platforms:

Gigabit Ethernet, 2 cards per node
Myrinet 2 G
Language: C, C++, Fortran, java
Compiler: Intel compiler, sun Java compiler
14
Science
Computation

Digital Biology

Aerospace

Resources
Exploration

15
High Performance Networks/Switches
a. Ethernet (10Mbps),
b. Fast Ethernet (100Mbps),
c. Gigabit Ethernet (1Gbps)
e. ATM
f. Myrinet (1.2Gbps)
g. Digital Memory Channel
ISSUES TO BE CONSIDERED
Cluster Networking
Cluster Software
Programming

Timing
Network Selection
Speed Selection
OS (Operating System )
 Three of the most commonly used OS are :


Windows
mainly used to build a High Availability Cluster or a NLB(Network Local
Balance) Cluster, provide services such as Database, File/Print,Web,Stream
Media .Support 2-4 SMP or 32 processors. Hardly used to build a Science
Computing Cluster



Redhat Linux
The most used OS for a Beowulf Cluster. provides High Performance and
Scalability / High Reliability / Low Cost ( get freely and uses inexpensive
commodity hardware )



SUN Solaris
Uses expensive and unpopular hardware
18
State of the art Operating Systems with
companies :
a. Linux
b. Microsoft NT
c. SUN Solaris
d. IBM AIX
e. HP UX (Illinois - PANDA)
20
 Calculation procedure for peak performance:
 No of nodes 64
 NO. of Master Nodes : 1
 Memory RAM: 4 GB
 Hard Disk Capacity/each node : 250GB
 Storage Cap. 4 TB
 CLUSTER Software : ROCKS version 4.3
 No .of processors and cores: 2 X 2 = 4(dual core +
dual socket)
 CPU speed : 2.6 GHz
21
Cluster  computer
ANY
QUERIES ???
Ad

More Related Content

What's hot (20)

Cluster computing
Cluster computingCluster computing
Cluster computing
Raja' Masa'deh
 
cluster computing
cluster computingcluster computing
cluster computing
anjalibhandari11011995
 
Cloud deployment models
Cloud deployment modelsCloud deployment models
Cloud deployment models
Ashok Kumar
 
Cloud Deployments Models
Cloud Deployments ModelsCloud Deployments Models
Cloud Deployments Models
Mohamed Sami El-Tahawy
 
Lecture5
Lecture5Lecture5
Lecture5
josephineusha
 
Wireless Networking Security
Wireless Networking SecurityWireless Networking Security
Wireless Networking Security
Anshuman Biswal
 
Introduction to Cloud Data Center and Network Issues
Introduction to Cloud Data Center and Network IssuesIntroduction to Cloud Data Center and Network Issues
Introduction to Cloud Data Center and Network Issues
Jason TC HOU (侯宗成)
 
Octave
OctaveOctave
Octave
Amar Myana
 
File models and file accessing models
File models and file accessing modelsFile models and file accessing models
File models and file accessing models
ishmecse13
 
Dmz
Dmz Dmz
Dmz
أحلام انصارى
 
High performance computing
High performance computingHigh performance computing
High performance computing
punjab engineering college, chandigarh
 
Imt 2000
Imt 2000Imt 2000
Imt 2000
bharanisamratelesela
 
Cluster computing ppt
Cluster computing pptCluster computing ppt
Cluster computing ppt
DC Graphics
 
Vanet ppt
Vanet pptVanet ppt
Vanet ppt
Akash Raghunath
 
Threat Modeling for IoT Systems
Threat Modeling for IoT SystemsThreat Modeling for IoT Systems
Threat Modeling for IoT Systems
Denim Group
 
Beowulf cluster
Beowulf clusterBeowulf cluster
Beowulf cluster
Yash Karanke
 
Distributed computing bsics
Distributed computing bsicsDistributed computing bsics
Distributed computing bsics
Deepak John
 
Introduction to Virtualization
Introduction to VirtualizationIntroduction to Virtualization
Introduction to Virtualization
Rahul Hada
 
CLOUD COMPUTING UNIT-1
CLOUD COMPUTING UNIT-1CLOUD COMPUTING UNIT-1
CLOUD COMPUTING UNIT-1
Dr K V Subba Reddy
 
Load balancing in Distributed Systems
Load balancing in Distributed SystemsLoad balancing in Distributed Systems
Load balancing in Distributed Systems
Richa Singh
 

Viewers also liked (6)

Information Technology Basics
Information Technology BasicsInformation Technology Basics
Information Technology Basics
National College of Business Administration and Economics
 
What we mean by meaning: new structural properties of information architectur...
What we mean by meaning: new structural properties of information architectur...What we mean by meaning: new structural properties of information architectur...
What we mean by meaning: new structural properties of information architectur...
Marsha Haverty
 
Cloud computing
Cloud computingCloud computing
Cloud computing
RobbySahoo
 
Mobile Fundraising (July 2012)
Mobile Fundraising (July 2012)Mobile Fundraising (July 2012)
Mobile Fundraising (July 2012)
Richard Otto
 
PPT on Cloud computing
PPT on Cloud computingPPT on Cloud computing
PPT on Cloud computing
Lakshita Mukul
 
Basic Concepts Of Information Technology (It)
Basic Concepts Of Information Technology (It)Basic Concepts Of Information Technology (It)
Basic Concepts Of Information Technology (It)
Dr. Dheeraj Mehrotra (National Awardee)
 
Ad

Similar to Cluster computer (20)

Parallel_and_Cluster_Computing.ppt
Parallel_and_Cluster_Computing.pptParallel_and_Cluster_Computing.ppt
Parallel_and_Cluster_Computing.ppt
MohmdUmer
 
Cluster Computing
Cluster ComputingCluster Computing
Cluster Computing
NIKHIL NAIR
 
Copy of Parallel_and_Cluster_Computing.pptx
Copy of Parallel_and_Cluster_Computing.pptxCopy of Parallel_and_Cluster_Computing.pptx
Copy of Parallel_and_Cluster_Computing.pptx
gopikahari7
 
59137949-Cluster-Computing (1).ppt .....
59137949-Cluster-Computing (1).ppt .....59137949-Cluster-Computing (1).ppt .....
59137949-Cluster-Computing (1).ppt .....
ananthupradeep00
 
Cluster Computers
Cluster ComputersCluster Computers
Cluster Computers
shopnil786
 
Cluster computing
Cluster computingCluster computing
Cluster computing
Shashwat Shriparv
 
Computer_Clustering_Technologies
Computer_Clustering_TechnologiesComputer_Clustering_Technologies
Computer_Clustering_Technologies
Manish Chopra
 
Clustering by AKASHMSHAH
Clustering by AKASHMSHAHClustering by AKASHMSHAH
Clustering by AKASHMSHAH
Akash M Shah
 
Cluster computing
Cluster computingCluster computing
Cluster computing
Venkat Sai Sharath Mudhigonda
 
Cluster computing
Cluster computingCluster computing
Cluster computing
pooja khatana
 
Maxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorialMaxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorial
madhuinturi
 
Distributed Computing
Distributed ComputingDistributed Computing
Distributed Computing
Sudarsun Santhiappan
 
Distributed Computing
Distributed ComputingDistributed Computing
Distributed Computing
Prashant Tiwari
 
Ap 06 4_10_simek
Ap 06 4_10_simekAp 06 4_10_simek
Ap 06 4_10_simek
Nguyen Vinh
 
Again music
Again musicAgain music
Again music
variable_orr
 
Computer cluster
Computer clusterComputer cluster
Computer cluster
Shiva Krishna Chandra Shekar
 
Computer cluster
Computer clusterComputer cluster
Computer cluster
Shiva Krishna Chandra Shekar
 
Clustering
ClusteringClustering
Clustering
Abhay Pai
 
Systems Support for Many Task Computing
Systems Support for Many Task ComputingSystems Support for Many Task Computing
Systems Support for Many Task Computing
Eric Van Hensbergen
 
Free Hardware & Networking Slides by ITE Infotech Private Limited
Free Hardware & Networking Slides by ITE Infotech Private LimitedFree Hardware & Networking Slides by ITE Infotech Private Limited
Free Hardware & Networking Slides by ITE Infotech Private Limited
Hemraj Singh Chouhan
 
Parallel_and_Cluster_Computing.ppt
Parallel_and_Cluster_Computing.pptParallel_and_Cluster_Computing.ppt
Parallel_and_Cluster_Computing.ppt
MohmdUmer
 
Cluster Computing
Cluster ComputingCluster Computing
Cluster Computing
NIKHIL NAIR
 
Copy of Parallel_and_Cluster_Computing.pptx
Copy of Parallel_and_Cluster_Computing.pptxCopy of Parallel_and_Cluster_Computing.pptx
Copy of Parallel_and_Cluster_Computing.pptx
gopikahari7
 
59137949-Cluster-Computing (1).ppt .....
59137949-Cluster-Computing (1).ppt .....59137949-Cluster-Computing (1).ppt .....
59137949-Cluster-Computing (1).ppt .....
ananthupradeep00
 
Cluster Computers
Cluster ComputersCluster Computers
Cluster Computers
shopnil786
 
Computer_Clustering_Technologies
Computer_Clustering_TechnologiesComputer_Clustering_Technologies
Computer_Clustering_Technologies
Manish Chopra
 
Clustering by AKASHMSHAH
Clustering by AKASHMSHAHClustering by AKASHMSHAH
Clustering by AKASHMSHAH
Akash M Shah
 
Maxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorialMaxwell siuc hpc_description_tutorial
Maxwell siuc hpc_description_tutorial
madhuinturi
 
Ap 06 4_10_simek
Ap 06 4_10_simekAp 06 4_10_simek
Ap 06 4_10_simek
Nguyen Vinh
 
Systems Support for Many Task Computing
Systems Support for Many Task ComputingSystems Support for Many Task Computing
Systems Support for Many Task Computing
Eric Van Hensbergen
 
Free Hardware & Networking Slides by ITE Infotech Private Limited
Free Hardware & Networking Slides by ITE Infotech Private LimitedFree Hardware & Networking Slides by ITE Infotech Private Limited
Free Hardware & Networking Slides by ITE Infotech Private Limited
Hemraj Singh Chouhan
 
Ad

More from Ashraful Hoda (7)

Group Discussion
Group DiscussionGroup Discussion
Group Discussion
Ashraful Hoda
 
Nanorobotics
NanoroboticsNanorobotics
Nanorobotics
Ashraful Hoda
 
Prasar barti "DD"
Prasar barti "DD"Prasar barti "DD"
Prasar barti "DD"
Ashraful Hoda
 
Principle of management
Principle of managementPrinciple of management
Principle of management
Ashraful Hoda
 
Management Strategy
Management StrategyManagement Strategy
Management Strategy
Ashraful Hoda
 
Nokia morph
Nokia morphNokia morph
Nokia morph
Ashraful Hoda
 
Gd pi tips
Gd pi tipsGd pi tips
Gd pi tips
Ashraful Hoda
 

Recently uploaded (20)

Q1 2025 Dropbox Earnings and Investor Presentation
Q1 2025 Dropbox Earnings and Investor PresentationQ1 2025 Dropbox Earnings and Investor Presentation
Q1 2025 Dropbox Earnings and Investor Presentation
Dropbox
 
AI You Can Trust: The Critical Role of Governance and Quality.pdf
AI You Can Trust: The Critical Role of Governance and Quality.pdfAI You Can Trust: The Critical Role of Governance and Quality.pdf
AI You Can Trust: The Critical Role of Governance and Quality.pdf
Precisely
 
DevOpsDays SLC - Platform Engineers are Product Managers.pptx
DevOpsDays SLC - Platform Engineers are Product Managers.pptxDevOpsDays SLC - Platform Engineers are Product Managers.pptx
DevOpsDays SLC - Platform Engineers are Product Managers.pptx
Justin Reock
 
GyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
GyrusAI - Broadcasting & Streaming Applications Driven by AI and MLGyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
GyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
Gyrus AI
 
How to Install & Activate ListGrabber - eGrabber
How to Install & Activate ListGrabber - eGrabberHow to Install & Activate ListGrabber - eGrabber
How to Install & Activate ListGrabber - eGrabber
eGrabber
 
Config 2025 presentation recap covering both days
Config 2025 presentation recap covering both daysConfig 2025 presentation recap covering both days
Config 2025 presentation recap covering both days
TrishAntoni1
 
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
James Anderson
 
Transcript: Canadian book publishing: Insights from the latest salary survey ...
Transcript: Canadian book publishing: Insights from the latest salary survey ...Transcript: Canadian book publishing: Insights from the latest salary survey ...
Transcript: Canadian book publishing: Insights from the latest salary survey ...
BookNet Canada
 
Shoehorning dependency injection into a FP language, what does it take?
Shoehorning dependency injection into a FP language, what does it take?Shoehorning dependency injection into a FP language, what does it take?
Shoehorning dependency injection into a FP language, what does it take?
Eric Torreborre
 
Slack like a pro: strategies for 10x engineering teams
Slack like a pro: strategies for 10x engineering teamsSlack like a pro: strategies for 10x engineering teams
Slack like a pro: strategies for 10x engineering teams
Nacho Cougil
 
machines-for-woodworking-shops-en-compressed.pdf
machines-for-woodworking-shops-en-compressed.pdfmachines-for-woodworking-shops-en-compressed.pdf
machines-for-woodworking-shops-en-compressed.pdf
AmirStern2
 
Mastering Testing in the Modern F&B Landscape
Mastering Testing in the Modern F&B LandscapeMastering Testing in the Modern F&B Landscape
Mastering Testing in the Modern F&B Landscape
marketing943205
 
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à GenèveUiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPathCommunity
 
Viam product demo_ Deploying and scaling AI with hardware.pdf
Viam product demo_ Deploying and scaling AI with hardware.pdfViam product demo_ Deploying and scaling AI with hardware.pdf
Viam product demo_ Deploying and scaling AI with hardware.pdf
camilalamoratta
 
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
SOFTTECHHUB
 
Financial Services Technology Summit 2025
Financial Services Technology Summit 2025Financial Services Technology Summit 2025
Financial Services Technology Summit 2025
Ray Bugg
 
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Raffi Khatchadourian
 
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Safe Software
 
IT484 Cyber Forensics_Information Technology
IT484 Cyber Forensics_Information TechnologyIT484 Cyber Forensics_Information Technology
IT484 Cyber Forensics_Information Technology
SHEHABALYAMANI
 
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Markus Eisele
 
Q1 2025 Dropbox Earnings and Investor Presentation
Q1 2025 Dropbox Earnings and Investor PresentationQ1 2025 Dropbox Earnings and Investor Presentation
Q1 2025 Dropbox Earnings and Investor Presentation
Dropbox
 
AI You Can Trust: The Critical Role of Governance and Quality.pdf
AI You Can Trust: The Critical Role of Governance and Quality.pdfAI You Can Trust: The Critical Role of Governance and Quality.pdf
AI You Can Trust: The Critical Role of Governance and Quality.pdf
Precisely
 
DevOpsDays SLC - Platform Engineers are Product Managers.pptx
DevOpsDays SLC - Platform Engineers are Product Managers.pptxDevOpsDays SLC - Platform Engineers are Product Managers.pptx
DevOpsDays SLC - Platform Engineers are Product Managers.pptx
Justin Reock
 
GyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
GyrusAI - Broadcasting & Streaming Applications Driven by AI and MLGyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
GyrusAI - Broadcasting & Streaming Applications Driven by AI and ML
Gyrus AI
 
How to Install & Activate ListGrabber - eGrabber
How to Install & Activate ListGrabber - eGrabberHow to Install & Activate ListGrabber - eGrabber
How to Install & Activate ListGrabber - eGrabber
eGrabber
 
Config 2025 presentation recap covering both days
Config 2025 presentation recap covering both daysConfig 2025 presentation recap covering both days
Config 2025 presentation recap covering both days
TrishAntoni1
 
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
GDG Cloud Southlake #42: Suresh Mathew: Autonomous Resource Optimization: How...
James Anderson
 
Transcript: Canadian book publishing: Insights from the latest salary survey ...
Transcript: Canadian book publishing: Insights from the latest salary survey ...Transcript: Canadian book publishing: Insights from the latest salary survey ...
Transcript: Canadian book publishing: Insights from the latest salary survey ...
BookNet Canada
 
Shoehorning dependency injection into a FP language, what does it take?
Shoehorning dependency injection into a FP language, what does it take?Shoehorning dependency injection into a FP language, what does it take?
Shoehorning dependency injection into a FP language, what does it take?
Eric Torreborre
 
Slack like a pro: strategies for 10x engineering teams
Slack like a pro: strategies for 10x engineering teamsSlack like a pro: strategies for 10x engineering teams
Slack like a pro: strategies for 10x engineering teams
Nacho Cougil
 
machines-for-woodworking-shops-en-compressed.pdf
machines-for-woodworking-shops-en-compressed.pdfmachines-for-woodworking-shops-en-compressed.pdf
machines-for-woodworking-shops-en-compressed.pdf
AmirStern2
 
Mastering Testing in the Modern F&B Landscape
Mastering Testing in the Modern F&B LandscapeMastering Testing in the Modern F&B Landscape
Mastering Testing in the Modern F&B Landscape
marketing943205
 
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à GenèveUiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à Genève
UiPathCommunity
 
Viam product demo_ Deploying and scaling AI with hardware.pdf
Viam product demo_ Deploying and scaling AI with hardware.pdfViam product demo_ Deploying and scaling AI with hardware.pdf
Viam product demo_ Deploying and scaling AI with hardware.pdf
camilalamoratta
 
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
The No-Code Way to Build a Marketing Team with One AI Agent (Download the n8n...
SOFTTECHHUB
 
Financial Services Technology Summit 2025
Financial Services Technology Summit 2025Financial Services Technology Summit 2025
Financial Services Technology Summit 2025
Ray Bugg
 
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...
Raffi Khatchadourian
 
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Integrating FME with Python: Tips, Demos, and Best Practices for Powerful Aut...
Safe Software
 
IT484 Cyber Forensics_Information Technology
IT484 Cyber Forensics_Information TechnologyIT484 Cyber Forensics_Information Technology
IT484 Cyber Forensics_Information Technology
SHEHABALYAMANI
 
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...
Markus Eisele
 

Cluster computer

  • 2.          Introduction History why cluster computing? Architecture Clustering Concept Several application Operating System Companies that use it High performance Clusters (HPC) 2
  • 3.  A computer cluster consists of a set of loosely connected computers that work together so that in many respects they can be viewed as a single system. Cluster consists of:  Nodes(master+computing)  Network  OS  Cluster middleware: It permits compute clustering programs to be portable to a wide variety of clusters. … Cluster Middle ware High Speed Local Network CPU CPU … CPU Cluster 3
  • 4. INTRODUCTION  Consists of many of the same or similar type of machines.  Tightly-coupled using dedicated network connections.  The components of a cluster are usually connected to each other through fast local area networks, each node running its own instance on an operating system.  All machines share resources.  They must trust each other so that does not require a password, otherwise you would need to do a manual start on each machine.
  • 5. HISTORY  The first commercial clustering product was ARCnet, developed by Data point in 1977.  Digital Equipment Corporation released their VAX cluster product in 1984 for the VAX/VMS operating system.  The ARCnet and VAX cluster products not only supported parallel computing, but also shared file systems and peripheral devices.  The idea was to provide the advantages of parallel processing, while maintaining data reliability and uniqueness.
  • 6. Through CLUSTERS  Data sharing  Message passing and communication  Task scheduling  Node failure management  Parallel programming  Debugging and monitoring
  • 8. ARCHITECTURE  It consists of a collection of interconnected stand-alone computers cooperatively working together a single, integrated computing resource.  A node:  a single or multiprocessor system with memory, I/O facilities, &OS  generally 2 or more computers (nodes) connected together in a single cabinet, or physically separated & connected via a LAN appear as a single system to users and applications provide a cost-effective way to gain features and benefits
  • 10. The components required to the development of low cost clusters are:     Processors Memory Networking components Motherboards, busses, and other sub-systems
  • 11. Beowulf cluster Start from 1994 Donald Becker of NASA assembled this cluster. Also called Beowulf cluster Applications like data mining, simulations, parallel processing, weather modeling, etc 
  • 12.  A Beowulf Cluster is a computer design that uses parallel processing across multiple computers to create cheap and powerful supercomputers. A Beowulf Cluster in practice is usually a collection of generic computers connected through an internal network.  A cluster has two types of computers, a master computer, and node computers.  When a large problem or set of data is given to a Beowulf cluster, the master computer first runs a program that breaks the problem into small discrete pieces; it then sends a piece to each node to compute. As nodes finish their tasks, the master computer continually sends more pieces to them until the entire problem has been computed. 12
  • 13. ( Ethernet,Myrinet….) + (MPI)  Master: or service node or front node ( used to interact with users and manage the cluster )  Nodes : a group of computers (computing node s)( keyboard, mouse, floppy, video…)  Communications between nodes on an interconnect network platform ( Ethernet, Myrinet….)  In order for the master and node computers to communicate, some sort message passing control structure is required. MPI,(Message Passing Interface) is the most commonly used such control. 13
  • 14. Brief Technical Parameters:  OS:  Service node:  Computing nodes:  System Memory:    CentOS 5 managed by Rochs-cluster 1 (Intel P4 2.4 GHz) 32(Intel P4 2.4- 2.8 GHz) 1 GB per node Network Platforms: Gigabit Ethernet, 2 cards per node Myrinet 2 G Language: C, C++, Fortran, java Compiler: Intel compiler, sun Java compiler 14
  • 16. High Performance Networks/Switches a. Ethernet (10Mbps), b. Fast Ethernet (100Mbps), c. Gigabit Ethernet (1Gbps) e. ATM f. Myrinet (1.2Gbps) g. Digital Memory Channel
  • 17. ISSUES TO BE CONSIDERED Cluster Networking Cluster Software Programming Timing Network Selection Speed Selection
  • 18. OS (Operating System )  Three of the most commonly used OS are :  Windows mainly used to build a High Availability Cluster or a NLB(Network Local Balance) Cluster, provide services such as Database, File/Print,Web,Stream Media .Support 2-4 SMP or 32 processors. Hardly used to build a Science Computing Cluster  Redhat Linux The most used OS for a Beowulf Cluster. provides High Performance and Scalability / High Reliability / Low Cost ( get freely and uses inexpensive commodity hardware )  SUN Solaris Uses expensive and unpopular hardware 18
  • 19. State of the art Operating Systems with companies : a. Linux b. Microsoft NT c. SUN Solaris d. IBM AIX e. HP UX (Illinois - PANDA)
  • 20. 20
  • 21.  Calculation procedure for peak performance:  No of nodes 64  NO. of Master Nodes : 1  Memory RAM: 4 GB  Hard Disk Capacity/each node : 250GB  Storage Cap. 4 TB  CLUSTER Software : ROCKS version 4.3  No .of processors and cores: 2 X 2 = 4(dual core + dual socket)  CPU speed : 2.6 GHz 21
  翻译: