SlideShare ist ein Scribd-Unternehmen logo
1 von 35
Accelerating MD..Simple Tweaks and Instant Clusters Mohd Shahir Shamsir Bioinformatics Research Group (BIRG) Faculty of Biosciences & Bioengineering Universiti Teknologi Malaysia INSPIRING CREATIVE AND INNOVATIVE MINDS
Summary Introduction to BIRG MD: What, Why and How Improving performance Simple Tweaks… Instant MD cluster– birgHPC Short video demo
Bioinformatics Research Group (BIRG)Faculty of Biosciences & Bioengineering Just Google Us…
MD… What? Why? How? COVERED!!
Performance of MD? Speed, speed, speed… Supercomputer IBM Roadrunner ~368 Flops Nankai Star 3.7 ns/day on 32 nodes (DPPC) HPCx 5.2 ns/day on 64 nodes (DPPC) New platform Cell-BE: GROMACS, 15x to Pentium 3.0GHz GPU: NAMD, 4 GPU = 16 CPUs
Microwulf  cluster 26 Gflops,  $2500, 11" x 12" x 17",  airline overhead baggage compliant
Simple Tweaks INSPIRING CREATIVE AND INNOVATIVE MINDS
Tweaks for MD? Hardware ↑ =  performance ↑ = $$$ ↑ OR Tweak Beowulf = performance ↑ = $$$ Pre-compiled vs self-compiled MPI libraries Test beds: 3 nodes GridMACS, 7 nodes Beowulf, 1 reference machine
Compilation Winner self-compile
Beowulf OpenMPIvs MPICH2 (pre) Winner MPICH2
Pre and Self Compiled MPI Self OpenMPI = MPICH2
What we found Single machine 66 % improvement  Parallel environment 64 % improvement  Compilation, software chosen affect performance
Instant MD, anyone? INSPIRING CREATIVE AND INNOVATIVE MINDS
Instant MD cluster Lots of under utilised computers in labs Idle mode after office hours, holidays, etc.
Instant MD cluster MD, parallel computing = high computing resources Solution? Supercomputers Dedicated computing cluster Problems? $$$ ??? (I don’t know this, I don’t know that…)
A + B = C What is A? Existing computers LAN connected, PXE-boot capable, CDROM/USB What is B? Linux Live CD Auto configuration What is C? Instant, out-of-the-box computing cluster!
birgHPC Free, open-source Linux distribution Based on PelicanHPC & Debian Live GROMACS, NAMD, mpiBLAST, ClustalW-MPI, PyMol, VMD Auto cluster config MPICH2 & OpenMPI Auto slots detections Ganglia monitoring Simple interface for job submission
Some Screenshots
Some Screenshots
Some Screenshots
Some Screenshots
Some Screenshots
Some Screeshots
Available at http://birg1.fbb.utm.my/birghpcor Just Google birghpc
Conclusion birgHPC instant cluster conversion Bioinformatics tools Auto configurations http://birg1.fbb.utm.my/birghpc ISOs guide
Acknowledgements Chew Teong Han - Alchemist Farizuawana – Graphics Joyce Tan – Testing Funding from you via LHDN via MOSTI Michael Creel for Pelican HPC
FAQs Boot sequence Head node -> run birgHPC_setup -> follow instructions -> boot compute node -> script on head node will show # of nodes detected, confirm -> done Headless compute nodes (no monitor) Have to get a monitor -> set boot sequence to netboot -> done
FAQs How to know compute nodes is up Follow the birgHPC boot sequence -> the birgHPC_setup script will show # nodes detected Cannot netboot Try http://etherboot.org/wiki/start Heterogeneous PCs ok? Ok (Thanks Micheal Creel) If 32 bit + 64 bit, use 32 bit PC as head node
FAQs Status monitoring Yes -> web browser -> localhost -> Ganglia Monitoring What will displayed on compute node? Just a simple login terminal with some warning not to use the nodes, etc Limitations? RAM, RAM, RAM … Everything is loaded to RAM, hence HDD size = RAM size
FAQs Headnode criteria Preferably big RAM because of shared folder (/home) 2 Ethernet port if you want internet connection (still work if you got only one eth) Guide? http://birg1.fbb.utm.my/birghpc Forum? No, but can always refer PelicanHPC forums
FAQs Multi users? No, designed for single user Future release, maybe SGE or PBS Installed on hardiskpermenantly? Not tested, technically possible, Google Performance On par with hardisk-installed cluster (tested up to 6 nodes)
FAQs I cannot boot from CD Refer user guide -> convert CD ISO to USB drive image -> boot from USB Can I use birgHPC along with existing DHCP Preferably no, DHCP will distribute IP and birgHPC head will distribute IP, causing confusion of IP addresses Alternative, boot DHCP as head node OR unplug DHCP, use another PC as head node
FAQs birgHPC criteria (PCs = compute, server = head)

Weitere ähnliche Inhalte

Andere mochten auch

Andere mochten auch (14)

ηΣυνταγήΤ
ηΣυνταγήΤηΣυνταγήΤ
ηΣυνταγήΤ
 
Canda 3.0 plr
Canda 3.0  plrCanda 3.0  plr
Canda 3.0 plr
 
Absolutely Gorgeous Photos By German
Absolutely Gorgeous Photos By GermanAbsolutely Gorgeous Photos By German
Absolutely Gorgeous Photos By German
 
Atm ΠΡΟΣ[1]
Atm  ΠΡΟΣ[1]Atm  ΠΡΟΣ[1]
Atm ΠΡΟΣ[1]
 
Rhto
RhtoRhto
Rhto
 
Nowthatswhaticallphotography
NowthatswhaticallphotographyNowthatswhaticallphotography
Nowthatswhaticallphotography
 
Photos Aerienne
Photos AeriennePhotos Aerienne
Photos Aerienne
 
Lexiko Oron
Lexiko OronLexiko Oron
Lexiko Oron
 
D
DD
D
 
Dysnohta(3)
Dysnohta(3)Dysnohta(3)
Dysnohta(3)
 
Unreal
UnrealUnreal
Unreal
 
Antarctic
AntarcticAntarctic
Antarctic
 
Aiding Computer Aided Drug Design
Aiding Computer Aided Drug DesignAiding Computer Aided Drug Design
Aiding Computer Aided Drug Design
 
Graphing quadratic equations
Graphing quadratic equationsGraphing quadratic equations
Graphing quadratic equations
 

Ähnlich wie Accelerating molecular dynamics simple tweaks to instant clusters

Introduction to Accelerators
Introduction to AcceleratorsIntroduction to Accelerators
Introduction to AcceleratorsDilum Bandara
 
GPU Accelerated Data Science with RAPIDS - ODSC West 2020
GPU Accelerated Data Science with RAPIDS - ODSC West 2020GPU Accelerated Data Science with RAPIDS - ODSC West 2020
GPU Accelerated Data Science with RAPIDS - ODSC West 2020John Zedlewski
 
digitaldesign-s20-lecture3b-fpga-afterlecture.pdf
digitaldesign-s20-lecture3b-fpga-afterlecture.pdfdigitaldesign-s20-lecture3b-fpga-afterlecture.pdf
digitaldesign-s20-lecture3b-fpga-afterlecture.pdfDuy-Hieu Bui
 
Kusto (Azure Data Explorer) Training for R&D - January 2019
Kusto (Azure Data Explorer) Training for R&D - January 2019 Kusto (Azure Data Explorer) Training for R&D - January 2019
Kusto (Azure Data Explorer) Training for R&D - January 2019 Tal Bar-Zvi
 
Systems@Scale 2021 BPF Performance Getting Started
Systems@Scale 2021 BPF Performance Getting StartedSystems@Scale 2021 BPF Performance Getting Started
Systems@Scale 2021 BPF Performance Getting StartedBrendan Gregg
 
Solving channel coding simulation and optimization problems using GPU
Solving channel coding simulation and optimization problems using GPUSolving channel coding simulation and optimization problems using GPU
Solving channel coding simulation and optimization problems using GPUUsatyuk Vasiliy
 
Interconnection Automation For All - Extended - MPS 2023
Interconnection Automation For All - Extended - MPS 2023Interconnection Automation For All - Extended - MPS 2023
Interconnection Automation For All - Extended - MPS 2023Chris Grundemann
 
Performance Wins with eBPF: Getting Started (2021)
Performance Wins with eBPF: Getting Started (2021)Performance Wins with eBPF: Getting Started (2021)
Performance Wins with eBPF: Getting Started (2021)Brendan Gregg
 
Nvidia in bioinformatics
Nvidia in bioinformaticsNvidia in bioinformatics
Nvidia in bioinformaticsShanker Trivedi
 
Intro to open source telemetry linux con 2016
Intro to open source telemetry   linux con 2016Intro to open source telemetry   linux con 2016
Intro to open source telemetry linux con 2016Matthew Broberg
 
Connected hardware for Software Engineers 101
Connected hardware for Software Engineers 101Connected hardware for Software Engineers 101
Connected hardware for Software Engineers 101Pance Cavkovski
 
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsRobotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsData Works MD
 
Laptop syllabus 1 month
Laptop syllabus 1 monthLaptop syllabus 1 month
Laptop syllabus 1 monthchiptroniks
 
Hadoop Query Performance Smackdown
Hadoop Query Performance SmackdownHadoop Query Performance Smackdown
Hadoop Query Performance SmackdownDataWorks Summit
 
Using GPUs to handle Big Data with Java by Adam Roberts.
Using GPUs to handle Big Data with Java by Adam Roberts.Using GPUs to handle Big Data with Java by Adam Roberts.
Using GPUs to handle Big Data with Java by Adam Roberts.J On The Beach
 
Laptop Chip level repairing(CPU section)
Laptop Chip level repairing(CPU section)Laptop Chip level repairing(CPU section)
Laptop Chip level repairing(CPU section)chiptroniks
 
laptop repairing institute
laptop repairing institutelaptop repairing institute
laptop repairing institutechiptroniks
 
laptop repairing institute
laptop repairing institutelaptop repairing institute
laptop repairing institutechiptroniks
 
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey GordeychikCODE BLUE
 

Ähnlich wie Accelerating molecular dynamics simple tweaks to instant clusters (20)

Introduction to Accelerators
Introduction to AcceleratorsIntroduction to Accelerators
Introduction to Accelerators
 
GPU Accelerated Data Science with RAPIDS - ODSC West 2020
GPU Accelerated Data Science with RAPIDS - ODSC West 2020GPU Accelerated Data Science with RAPIDS - ODSC West 2020
GPU Accelerated Data Science with RAPIDS - ODSC West 2020
 
digitaldesign-s20-lecture3b-fpga-afterlecture.pdf
digitaldesign-s20-lecture3b-fpga-afterlecture.pdfdigitaldesign-s20-lecture3b-fpga-afterlecture.pdf
digitaldesign-s20-lecture3b-fpga-afterlecture.pdf
 
Kusto (Azure Data Explorer) Training for R&D - January 2019
Kusto (Azure Data Explorer) Training for R&D - January 2019 Kusto (Azure Data Explorer) Training for R&D - January 2019
Kusto (Azure Data Explorer) Training for R&D - January 2019
 
Systems@Scale 2021 BPF Performance Getting Started
Systems@Scale 2021 BPF Performance Getting StartedSystems@Scale 2021 BPF Performance Getting Started
Systems@Scale 2021 BPF Performance Getting Started
 
Solving channel coding simulation and optimization problems using GPU
Solving channel coding simulation and optimization problems using GPUSolving channel coding simulation and optimization problems using GPU
Solving channel coding simulation and optimization problems using GPU
 
Interconnection Automation For All - Extended - MPS 2023
Interconnection Automation For All - Extended - MPS 2023Interconnection Automation For All - Extended - MPS 2023
Interconnection Automation For All - Extended - MPS 2023
 
Performance Wins with eBPF: Getting Started (2021)
Performance Wins with eBPF: Getting Started (2021)Performance Wins with eBPF: Getting Started (2021)
Performance Wins with eBPF: Getting Started (2021)
 
Nvidia in bioinformatics
Nvidia in bioinformaticsNvidia in bioinformatics
Nvidia in bioinformatics
 
Intro to open source telemetry linux con 2016
Intro to open source telemetry   linux con 2016Intro to open source telemetry   linux con 2016
Intro to open source telemetry linux con 2016
 
Connected hardware for Software Engineers 101
Connected hardware for Software Engineers 101Connected hardware for Software Engineers 101
Connected hardware for Software Engineers 101
 
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson KitsRobotics and Machine Learning: Working with NVIDIA Jetson Kits
Robotics and Machine Learning: Working with NVIDIA Jetson Kits
 
Laptop syllabus 1 month
Laptop syllabus 1 monthLaptop syllabus 1 month
Laptop syllabus 1 month
 
Hadoop Query Performance Smackdown
Hadoop Query Performance SmackdownHadoop Query Performance Smackdown
Hadoop Query Performance Smackdown
 
Using GPUs to handle Big Data with Java by Adam Roberts.
Using GPUs to handle Big Data with Java by Adam Roberts.Using GPUs to handle Big Data with Java by Adam Roberts.
Using GPUs to handle Big Data with Java by Adam Roberts.
 
Laptop Chip level repairing(CPU section)
Laptop Chip level repairing(CPU section)Laptop Chip level repairing(CPU section)
Laptop Chip level repairing(CPU section)
 
laptop repairing institute
laptop repairing institutelaptop repairing institute
laptop repairing institute
 
laptop repairing institute
laptop repairing institutelaptop repairing institute
laptop repairing institute
 
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik
[CB20] Vulnerabilities of Machine Learning Infrastructure by Sergey Gordeychik
 
Pclr syllabus 1 month
Pclr syllabus  1 monthPclr syllabus  1 month
Pclr syllabus 1 month
 

Kürzlich hochgeladen

Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfciinovamais
 
Class 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdfClass 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdfAyushMahapatra5
 
fourth grading exam for kindergarten in writing
fourth grading exam for kindergarten in writingfourth grading exam for kindergarten in writing
fourth grading exam for kindergarten in writingTeacherCyreneCayanan
 
Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Celine George
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfagholdier
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdfQucHHunhnh
 
Sanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfSanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfsanyamsingh5019
 
Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Disha Kariya
 
General AI for Medical Educators April 2024
General AI for Medical Educators April 2024General AI for Medical Educators April 2024
General AI for Medical Educators April 2024Janet Corral
 
Measures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDMeasures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDThiyagu K
 
Disha NEET Physics Guide for classes 11 and 12.pdf
Disha NEET Physics Guide for classes 11 and 12.pdfDisha NEET Physics Guide for classes 11 and 12.pdf
Disha NEET Physics Guide for classes 11 and 12.pdfchloefrazer622
 
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdfBASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdfSoniaTolstoy
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityGeoBlogs
 
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...christianmathematics
 
9548086042 for call girls in Indira Nagar with room service
9548086042  for call girls in Indira Nagar  with room service9548086042  for call girls in Indira Nagar  with room service
9548086042 for call girls in Indira Nagar with room servicediscovermytutordmt
 
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in DelhiRussian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhikauryashika82
 
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxSOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxiammrhaywood
 
Beyond the EU: DORA and NIS 2 Directive's Global Impact
Beyond the EU: DORA and NIS 2 Directive's Global ImpactBeyond the EU: DORA and NIS 2 Directive's Global Impact
Beyond the EU: DORA and NIS 2 Directive's Global ImpactPECB
 

Kürzlich hochgeladen (20)

Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
Class 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdfClass 11th Physics NEET formula sheet pdf
Class 11th Physics NEET formula sheet pdf
 
fourth grading exam for kindergarten in writing
fourth grading exam for kindergarten in writingfourth grading exam for kindergarten in writing
fourth grading exam for kindergarten in writing
 
Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17Advanced Views - Calendar View in Odoo 17
Advanced Views - Calendar View in Odoo 17
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdf
 
1029 - Danh muc Sach Giao Khoa 10 . pdf
1029 -  Danh muc Sach Giao Khoa 10 . pdf1029 -  Danh muc Sach Giao Khoa 10 . pdf
1029 - Danh muc Sach Giao Khoa 10 . pdf
 
Sanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdfSanyam Choudhary Chemistry practical.pdf
Sanyam Choudhary Chemistry practical.pdf
 
Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..
 
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
 
Mattingly "AI & Prompt Design: The Basics of Prompt Design"
Mattingly "AI & Prompt Design: The Basics of Prompt Design"Mattingly "AI & Prompt Design: The Basics of Prompt Design"
Mattingly "AI & Prompt Design: The Basics of Prompt Design"
 
General AI for Medical Educators April 2024
General AI for Medical Educators April 2024General AI for Medical Educators April 2024
General AI for Medical Educators April 2024
 
Measures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDMeasures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SD
 
Disha NEET Physics Guide for classes 11 and 12.pdf
Disha NEET Physics Guide for classes 11 and 12.pdfDisha NEET Physics Guide for classes 11 and 12.pdf
Disha NEET Physics Guide for classes 11 and 12.pdf
 
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdfBASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activity
 
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
 
9548086042 for call girls in Indira Nagar with room service
9548086042  for call girls in Indira Nagar  with room service9548086042  for call girls in Indira Nagar  with room service
9548086042 for call girls in Indira Nagar with room service
 
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in DelhiRussian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
 
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptxSOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
SOCIAL AND HISTORICAL CONTEXT - LFTVD.pptx
 
Beyond the EU: DORA and NIS 2 Directive's Global Impact
Beyond the EU: DORA and NIS 2 Directive's Global ImpactBeyond the EU: DORA and NIS 2 Directive's Global Impact
Beyond the EU: DORA and NIS 2 Directive's Global Impact
 

Accelerating molecular dynamics simple tweaks to instant clusters

  • 1. Accelerating MD..Simple Tweaks and Instant Clusters Mohd Shahir Shamsir Bioinformatics Research Group (BIRG) Faculty of Biosciences & Bioengineering Universiti Teknologi Malaysia INSPIRING CREATIVE AND INNOVATIVE MINDS
  • 2. Summary Introduction to BIRG MD: What, Why and How Improving performance Simple Tweaks… Instant MD cluster– birgHPC Short video demo
  • 3. Bioinformatics Research Group (BIRG)Faculty of Biosciences & Bioengineering Just Google Us…
  • 4. MD… What? Why? How? COVERED!!
  • 5. Performance of MD? Speed, speed, speed… Supercomputer IBM Roadrunner ~368 Flops Nankai Star 3.7 ns/day on 32 nodes (DPPC) HPCx 5.2 ns/day on 64 nodes (DPPC) New platform Cell-BE: GROMACS, 15x to Pentium 3.0GHz GPU: NAMD, 4 GPU = 16 CPUs
  • 6.
  • 7.
  • 8. Microwulf cluster 26 Gflops, $2500, 11" x 12" x 17", airline overhead baggage compliant
  • 9. Simple Tweaks INSPIRING CREATIVE AND INNOVATIVE MINDS
  • 10. Tweaks for MD? Hardware ↑ = performance ↑ = $$$ ↑ OR Tweak Beowulf = performance ↑ = $$$ Pre-compiled vs self-compiled MPI libraries Test beds: 3 nodes GridMACS, 7 nodes Beowulf, 1 reference machine
  • 12. Beowulf OpenMPIvs MPICH2 (pre) Winner MPICH2
  • 13. Pre and Self Compiled MPI Self OpenMPI = MPICH2
  • 14. What we found Single machine 66 % improvement Parallel environment 64 % improvement Compilation, software chosen affect performance
  • 15. Instant MD, anyone? INSPIRING CREATIVE AND INNOVATIVE MINDS
  • 16. Instant MD cluster Lots of under utilised computers in labs Idle mode after office hours, holidays, etc.
  • 17. Instant MD cluster MD, parallel computing = high computing resources Solution? Supercomputers Dedicated computing cluster Problems? $$$ ??? (I don’t know this, I don’t know that…)
  • 18. A + B = C What is A? Existing computers LAN connected, PXE-boot capable, CDROM/USB What is B? Linux Live CD Auto configuration What is C? Instant, out-of-the-box computing cluster!
  • 19. birgHPC Free, open-source Linux distribution Based on PelicanHPC & Debian Live GROMACS, NAMD, mpiBLAST, ClustalW-MPI, PyMol, VMD Auto cluster config MPICH2 & OpenMPI Auto slots detections Ganglia monitoring Simple interface for job submission
  • 27. Conclusion birgHPC instant cluster conversion Bioinformatics tools Auto configurations http://birg1.fbb.utm.my/birghpc ISOs guide
  • 28. Acknowledgements Chew Teong Han - Alchemist Farizuawana – Graphics Joyce Tan – Testing Funding from you via LHDN via MOSTI Michael Creel for Pelican HPC
  • 29. FAQs Boot sequence Head node -> run birgHPC_setup -> follow instructions -> boot compute node -> script on head node will show # of nodes detected, confirm -> done Headless compute nodes (no monitor) Have to get a monitor -> set boot sequence to netboot -> done
  • 30. FAQs How to know compute nodes is up Follow the birgHPC boot sequence -> the birgHPC_setup script will show # nodes detected Cannot netboot Try http://etherboot.org/wiki/start Heterogeneous PCs ok? Ok (Thanks Micheal Creel) If 32 bit + 64 bit, use 32 bit PC as head node
  • 31. FAQs Status monitoring Yes -> web browser -> localhost -> Ganglia Monitoring What will displayed on compute node? Just a simple login terminal with some warning not to use the nodes, etc Limitations? RAM, RAM, RAM … Everything is loaded to RAM, hence HDD size = RAM size
  • 32. FAQs Headnode criteria Preferably big RAM because of shared folder (/home) 2 Ethernet port if you want internet connection (still work if you got only one eth) Guide? http://birg1.fbb.utm.my/birghpc Forum? No, but can always refer PelicanHPC forums
  • 33. FAQs Multi users? No, designed for single user Future release, maybe SGE or PBS Installed on hardiskpermenantly? Not tested, technically possible, Google Performance On par with hardisk-installed cluster (tested up to 6 nodes)
  • 34. FAQs I cannot boot from CD Refer user guide -> convert CD ISO to USB drive image -> boot from USB Can I use birgHPC along with existing DHCP Preferably no, DHCP will distribute IP and birgHPC head will distribute IP, causing confusion of IP addresses Alternative, boot DHCP as head node OR unplug DHCP, use another PC as head node
  • 35. FAQs birgHPC criteria (PCs = compute, server = head)