SlideShare a Scribd company logo
Docker and
the K computer
Bryzgalov Peter @ HPC Usability Research Team
RIKEN AICS
petr.bryzgalov@riken.jp
2015.08.19 Docker Meetup Tokyo #5
K computer
K computer
Nodes > 80 000
Cores > 700 000
Linpack Performance 10,510 TFlop/s
Docker merits for HPC
• Runtime — customise software stack, easy to
test new software, more efficient resource
management, administrators do not have to
worry about security patches for user-space
software, etc.
• Compile time — eliminate dependencies on
compilers and libraries versions, etc.
4
Docker merits for HPC
• Runtime — customise software stack, easy to
test new software, more efficient resource
management, administrators do not have to
worry about security patches for user-space
software, etc.
• Compile time — eliminate dependencies on
compilers and libraries versions, etc.
5
Docker at Compile time
• Install cross-compiler in container
6
Docker at Compile time
• Install cross-compiler in container
Why cross-compiler?
7
K architecture
• Login nodes and Compute nodes
• Compilation on Login nodes
• Execution on Compute nodes
8
K architecture
• Login nodes and Compute nodes
• Compilation on Login nodes
• Execution on Compute nodes
• Login nodes — x86_64
• Compute nodes — sparc64fx
9
Docker at Compile time
Done :
• Installed cross-compilers in container.
• Developed system for using it online.
10
Docker at Compile time
Workflow :
• User runs a script instead of “make”
command on his computer.
• Source code folder automatically mounted
into container with compilers.
• “make” command automatically executed in
container.
11
Docker at Compile time
1.Container with cross-compilers runs on server.
12
Docker at Compile time
2.Source code folder automatically mounted
into container.
13
Docker at Compile time
3.“make” command automatically executed in
container.
14
Docker at Compile time
Result :
• Binaries are built and saved on user
computer.
15
Docker at Compile time
Issues :
• Compiler license.
• Customise software stack (instal third-party
software in container).
16
Docker merits for HPC
• Runtime — customise software stack, easy to
test new software, more efficient resource
management, administrators do not have to
worry about security patches for user-space
software, etc.
• Compile time — eliminate dependencies on
compilers and libraries versions, etc.
17
Docker at Runtime
• Install Docker on Compute nodes
18
Docker at Runtime
• Install Docker on Compute nodes
• K compute nodes:
19
Docker at Runtime
• Install Docker on Compute nodes
• K compute nodes:
20
Docker at Runtime
• Install Docker on Compute nodes
• K compute nodes:
20
CPU sparc64fx
OS kernel 2.6.25
Docker at Runtime
• Install Docker on Compute nodes
• K compute nodes:
• Docker requires 3.8 (3.10 for latest version)
21
CPU sparc64fx
OS kernel 2.6.25
Update OS kernel
on Compute Nodes
• 2.6.25 -> 3.8
22
Update OS kernel
on Compute Nodes
• 2.6.25 -> 3.8
• OS kernel on K CN:

vanilla 2.6.25.8 + Fujitsu patch
• Apply Fujitsu patch to vanilla 3.8
23
Patch vanilla kernel
• Take a new vanilla kernel source code and
apply Fujitsu patch
24
Patch vanilla kernel
• Take a new vanilla kernel source code and
apply Fujitsu patch
• Patch fails starting from 2.6.26
25
Patch vanilla kernel
• How many files need to be fixed?
26
Patch vanilla kernel
27
Patch vanilla kernel
28
84f i l e s
Patch vanilla kernel
29
We need
help!
Maeda Toshiyuki Team leader
Bryzgalov Peter Researcher
Thank you!
HPC Usability Research Team
www.hpcu.aics.riken.jp
comp.photo777.org
petr.bryzgalov@riken.jp
This presentation has more slides but I don’t have time to show them.
Docker merits for HPC
• Runtime — customise software stack, easy to
test new software, more efficient resource
management, administrators do not have to
worry about security patches for user-space
software, etc.
• Compile time — eliminate dependencies on
compilers and libraries versions.
31
Docker at Compile time
K workflow
• Edit Source code
• Cross-compile on Login Nodes
• Execute on Compute Nodes
32
Docker at Compile time
K workflow
• Edit Source code
• Cross-compile on Login Nodes
• Execute on Compute Nodes
Dependencies
• Compilers, libraries, … — one environment for all, it changes
over time, application that compiles now may not compile in
the future (like ABYSS).
33
Docker at Compile time
K workflow
• Edit Source code
• Cross-compile on Login Nodes
• Execute on Compute Nodes
Dependencies
• Compilers, libraries, … — one environment for all, it changes
over time, application that compiles now may not compile in
the future (like ABYSS).
33
Cross-compile in Docker container
Compile in container
• Run container on User Computer
• Run container on Login Node
• Run container on Server
34
Compile in container
• Run container on User Computer
• Run container on Login Node
• Run container on Server
34
Compiler licence limitations
Compile in container
• Run container on User Computer
• Run container on Login Node
• Run container on Server
34
Compiler licence limitations
“K” administration policies
Compile in container
• Run container on User Computer
• Run container on Login Node
• Run container on Server
35
Run compiler on server
in Docker container
Requirements
• Multiple environments (containers)
• Multiple users
• One machine
• Limited resources — stop containers when they are not used
Tasks
• Copy source code into containers
• Copy binaries from containers
36
Docker-based IaaS
Overview
• One host multi tenant IaaS
• One user — one container
• Users login with SSH
• Mount local folders into containers with SSHFS
37
github.com/pyotr777/dockerIaaSTools
Docker-based IaaS
Overview
38
github.com/pyotr777/dockerIaaSTools
Docker-based IaaS
Application
Challenge
• K-scope — tool for Fortran program static analysis.
• Depends on Omni XMP compiler front-end, which produces
AST (XML files).
• Omni XMP depends on many packages and is hard to install.
39
omni-compiler.org/xcalablemp.html
www.aics.riken.jp/ungi/soft/kscope
github.com/pyotr777/dockerIaaSTools
Docker-based IaaS
Application
Our solution
• Script that mounts local source code folder in container and
executes “make” command inside the container.
• K-scope executes the script instead of make command.
40
github.com/pyotr777/dockerIaaSTools
Docker-based IaaS
Application
Our solution
41
Docker-based IaaS
Application
Our solution
• Script that mounts local source code folder in container and
executes “make” command inside the container.
• K-scope executes the script instead of make command.
• After script finishes Intermediate code files are in local source
folder — same effect as calling “make” command locally.
42
github.com/pyotr777/dockerIaaSTools

More Related Content

What's hot

What's hot (20)

Kubernetes in Docker
Kubernetes in DockerKubernetes in Docker
Kubernetes in Docker
 
Building Your Docker Swarm Tech Stack
Building Your Docker Swarm Tech StackBuilding Your Docker Swarm Tech Stack
Building Your Docker Swarm Tech Stack
 
DCEU 18: State of the Docker Engine
DCEU 18: State of the Docker EngineDCEU 18: State of the Docker Engine
DCEU 18: State of the Docker Engine
 
OSCON: System software goes weird
OSCON: System software goes weirdOSCON: System software goes weird
OSCON: System software goes weird
 
Kubernetes deep dive - - Huawei 2015-10
Kubernetes deep dive - - Huawei 2015-10Kubernetes deep dive - - Huawei 2015-10
Kubernetes deep dive - - Huawei 2015-10
 
DockerCon EU 2015: The Latest in Docker Engine
DockerCon EU 2015: The Latest in Docker EngineDockerCon EU 2015: The Latest in Docker Engine
DockerCon EU 2015: The Latest in Docker Engine
 
Docker Online Meetup: Announcing Docker CE + EE
Docker Online Meetup: Announcing Docker CE + EEDocker Online Meetup: Announcing Docker CE + EE
Docker Online Meetup: Announcing Docker CE + EE
 
DockerCon EU 2015: Docker and PCI-DSS - Lessons learned in a security sensiti...
DockerCon EU 2015: Docker and PCI-DSS - Lessons learned in a security sensiti...DockerCon EU 2015: Docker and PCI-DSS - Lessons learned in a security sensiti...
DockerCon EU 2015: Docker and PCI-DSS - Lessons learned in a security sensiti...
 
Docker Platform Internals: Taking runtimes and image creation to the next lev...
Docker Platform Internals: Taking runtimes and image creation to the next lev...Docker Platform Internals: Taking runtimes and image creation to the next lev...
Docker Platform Internals: Taking runtimes and image creation to the next lev...
 
Why I wish I'd Heard of Docker when I was 12 - Finnian Anderson
Why I wish I'd Heard of Docker when I was 12 - Finnian AndersonWhy I wish I'd Heard of Docker when I was 12 - Finnian Anderson
Why I wish I'd Heard of Docker when I was 12 - Finnian Anderson
 
Docker?!?! But I'm a SysAdmin
Docker?!?! But I'm a SysAdminDocker?!?! But I'm a SysAdmin
Docker?!?! But I'm a SysAdmin
 
Docker 1.11 Meetup: Containerd and runc, by Arnaud Porterie and Michael Crosby
Docker 1.11 Meetup: Containerd and runc, by Arnaud Porterie and Michael Crosby Docker 1.11 Meetup: Containerd and runc, by Arnaud Porterie and Michael Crosby
Docker 1.11 Meetup: Containerd and runc, by Arnaud Porterie and Michael Crosby
 
Introduction to OCI Image Technologies Serving Container
Introduction to OCI Image Technologies Serving ContainerIntroduction to OCI Image Technologies Serving Container
Introduction to OCI Image Technologies Serving Container
 
DCSF 19 Kubernetes and Container Storage Interface Update
DCSF 19 Kubernetes and Container Storage Interface UpdateDCSF 19 Kubernetes and Container Storage Interface Update
DCSF 19 Kubernetes and Container Storage Interface Update
 
Persistent Data Storage for Docker Containers by Andre Moruga
Persistent Data Storage for Docker Containers by Andre MorugaPersistent Data Storage for Docker Containers by Andre Moruga
Persistent Data Storage for Docker Containers by Andre Moruga
 
DockerCon EU 2015: What's New with Docker Trusted Registry
DockerCon EU 2015: What's New with Docker Trusted RegistryDockerCon EU 2015: What's New with Docker Trusted Registry
DockerCon EU 2015: What's New with Docker Trusted Registry
 
Dockercon 16 Wrap-up (Docker for Mac and Win, Docker 1.12, Swarm Mode, etc.)
Dockercon 16 Wrap-up (Docker for Mac and Win, Docker 1.12, Swarm Mode, etc.)Dockercon 16 Wrap-up (Docker for Mac and Win, Docker 1.12, Swarm Mode, etc.)
Dockercon 16 Wrap-up (Docker for Mac and Win, Docker 1.12, Swarm Mode, etc.)
 
DCSF 19 Online Feature Extraction and Event Generation for Computer-Animal In...
DCSF 19 Online Feature Extraction and Event Generation for Computer-Animal In...DCSF 19 Online Feature Extraction and Event Generation for Computer-Animal In...
DCSF 19 Online Feature Extraction and Event Generation for Computer-Animal In...
 
DockerCon 2017 - Cilium - Network and Application Security with BPF and XDP
DockerCon 2017 - Cilium - Network and Application Security with BPF and XDPDockerCon 2017 - Cilium - Network and Application Security with BPF and XDP
DockerCon 2017 - Cilium - Network and Application Security with BPF and XDP
 
Tales of Training: Scaling CodeLabs with Swarm Mode and Docker-Compose
Tales of Training: Scaling CodeLabs with Swarm Mode and Docker-ComposeTales of Training: Scaling CodeLabs with Swarm Mode and Docker-Compose
Tales of Training: Scaling CodeLabs with Swarm Mode and Docker-Compose
 

Viewers also liked

HPC Application Profiling and Analysis
HPC Application Profiling and AnalysisHPC Application Profiling and Analysis
HPC Application Profiling and Analysis
Rishi Pathak
 
Symposium on HPC Applications – IIT Kanpur
Symposium on HPC Applications – IIT KanpurSymposium on HPC Applications – IIT Kanpur
Symposium on HPC Applications – IIT Kanpur
Rishi Pathak
 
HPC Application Profiling & Analysis
HPC Application Profiling & AnalysisHPC Application Profiling & Analysis
HPC Application Profiling & Analysis
Rishi Pathak
 
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
Geoffrey Fox
 
Oscon London 2016 - Docker from Development to Production
Oscon London 2016 - Docker from Development to ProductionOscon London 2016 - Docker from Development to Production
Oscon London 2016 - Docker from Development to Production
Patrick Chanezon
 

Viewers also liked (20)

AcuSolve Optimizations for Scale - Hpc advisory council
AcuSolve Optimizations for Scale   - Hpc advisory councilAcuSolve Optimizations for Scale   - Hpc advisory council
AcuSolve Optimizations for Scale - Hpc advisory council
 
AWS for Semiconductor and Electronics Design | Hsinchu, April 10
AWS for Semiconductor and Electronics Design | Hsinchu, April 10AWS for Semiconductor and Electronics Design | Hsinchu, April 10
AWS for Semiconductor and Electronics Design | Hsinchu, April 10
 
HPC in AWS - Technical Workshop
HPC in AWS - Technical WorkshopHPC in AWS - Technical Workshop
HPC in AWS - Technical Workshop
 
Performance Optimization of HPC Applications: From Hardware to Source Code
Performance Optimization of HPC Applications: From Hardware to Source CodePerformance Optimization of HPC Applications: From Hardware to Source Code
Performance Optimization of HPC Applications: From Hardware to Source Code
 
HPC Application Profiling and Analysis
HPC Application Profiling and AnalysisHPC Application Profiling and Analysis
HPC Application Profiling and Analysis
 
Symposium on HPC Applications – IIT Kanpur
Symposium on HPC Applications – IIT KanpurSymposium on HPC Applications – IIT Kanpur
Symposium on HPC Applications – IIT Kanpur
 
51 Use Cases and implications for HPC & Apache Big Data Stack
51 Use Cases and implications for HPC & Apache Big Data Stack51 Use Cases and implications for HPC & Apache Big Data Stack
51 Use Cases and implications for HPC & Apache Big Data Stack
 
HPC Application Profiling & Analysis
HPC Application Profiling & AnalysisHPC Application Profiling & Analysis
HPC Application Profiling & Analysis
 
HPC Best Practices: Application Performance Optimization
HPC Best Practices: Application Performance OptimizationHPC Best Practices: Application Performance Optimization
HPC Best Practices: Application Performance Optimization
 
IDC Perspectives on Big Data Outside of HPC
IDC Perspectives on Big Data Outside of HPCIDC Perspectives on Big Data Outside of HPC
IDC Perspectives on Big Data Outside of HPC
 
Autodesk maya presentation
Autodesk maya presentationAutodesk maya presentation
Autodesk maya presentation
 
Building Distributed Systems without Docker, Using Docker Plumbing Projects -...
Building Distributed Systems without Docker, Using Docker Plumbing Projects -...Building Distributed Systems without Docker, Using Docker Plumbing Projects -...
Building Distributed Systems without Docker, Using Docker Plumbing Projects -...
 
Intro to maya
Intro to mayaIntro to maya
Intro to maya
 
Big data analytics use case and software
Big data analytics use case and softwareBig data analytics use case and software
Big data analytics use case and software
 
Big Data and High Performance Computing Solutions in the AWS Cloud
Big Data and High Performance Computing Solutions in the AWS CloudBig Data and High Performance Computing Solutions in the AWS Cloud
Big Data and High Performance Computing Solutions in the AWS Cloud
 
Docker Platform and Ecosystem
Docker Platform and EcosystemDocker Platform and Ecosystem
Docker Platform and Ecosystem
 
AWS Cloud for HPC and Big Data
AWS Cloud for HPC and Big DataAWS Cloud for HPC and Big Data
AWS Cloud for HPC and Big Data
 
Intro to High Performance Computing in the AWS Cloud
Intro to High Performance Computing in the AWS CloudIntro to High Performance Computing in the AWS Cloud
Intro to High Performance Computing in the AWS Cloud
 
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
Multi-faceted Classification of Big Data Use Cases and Proposed Architecture ...
 
Oscon London 2016 - Docker from Development to Production
Oscon London 2016 - Docker from Development to ProductionOscon London 2016 - Docker from Development to Production
Oscon London 2016 - Docker from Development to Production
 

Similar to Docker and the K computer

Docker Hub: Past, Present and Future by Ken Cochrane & BC Wong
Docker Hub: Past, Present and Future by Ken Cochrane & BC WongDocker Hub: Past, Present and Future by Ken Cochrane & BC Wong
Docker Hub: Past, Present and Future by Ken Cochrane & BC Wong
Docker, Inc.
 
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
IBM France Lab
 

Similar to Docker and the K computer (20)

Efficient Parallel Testing with Docker by Laura Frank
Efficient Parallel Testing with Docker by Laura FrankEfficient Parallel Testing with Docker by Laura Frank
Efficient Parallel Testing with Docker by Laura Frank
 
Continuous Integration with Cloud Foundry Concourse and Docker on OpenPOWER
Continuous Integration with Cloud Foundry Concourse and Docker on OpenPOWERContinuous Integration with Cloud Foundry Concourse and Docker on OpenPOWER
Continuous Integration with Cloud Foundry Concourse and Docker on OpenPOWER
 
Adf with docker
Adf with dockerAdf with docker
Adf with docker
 
Docker Global Hack Day #3
Docker Global Hack Day #3 Docker Global Hack Day #3
Docker Global Hack Day #3
 
A Tail of Two Containers: How docker made ci great again
A Tail of Two Containers: How docker made ci great againA Tail of Two Containers: How docker made ci great again
A Tail of Two Containers: How docker made ci great again
 
Docker Hub: Past, Present and Future by Ken Cochrane & BC Wong
Docker Hub: Past, Present and Future by Ken Cochrane & BC WongDocker Hub: Past, Present and Future by Ken Cochrane & BC Wong
Docker Hub: Past, Present and Future by Ken Cochrane & BC Wong
 
Bare-metal, Docker Containers, and Virtualization: The Growing Choices for Cl...
Bare-metal, Docker Containers, and Virtualization: The Growing Choices for Cl...Bare-metal, Docker Containers, and Virtualization: The Growing Choices for Cl...
Bare-metal, Docker Containers, and Virtualization: The Growing Choices for Cl...
 
Webinar: From Development to Production with Docker and MongoDB
Webinar: From Development to Production with Docker and MongoDBWebinar: From Development to Production with Docker and MongoDB
Webinar: From Development to Production with Docker and MongoDB
 
DCEU 18: Building Your Development Pipeline
DCEU 18: Building Your Development PipelineDCEU 18: Building Your Development Pipeline
DCEU 18: Building Your Development Pipeline
 
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
IBM Bluemix Paris Meetup #14 - Le Village by CA - 20160413 - Introduction à D...
 
DockerCon 2015: Docker Engine Breakout Session
DockerCon 2015: Docker Engine Breakout SessionDockerCon 2015: Docker Engine Breakout Session
DockerCon 2015: Docker Engine Breakout Session
 
DockerCon US 2015 - Engine Breakout Session
DockerCon US 2015 - Engine Breakout SessionDockerCon US 2015 - Engine Breakout Session
DockerCon US 2015 - Engine Breakout Session
 
Containerization using docker and its applications
Containerization using docker and its applicationsContainerization using docker and its applications
Containerization using docker and its applications
 
Containerization using docker and its applications
Containerization using docker and its applicationsContainerization using docker and its applications
Containerization using docker and its applications
 
Django and Docker
Django and DockerDjango and Docker
Django and Docker
 
Docker at Djangocon 2013 | Talk by Ken Cochrane
Docker at Djangocon 2013 | Talk by Ken CochraneDocker at Djangocon 2013 | Talk by Ken Cochrane
Docker at Djangocon 2013 | Talk by Ken Cochrane
 
Building Efficient Parallel Testing Platforms with Docker
Building Efficient Parallel Testing Platforms with DockerBuilding Efficient Parallel Testing Platforms with Docker
Building Efficient Parallel Testing Platforms with Docker
 
5 Ways to Secure Your Containers for Docker and Beyond
5 Ways to Secure Your Containers for Docker and Beyond5 Ways to Secure Your Containers for Docker and Beyond
5 Ways to Secure Your Containers for Docker and Beyond
 
Docker Orchestration: Welcome to the Jungle! Devoxx & Docker Meetup Tour Nov ...
Docker Orchestration: Welcome to the Jungle! Devoxx & Docker Meetup Tour Nov ...Docker Orchestration: Welcome to the Jungle! Devoxx & Docker Meetup Tour Nov ...
Docker Orchestration: Welcome to the Jungle! Devoxx & Docker Meetup Tour Nov ...
 
Docker Containers Deep Dive
Docker Containers Deep DiveDocker Containers Deep Dive
Docker Containers Deep Dive
 

Recently uploaded

AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
Alluxio, Inc.
 

Recently uploaded (20)

AI/ML Infra Meetup | Reducing Prefill for LLM Serving in RAG
AI/ML Infra Meetup | Reducing Prefill for LLM Serving in RAGAI/ML Infra Meetup | Reducing Prefill for LLM Serving in RAG
AI/ML Infra Meetup | Reducing Prefill for LLM Serving in RAG
 
Breaking the Code : A Guide to WhatsApp Business API.pdf
Breaking the Code : A Guide to WhatsApp Business API.pdfBreaking the Code : A Guide to WhatsApp Business API.pdf
Breaking the Code : A Guide to WhatsApp Business API.pdf
 
OpenFOAM solver for Helmholtz equation, helmholtzFoam / helmholtzBubbleFoam
OpenFOAM solver for Helmholtz equation, helmholtzFoam / helmholtzBubbleFoamOpenFOAM solver for Helmholtz equation, helmholtzFoam / helmholtzBubbleFoam
OpenFOAM solver for Helmholtz equation, helmholtzFoam / helmholtzBubbleFoam
 
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
Gamify Your Mind; The Secret Sauce to Delivering Success, Continuously Improv...
 
Studiovity film pre-production and screenwriting software
Studiovity film pre-production and screenwriting softwareStudiovity film pre-production and screenwriting software
Studiovity film pre-production and screenwriting software
 
AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
AI/ML Infra Meetup | Improve Speed and GPU Utilization for Model Training & S...
 
Exploring Innovations in Data Repository Solutions - Insights from the U.S. G...
Exploring Innovations in Data Repository Solutions - Insights from the U.S. G...Exploring Innovations in Data Repository Solutions - Insights from the U.S. G...
Exploring Innovations in Data Repository Solutions - Insights from the U.S. G...
 
Designing for Privacy in Amazon Web Services
Designing for Privacy in Amazon Web ServicesDesigning for Privacy in Amazon Web Services
Designing for Privacy in Amazon Web Services
 
top nidhi software solution freedownload
top nidhi software solution freedownloadtop nidhi software solution freedownload
top nidhi software solution freedownload
 
Innovating Inference - Remote Triggering of Large Language Models on HPC Clus...
Innovating Inference - Remote Triggering of Large Language Models on HPC Clus...Innovating Inference - Remote Triggering of Large Language Models on HPC Clus...
Innovating Inference - Remote Triggering of Large Language Models on HPC Clus...
 
AI/ML Infra Meetup | Perspective on Deep Learning Framework
AI/ML Infra Meetup | Perspective on Deep Learning FrameworkAI/ML Infra Meetup | Perspective on Deep Learning Framework
AI/ML Infra Meetup | Perspective on Deep Learning Framework
 
Field Employee Tracking System| MiTrack App| Best Employee Tracking Solution|...
Field Employee Tracking System| MiTrack App| Best Employee Tracking Solution|...Field Employee Tracking System| MiTrack App| Best Employee Tracking Solution|...
Field Employee Tracking System| MiTrack App| Best Employee Tracking Solution|...
 
Vitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume MontevideoVitthal Shirke Microservices Resume Montevideo
Vitthal Shirke Microservices Resume Montevideo
 
Paketo Buildpacks : la meilleure façon de construire des images OCI? DevopsDa...
Paketo Buildpacks : la meilleure façon de construire des images OCI? DevopsDa...Paketo Buildpacks : la meilleure façon de construire des images OCI? DevopsDa...
Paketo Buildpacks : la meilleure façon de construire des images OCI? DevopsDa...
 
A Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdfA Comprehensive Look at Generative AI in Retail App Testing.pdf
A Comprehensive Look at Generative AI in Retail App Testing.pdf
 
How Recreation Management Software Can Streamline Your Operations.pptx
How Recreation Management Software Can Streamline Your Operations.pptxHow Recreation Management Software Can Streamline Your Operations.pptx
How Recreation Management Software Can Streamline Your Operations.pptx
 
Agnieszka Andrzejewska - BIM School Course in Kraków
Agnieszka Andrzejewska - BIM School Course in KrakówAgnieszka Andrzejewska - BIM School Course in Kraków
Agnieszka Andrzejewska - BIM School Course in Kraków
 
Developing Distributed High-performance Computing Capabilities of an Open Sci...
Developing Distributed High-performance Computing Capabilities of an Open Sci...Developing Distributed High-performance Computing Capabilities of an Open Sci...
Developing Distributed High-performance Computing Capabilities of an Open Sci...
 
Cyaniclab : Software Development Agency Portfolio.pdf
Cyaniclab : Software Development Agency Portfolio.pdfCyaniclab : Software Development Agency Portfolio.pdf
Cyaniclab : Software Development Agency Portfolio.pdf
 
BoxLang: Review our Visionary Licenses of 2024
BoxLang: Review our Visionary Licenses of 2024BoxLang: Review our Visionary Licenses of 2024
BoxLang: Review our Visionary Licenses of 2024
 

Docker and the K computer

  • 1. Docker and the K computer Bryzgalov Peter @ HPC Usability Research Team RIKEN AICS petr.bryzgalov@riken.jp 2015.08.19 Docker Meetup Tokyo #5
  • 3. K computer Nodes > 80 000 Cores > 700 000 Linpack Performance 10,510 TFlop/s
  • 4. Docker merits for HPC • Runtime — customise software stack, easy to test new software, more efficient resource management, administrators do not have to worry about security patches for user-space software, etc. • Compile time — eliminate dependencies on compilers and libraries versions, etc. 4
  • 5. Docker merits for HPC • Runtime — customise software stack, easy to test new software, more efficient resource management, administrators do not have to worry about security patches for user-space software, etc. • Compile time — eliminate dependencies on compilers and libraries versions, etc. 5
  • 6. Docker at Compile time • Install cross-compiler in container 6
  • 7. Docker at Compile time • Install cross-compiler in container Why cross-compiler? 7
  • 8. K architecture • Login nodes and Compute nodes • Compilation on Login nodes • Execution on Compute nodes 8
  • 9. K architecture • Login nodes and Compute nodes • Compilation on Login nodes • Execution on Compute nodes • Login nodes — x86_64 • Compute nodes — sparc64fx 9
  • 10. Docker at Compile time Done : • Installed cross-compilers in container. • Developed system for using it online. 10
  • 11. Docker at Compile time Workflow : • User runs a script instead of “make” command on his computer. • Source code folder automatically mounted into container with compilers. • “make” command automatically executed in container. 11
  • 12. Docker at Compile time 1.Container with cross-compilers runs on server. 12
  • 13. Docker at Compile time 2.Source code folder automatically mounted into container. 13
  • 14. Docker at Compile time 3.“make” command automatically executed in container. 14
  • 15. Docker at Compile time Result : • Binaries are built and saved on user computer. 15
  • 16. Docker at Compile time Issues : • Compiler license. • Customise software stack (instal third-party software in container). 16
  • 17. Docker merits for HPC • Runtime — customise software stack, easy to test new software, more efficient resource management, administrators do not have to worry about security patches for user-space software, etc. • Compile time — eliminate dependencies on compilers and libraries versions, etc. 17
  • 18. Docker at Runtime • Install Docker on Compute nodes 18
  • 19. Docker at Runtime • Install Docker on Compute nodes • K compute nodes: 19
  • 20. Docker at Runtime • Install Docker on Compute nodes • K compute nodes: 20
  • 21. Docker at Runtime • Install Docker on Compute nodes • K compute nodes: 20 CPU sparc64fx OS kernel 2.6.25
  • 22. Docker at Runtime • Install Docker on Compute nodes • K compute nodes: • Docker requires 3.8 (3.10 for latest version) 21 CPU sparc64fx OS kernel 2.6.25
  • 23. Update OS kernel on Compute Nodes • 2.6.25 -> 3.8 22
  • 24. Update OS kernel on Compute Nodes • 2.6.25 -> 3.8 • OS kernel on K CN:
 vanilla 2.6.25.8 + Fujitsu patch • Apply Fujitsu patch to vanilla 3.8 23
  • 25. Patch vanilla kernel • Take a new vanilla kernel source code and apply Fujitsu patch 24
  • 26. Patch vanilla kernel • Take a new vanilla kernel source code and apply Fujitsu patch • Patch fails starting from 2.6.26 25
  • 27. Patch vanilla kernel • How many files need to be fixed? 26
  • 31. Maeda Toshiyuki Team leader Bryzgalov Peter Researcher Thank you! HPC Usability Research Team www.hpcu.aics.riken.jp comp.photo777.org petr.bryzgalov@riken.jp This presentation has more slides but I don’t have time to show them.
  • 32. Docker merits for HPC • Runtime — customise software stack, easy to test new software, more efficient resource management, administrators do not have to worry about security patches for user-space software, etc. • Compile time — eliminate dependencies on compilers and libraries versions. 31
  • 33. Docker at Compile time K workflow • Edit Source code • Cross-compile on Login Nodes • Execute on Compute Nodes 32
  • 34. Docker at Compile time K workflow • Edit Source code • Cross-compile on Login Nodes • Execute on Compute Nodes Dependencies • Compilers, libraries, … — one environment for all, it changes over time, application that compiles now may not compile in the future (like ABYSS). 33
  • 35. Docker at Compile time K workflow • Edit Source code • Cross-compile on Login Nodes • Execute on Compute Nodes Dependencies • Compilers, libraries, … — one environment for all, it changes over time, application that compiles now may not compile in the future (like ABYSS). 33 Cross-compile in Docker container
  • 36. Compile in container • Run container on User Computer • Run container on Login Node • Run container on Server 34
  • 37. Compile in container • Run container on User Computer • Run container on Login Node • Run container on Server 34 Compiler licence limitations
  • 38. Compile in container • Run container on User Computer • Run container on Login Node • Run container on Server 34 Compiler licence limitations “K” administration policies
  • 39. Compile in container • Run container on User Computer • Run container on Login Node • Run container on Server 35
  • 40. Run compiler on server in Docker container Requirements • Multiple environments (containers) • Multiple users • One machine • Limited resources — stop containers when they are not used Tasks • Copy source code into containers • Copy binaries from containers 36
  • 41. Docker-based IaaS Overview • One host multi tenant IaaS • One user — one container • Users login with SSH • Mount local folders into containers with SSHFS 37 github.com/pyotr777/dockerIaaSTools
  • 43. Docker-based IaaS Application Challenge • K-scope — tool for Fortran program static analysis. • Depends on Omni XMP compiler front-end, which produces AST (XML files). • Omni XMP depends on many packages and is hard to install. 39 omni-compiler.org/xcalablemp.html www.aics.riken.jp/ungi/soft/kscope github.com/pyotr777/dockerIaaSTools
  • 44. Docker-based IaaS Application Our solution • Script that mounts local source code folder in container and executes “make” command inside the container. • K-scope executes the script instead of make command. 40 github.com/pyotr777/dockerIaaSTools
  • 46. Docker-based IaaS Application Our solution • Script that mounts local source code folder in container and executes “make” command inside the container. • K-scope executes the script instead of make command. • After script finishes Intermediate code files are in local source folder — same effect as calling “make” command locally. 42 github.com/pyotr777/dockerIaaSTools