Changes between Version 2 and Version 3 of Public/User_Guide/Tutorial1/MSA_Idea


Ignore:
Timestamp:
Jan 30, 2020, 12:04:46 PM (4 years ago)
Author:
Anke Kreuzer
Comment:

—

Legend:

Unmodified
Added
Removed
Modified

  • Public/User_Guide/Tutorial1/MSA_Idea

    v2 v3  
    1919=== Traditional heterogeneous cluster ===
    2020* Attach accelerators (e.g. GPUs) to each CPU
    21 [[Image(HetCluster.PNG, 300px)]]
     21[[Image(HetCluster.PNG, 400px)]]
    2222 '''[[span(style=color: #00FF00, + )]]''' Energy efficient, easy management\\
    2323 '''[[span(style=color: #FF0000, -  )]]''' Static assignment of accelerators to CPUs\\
    … …  
    2525The MSA developed in DEEP-EST builds on the so-called Cluster-Booster architecture.
    2626=== Cluster-Booster architecture ===
    27 [[Image(CB.PNG, 300px)]]
     27[[Image(CB.PNG, 500px)]]
    2828 '''[[span(style=color: #00FF00, + )]]''' Energy efficient, high flexibility, dynamic ressource assignment
    2929==== Does this work? ====
    30 The Cluster-Booster architecture was first conceptualized and proven with prototypes in the DEEP project. It is a combination of a standard HPC Cluster and a tightly connected HPC Booster built of many-core processors or accelerators. The second project DEEP-ER evolved this architecture to address two significant Exascale computing challenges: highly scalable and efficient parallel I/O and system resiliency. Co-Design was the key to tackle these challenges – through thoroughly integrated development of new hardware and software components, fine-tuned with actual HPC applications in mind. Results of these two projects showed: '''Yes, it works! '''
     30The Cluster-Booster architecture was first conceptualized and proven with prototypes in the DEEP project.\\
     31It is a combination of a standard HPC Cluster and a tightly connected HPC Booster built of many-core\\
     32processors or accelerators. The second project DEEP-ER evolved this architecture to address two significant\\
     33Exascale computing challenges: highly scalable and efficient parallel I/O and system resiliency. Co-Design\\
     34was the key to tackle these challenges – through thoroughly integrated development of new hardware and software\\
     35components, fine-tuned with actual HPC applications in mind. Results of these two projects showed: '''Yes, it works! '''
    3136== 4) Towards a modular supercomputing architecture - The theory ==
    32 The idea of a MSA is a generalization to any number of specialized modules to address diverse application needs. An example could be a system arranged like this:\\
    33 [[Image(Idea.png, 300px)]]\\
    34 Multiple specialized modules will allow a wide range of different applications to efficiently use the system. Each application has its own way of using the MSA. One scenario where the workflos of several applications are distributed over the system could be:\\
    35 [[Image(Workflows.png, 300px)]] \\
     37The idea of a MSA is a generalization to any number of specialized modules to address diverse application needs.\\
     38An example could be a system arranged like this:\\
     39[[Image(Idea.png, 400px)]]\\
     40Multiple specialized modules will allow a wide range of different applications to efficiently use the system. Each\\
     41application has its own way of using the MSA. One scenario where the workflos of several applications are distributed\\
     42over the system could be:\\
     43[[Image(Workflows.png, 400px)]] \\
     44== 5) The DEEP-EST MSA prototype ==
     45The DEEP-EST project expands the Cluster-Booster architecture by adding a new module to the system: The Data Analytics Module\\
     46[[Image(ModularSupercomputerArchitecture.png, 400px)]]