Skip to content
GitLab
Projects Groups Topics Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
  • Register
  • Sign in
  • E ear_learning
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributor statistics
    • Graph
    • Compare revisions
  • Issues 0
    • Issues 0
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 0
    • Merge requests 0
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Packages and registries
    • Packages and registries
    • Package Registry
    • Container Registry
    • Terraform modules
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • EAR_teamEAR_team
  • ear_learning
  • Wiki
  • Home

Home · Changes

Page history
Updated indexes. authored May 28, 2019 by xgomez's avatar xgomez
Hide whitespace changes
Inline Side-by-side
Home.md
View page @ 58b35439
...@@ -2,6 +2,14 @@ Overview of the learning phase ...@@ -2,6 +2,14 @@ Overview of the learning phase
------------------------------ ------------------------------
This is a phase prior to the normal EAR utilization, and it is called learning phase since is a kind of hardware characterization of the nodes. This is a phase prior to the normal EAR utilization, and it is called learning phase since is a kind of hardware characterization of the nodes.
During the learning phase a matrix of coefficients are computed and stored on all nodes. This coefficients next to some measurements, are used to predict the performance and energy costs of each repetitive sections of each application. To computed them, a set of stress tools (or kernels), are included in this package and are executed for a selected range of system frequencies. During the learning phase a matrix of coefficients are computed and stored on all nodes. This coefficients next to some measurements, are used to predict the performance and energy costs of each repetitive sections of each application. To compute them, a set of kernels (stress tools), are included in this package and are executed for a selected range of system frequencies.
A set of scripts to simplify the learning phase is provided in the `scripts` folder. It is also included the code of these stressing tools in the folder `kernels`. A set of scripts to simplify the learning phase is provided in the `scripts` folder. It is also included the code of these stressing tools in the folder `kernels`. It is required that the variable `$EAR_INSTALL_PATH` is set in the environment, by loading the EAR module or just defining it manually.
\ No newline at end of file
Learning phase steps
--------------------
1) Compile the kernels as described in the [compile guide](1-˗-Kernels-compile)
2) Make a lists files of the nodes where these kernels will run to following the [lists guidelines](2-˗-Nodes-lists-guidelines).
3) Test these kernels in a node through [test guide](3-˗-Kernels-test).
4) Execute these kernels using all your nodes lists as described in the [launch guide](4-˗-Kernels-launch).
5) Generate the coefficient files following the [coefficient files guide](5-˗-Coefficients-compute)
\ No newline at end of file
Clone repository
  • 1 ˗ Kernels compile
  • 2 ˗ Nodes lists guidelines
  • 3 ˗ Kernels test
  • 4 ˗ Kernels launch
  • 5 ˗ Coefficients compute
  • Home