Merge pull request #23 from Exabyte-io/add-another-hpc-rewote

Add containers-focused HPC rewote

Cloud-Infrastructure.md

@@ -13,8 +13,7 @@ The aim of this task is to create a CI/CD pipeline (github workflow) that includ
 
 # Expectations
 
-- The application may be relatively simple - e.g. Linpack, this is focused more on application infrastructure and DevOps, but the tests must actually verify functionality
-- Correctly passes the tests in automation and displays coverage metrics
+- The application may be relatively simple - e.g. Linpack, this is focused more on infrastructure
 - Clean workflow logic
 
 # Timeline
@@ -23,7 +22,7 @@ We leave exact timing to the candidate. Should fit Within 5 days total.
 
 # User story
 
-As a user of this CI/CD pipelin I can:
+As a user of this CI/CD pipeline I can:
 
 - initiate tests for a specific number of scenarios: e.g. 2 nodes, 16 core per node
 - select the instance type to be used 

Containerization-HPC.md

@@ -0,0 +1,44 @@
+# Containerization / Benchmarks (HPC)
+
+> Ideal candidate: skilled HPC engineer versed in HPC, and Containers
+
+# Overview
+
+The aim of this task is to build an HPC compatible container (i.e. [Singularity](https://sylabs.io/guides/3.5/user-guide/introduction.html)) and test its performance in comparison with a native installation (no containerization) for a set of distributed memory calculations.
+
+# Requirements
+
+1. A working deployment pipeline - using any preferred tool such as SaltStack, Terraform, CloudFormation - for building out the computational infrastructure
+2. A pipeline for building the HPC compatible container
+3. A set of benchmarks for one or more HPC application on one or more cloud instance type
+
+# Expectations
+
+- The application may be relatively simple - e.g. Linpack, this is focused more on infrastructure
+- Repeatable approach (no manual setup "in console")
+- Clean workflow logic
+
+# Timeline
+
+We leave exact timing to the candidate. Should fit Within 5 days total.
+
+# User story
+
+As a user of this pipeline I can:
+
+- build an HPC-compatible container for an HPC executable/code
+- run test calculations to assert working state of this container
+- (optional) compare the behavior of this container with a OS native installation
+
+# Notes
+
+- Commit early and often
+
+# Suggestions
+
+We suggest:
+
+- using AWS as the cloud provider
+- using Exabench as the source of benchmarks: https://github.com/Exabyte-io/exabyte-benchmarks-suite
+- using CentOS or similar as operating system
+- using Salstack, or Terraform, for infrastructure management

README.md

@@ -20,6 +20,7 @@ Each file represents an assignment similar to what one would get when hired.
 | Back-End / Ops | [Parallel Uploader](Parallel-File-Uploader.md)   | Python, OOD, Threading, Objectstore |
 | CI/CD, DevOps  | [End-to-End Tests](End-To-End-Tests.md)    | BDD tests, CI/CD workflows, Cypress |
 | HPC, Cloud Inf | [Cloud HPC Bench.](Cloud-Infrastructure.md)    | HPC Cluster, Linpack, Benchmarks |
+| HPC, Containers| [Conteinerized HPC](Conteinerization-HPC.md)    | HPC Cluster, Containers, Benchmarks |
 
 ## Usage