Scribd
Upload
4 views1 page

Task 2

The document outlines a task to install and utilize cloud infrastructure and platforms like Docker for solving problems using parallel processing with TensorFlow or PyTorch. It includes instructions for running distributed programs, modifying them for task or data parallelism, and applying them across various computing environments. A report is required detailing the installation process, algorithm explanations, deployment environments, performance comparisons, and findings, while adhering to specific guidelines.

Uploaded by

fixelcloud.sameen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
4 views1 page

Task 2

The document outlines a task to install and utilize cloud infrastructure and platforms like Docker for solving problems using parallel processing with TensorFlow or PyTorch. It includes instructions for running distributed programs, modifying them for task or data parallelism, and applying them across various computing environments. A report is required detailing the installation process, algorithm explanations, deployment environments, performance comparisons, and findings, while adhering to specific guidelines.

Uploaded by

fixelcloud.sameen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

Task 2

Objective:

 To install and use cloud infrastructure & platforms such as Docker for
solving a problem.
 To apply parallel processing of TensorFlow (data parallelism or model
parallelism) or PyTorch (Distributed Data Parallel) of training machine
learning algorithm on multi containers or multi nodes.
Instruction:

1. Try to install and run an example of distributed program with the proper
installation of middleware on multiple machines, VMs or dockers or with
public cloud instants.
2. Modify the program to show the task or data parallelism in solving a problem.
3. Apply your code with CPU, GPU, multi-CPU and multi-GPU.
4. Write a report.

Report Guidelines:
1. The report should not exceed 7 pages. Report can be in any journal format,
2. Describe your activities such as installation process and division of tasks.
3. Explain your algorithm or codes.
4. Describe the deployment environment such as how many machines are used.
5. Describe your data sources.
6. Compare your solution in terms of performance that on single node and multi-
nodes, single container and multi-containers.
7. Show your output.
8. Discuss your findings, share your experiences.
9. Attach source code.
10. References: You can refer to GitHub and need to cite them.

You might also like