Exploring GPUs in Google Compute Engine - GCE | Scribe

    Exploring GPUs in Google Compute Engine - GCE

    • Hafeez Baig |
    • 12 steps |
    • 42 seconds
    1
    Sign in into the **Google Cloud Platform**
    2
    Type "**Compute Engine**" in the search bar and click on the **Compute Engine** option
    information ordinal icon
    **What is a Compute Engine?** Google Compute Engine (GCE) is an Infrastructure-as-a-Service (IaaS) offering from Google Cloud Platform (GCP) that allows users to run virtual machines (VMs) on Google's infrastructure. It provides scalable and flexible computing resources, high-performance persistent storage, advanced networking capabilities, and integration with other GCP services, making it suitable for a variety of workloads including web hosting, data processing, and high-performance computing.
    3
    **VM instances** wizard will open, click on the **CREATE INSTANCE** button on the top left to create a new instance
    information ordinal icon
    **What is an Instance?**\ \ An instance is a virtual machine (VM) running on Google Cloud Platform's Compute Engine, providing the necessary resources and environment to run applications and workloads.
    4
    **Create an instance** wizard will open, give the Name as - **"instance2"**
    5
    Scroll to the **Machine configuration** section, click on the **GPUs** option
    information ordinal icon
    **What are GPUs?** GPUs, or Graphics Processing Units, are specialized hardware components designed to accelerate the processing of graphics and complex mathematical computations. Initially developed for rendering images and video in gaming and graphic design, GPUs are now widely used in various fields like scientific computing, machine learning, and artificial intelligence due to their ability to handle parallel processing efficiently. They significantly speed up tasks that involve large-scale data processing and computationally intensive operations.
    6
    Scroll to the **GPU type** section, here you can select the GPU type as per your requirement **Note:** Example - NVIDIA T4 which is one of the GPU type
    information ordinal icon
    **What is GPU type NVIDIA T4?**\ \ The NVIDIA T4 is a type of GPU (Graphics Processing Unit) commonly used in cloud computing and data centers for various applications such as machine learning, artificial intelligence, and high-performance computing. It features NVIDIA's Turing architecture and is known for its efficiency in both training and inference tasks. The T4 GPU is designed to provide accelerated computing performance with optimized power consumption, making it suitable for a wide range of compute-intensive workloads in cloud environments like Google Cloud Platform (GCP).
    7
    Click on the **Number of GPUs** dropdown, here you can select and set the number of GPUs you need as per your requirement
    8
    You can also configure GPUs using the General Purpose VM machines, scroll to the **Machine configuration** section and click on the **General purpose** option
    information ordinal icon
    **What are General purpose Machines?**\ \ General purpose machines in Google Cloud Platform are VM instances optimized for a balanced mix of compute, memory, and network resources, suitable for a wide variety of workloads such as web servers, applications, and development environments.
    9
    Select the Series **E2** option
    information ordinal icon
    **What is E2 in General Purpose Machine?**\ \ E2 instances in Google Cloud Platform's Compute Engine are part of the general-purpose machine family optimized for cost-effective performance. They offer a balanced mix of CPU and memory resources suitable for a wide range of applications, from web servers to enterprise applications, providing flexibility and cost efficiency.
    10
    Scroll to the **Machine type PRESET** section and click on the dropdown
    11
    Select the option **High CPU** from the dropdown
    information ordinal icon
    **What is High CPU?** In Google Cloud Platform (GCP), High CPU machine types refer to virtual machine configurations that emphasize higher CPU resources relative to memory. These types of machine configurations are designed for compute-intensive workloads that require more processing power than memory capacity. They are suitable for tasks such as data processing, computational modeling, and high-performance computing where CPU performance is critical. High CPU machine types typically offer a higher ratio of vCPUs (virtual CPUs) to memory compared to general-purpose machine types, optimizing them for compute-bound applications.
    12
    On the right side you can select the **High CPU** type for your VM instance
    information ordinal icon
    Congratulations on completing this lab and Exploring GPUs in Google Compute Engine!