GCP's compute services

1. GCP's compute services

Hello and welcome! In this video we will further explore the suite of compute services offered by GCP.

2. Engines and functions

We have come across Compute Engine, App Engine, and Cloud Functions as key services that exemplify GCP's compute offering. Compute Engine offers customizable virtual machines for applications and computational tasks, a server-based approach that can satisfy very specific requirements. Though this requires specification of hardware in the form of VM configuration. If we want to avoid this configuration and go serverless, we have App Engine. App Engine provides a platform for efficient app development and scaling. Another option is Cloud Functions which enables event-driven computing for lightweight, task-specific applications.

3. Serverless spectrum

From Cloud Functions to App Engine to Compute Engine, notice how we progressively increase the amount of resource management required. Consequently, we also increase the amount of control we have over the hardware and software environment used to run our computation or application.

4. Microservices in containers

Focusing on just the software environment, we learned about containers. These virtual packages contain all the software, code, and data needed to run an application. Thus, they allow us to break bulky hard-to-manage applications into smaller ones running in containers. Each containerized application provides manageable microservices, which together make up your service.

5. The photo-sharing app

Imagine a scenario where your company runs a photo-sharing platform. You have deployed containerized microservices to handle photo uploads, downloads, and recommendations. If there is a surge in the number of uploads, the upload microservice needs to be scaled up. That is, an increase is required in the number of containers running this service. The uploads microservice also needs to trigger an update of the recommendations microservice so that fresh content is recommended to users. With a microservices architecture, container orchestration is crucial.

6. Kubernetes

This is where Kubernetes comes into action. Kubernetes is an open-source software that orchestrates containers. It functions harmoniously with all GCP services as it was invented by Google. GCP’s Kubernetes Engine or GKE harnesses Kubernetes and delivers a fully-managed solution for container orchestration. It automates scaling and updates, letting businesses focus solely on development. And of course, it integrates seamlessly with other GCP services like Cloud Storage and BigQuery.

7. Handling the upload surge

From our photo-sharing example, when there is a surge in uploads, GKE will automatically increase the number of upload microservice containers. Conversely, it can scale down the microservice when demand is low, cutting service costs. The containers can pass messages to each other in the GKE environment. In this case, uploads trigger an update of the recommendation microservice.

8. Complicated architectures

Now let’s say the data needed for your photo-sharing app is present in a server that you have on-premises for regulatory reasons. This server runs analytics on sensitive user data. The on-premises infrastructure needs to communicate with those microservices running in the cloud.

9. Bridging platforms

GCP offers Anthos for this situation. Anthos is designed for consistent deployment, scaling, and management of applications across various environments. This includes GCP, other cloud providers, as well as on-premises infrastructure. This unified management approach simplifies operations and reduces the risk of compliance violations. The only condition is that the application must have a container-based architecture. To this end, Anthos also facilitates the modernization of legacy applications by shifting them to container-based architectures.

10. Let's practice!

Now let’s review what we know about GCP’s compute services.

Create Your Free Account

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.