GCP Cloud Run

In the digital age, where competition is fierce and consumer tastes change quickly, Time to Market (TTM) takes on fundamental importance for the success of digital products.

Google Cloud Run is a serverless platform that lets you run stateless containers. It takes care of all the infrastructure management such as scaling, patching, and provisioning. With Cloud Run, developers can focus on writing code instead of worrying about the underlying infrastructure.

Cloud Run supports both fully managed and Knative-based environments. Fully managed Cloud Run environments are fully managed by Google, meaning that you don't have to worry about managing the infrastructure. Cloud Run environments are based on the open-source Knative project, which provides a set of middleware components for building modern, container-based applications.

TTM is the time between the conception of a digital product and its launch on the market. A rapid TTM allows you to:

• Seizing market opportunities: in a constantly evolving context, being the first to respond to new customer needs can lead to a significant competitive advantage
• Anticipating the competition: placing an innovative product on the market before competitors can allow you to gain a market share important market and retain customers.
• Reduce development costs: a streamlined and rapid development process allows you to minimize waste and optimize the use of resources.

Digital products have a shorter life cycle than traditional products. For this reason, it is crucial to get them to market as soon as possible to make the most of their window of opportunity. Furthermore, the digital world is characterized by a high speed of innovation. If a product is not launched on the market quickly, it risks being quickly overtaken by more innovative solutions proposed by competitors.

In this context, Serverless approach is becoming more and more efficient to enable companies to release digital products faster then ever before, and if we talk about serverless, we cannot think about Google Cloud Run Service.

What is Google Cloud Run?

To learn about this service (that is my first choice-to-go for containerized applications), I’d like to interview a Customer Engineer Manager of Google Cloud Italy: William Bariselli.

Google Cloud Run is a serverless platform that lets you run stateless containers. It takes care of all the infrastructure management such as scaling, patching, and provisioning. With Cloud Run, developers can focus on writing code instead of worrying about the underlying infrastructure.

The Interview

Q: Hi , it’s a very pleasure to have a chance to interview you about Cloud Run and hosting you on my newsletter. we know each other for a long time, but would you introduce yourself for the reader that does not know you?

A: Ciao Francesco and thanks for hosting me. I'm William Bariselli, and I'm passionate about all things tech! I've been working in IT for more than 20 years. Three years ago, my passion led me to join Google Cloud as Principal Architect with a particular focus on supporting our Enterprise customers in the adoption of cloud native platforms that, as we will see, are the real enablers of a sustainable digital transformation and I’m now leading a team of Customer Engineers working on our Customers in Financial Services industry.

Q: So, let’s talk about Cloud Run. Can you describe what kind of service it is?

A: Imagine this – you've got some code, and you need to get it running online. You could set up your own servers, configure everything just right, and make sure it all stays up and running. Or... you could let Google do all the heavy lifting for you. That's where Cloud Run steps in. The real beauty of Cloud Run is its simplicity and flexibility. You can write your code in any language – Python, Go, Node.js, Ruby, Java, .NET, you name it. Just package it up as a container, and Cloud Run takes care of the rest. It also automatically scales up or down based on demand, so you only pay for the resources you actually use. Cloud Run basically says, "Forget all that! Package your code into a neat little container, and I'll handle the rest." Seriously, it runs your code, scales it up or down depending on how busy things get, and lets you focus on what you do best – building awesome stuff.

Q: what are the main use cases that can addressed with this service?

A: First up, we have APIs and microservices. If you're building a microservices architecture, Cloud Run's stateless nature and ability to scale quickly make it a perfect fit. You can deploy individual microservices as containers, and they'll communicate with each other effortlessly. Another great use case is web applications. Cloud Run is a fantastic choice for building and hosting modern, scalable web apps. You can use your favorite web frameworks like React, Angular, or Vue.js, and Cloud Run handles the deployment and scaling seamlessly. Cloud Run also excels at handling event-driven applications. It can be triggered by events from various sources, like Cloud Pub/Sub messages or changes in Cloud Storage buckets. This makes it ideal for processing data in real time or responding to events quickly.

And it doesn't stop there! Cloud Run can also be used for background tasks and batch processing jobs. You can run your scripts or data processing workloads in containers, and Cloud Run will ensure they execute reliably and efficiently.

Q: what are the benefit of Cloud Run?

A: First and foremost, simplicity is a major advantage of Cloud Run. You can focus on writing your code and leave the server management to Google. There's no need to provision or maintain infrastructure, which frees up your team to focus on development.

Next, Cloud Run offers unparalleled scalability. It can automatically scale your application up or down based on demand, so you never have to worry about being overwhelmed by traffic spikes or paying for idle resources.

Another key benefit is speed. Cloud Run can start up your containers in milliseconds, delivering near-instantaneous responses to user requests. This is crucial for real-time applications and those requiring low latency.

Cloud Run is also incredibly flexible. You can use any programming language, framework, or libraries you like, as long as you can package your code into a container. This gives you complete freedom to choose the tools that work best for you.

And we can't forget about cost-effectiveness. Cloud Run operates on a pay-per-use model, meaning you only pay for the resources your application actually consumes. This can lead to significant cost savings compared to traditional server-based approaches.

Q: usually, there are also limitations or constraints in such platforms. are there any key point we have to keep in mind when we move to Cloud Run?

A: Firstly, cold starts can be a consideration. When your Cloud Run service hasn't been used for a while, the first request may experience a slight delay as the container starts up. While Cloud Run optimizations mitigate this, it's something to be aware of, especially for real-time applications. Secondly, there are resource constraints. While Cloud Run scales automatically, there are limits on CPU, memory, and request duration. For extremely resource-intensive workloads, you might need to explore other Google Cloud options like GKE Autopilot. Finally, networking can be a bit nuanced. Cloud Run operates within a managed environment, so if you need fine-grained control over network configurations or direct access to certain services, it might require some additional setup or exploration of alternative solutions.

Q: What are the best practices to be applied for Cloud Run Development?

A: First things first, let's talk about container optimization. Cloud Run thrives on lightweight and efficient containers. Keep your container images small by using minimal base images, removing unnecessary dependencies, and optimizing your code for size, then consider statelessness. Cloud Run is designed for stateless services, so avoid storing session data or other persistent information in your containers. Instead, leverage external services like Cloud Memorystore or Cloud Firestore for data storage.

Before I mentioned cold starts, well, to minimize cold starts, consider using minimum instances. This feature keeps a few instances of your service warm, even when there's no traffic, reducing startup latency for subsequent requests, and of course, monitoring is key. Set up robust monitoring and logging to track your application's performance, errors, and resource usage. This will help you identify bottlenecks, troubleshoot issues, and optimize your Cloud Run services.

Q: Speaking about competitors, what are the main alternatives to Cloud Run, on GCP itself or on other Cloud Provider such Azure or AWS?

A: Let's start with GCP itself. While Cloud Run is a popular choice, Google offers other serverless options like Cloud Functions and Google Kubernetes Engine (GKE). Cloud Functions are great for simple, event-driven tasks, while GKE provides a more comprehensive platform for building and running your containers directly from Kubernetes creator.

Moving on to AWS, the primary competitor to Cloud Run is AWS Fargate. Fargate is a serverless compute engine for containers, similar to Cloud Run. It offers seamless scalability and integration with other AWS services. However, Fargate can be more complex to set up and manage compared to Cloud Run's simplicity. Over on Azure, the closest counterpart to Cloud Run is Azure Container Instances (ACI). ACI allows you to run containers on demand without managing servers. It's a good choice for simpler use cases, but it may lack some of the advanced features and scalability options that Cloud Run offers.

Q: can you tell us some success Case of Cloud Run usage?

A: Absolutely! Here are a few noteworthy success stories of companies leveraging Cloud Run to achieve their goals:

1. Veolia: This global leader in optimized resource management migrated their legacy monolith application to a microservices architecture on Google Cloud Run. This transition enabled them to scale their services independently, reduce operational overhead, and accelerate innovation.
2. Wind River: A provider of software for intelligent systems, Wind River utilized Cloud Run to build a highly scalable and reliable platform for IoT device management. This allowed them to efficiently process massive amounts of data from millions of devices in real time.
3. ENI Plenitude: Italian Energy Provider using Cloud Run to execute the microservices the support all the digital channels, serving more than 9 million customers, enabling to rapidly iterate and deploy new features while maintaining a lean and agile development process.
4. Many Startups: Numerous startups have embraced Cloud Run for its simplicity, scalability, and cost-effectiveness. It allows them to quickly launch and scale their applications without the need for extensive infrastructure management.

These examples illustrate how Cloud Run has empowered companies across various industries to achieve greater efficiency, scalability, and cost savings. If you're looking for a platform to streamline your application development and deployment, Cloud Run could be the answer.

Q: What is the average Costs for this service ?

A: The good news is, Cloud Run follows a pay-as-you-go model, meaning you only pay for the resources your application consumes. Cloud Run only charges you for the resources you use, rounded up to the nearest 100 milliseconds. This makes it very cost-effective for many use cases.

So, what's the average cost? Well, it can vary significantly depending on your application's usage patterns and resource requirements. However, Google provides a handy pricing calculator on their website that allows you to estimate your costs based on your specific needs. you can find detailed information here https://cloud.google.com/run/pricing

Conclusions

Cloud run is a powerful tool to compress the time-to-market at its minimum and remove complexity related to infrastructure, in order to let the team being focused only of adding values to the digital product.

For more informations, you can always refer to official documentation here: https://cloud.google.com/run/docs/.

If you want to know about it, please download the episode from my podcast: MODA - Modern Digital Architecture.

Follow my poscast also on Spotify and YouTube now!

?? William Bariselli: Experienced cloud architect turned leader, now heading Google Cloud's Italian Customer Engineering team for Financial Services. 20+ years in the tech industry (Oracle roots) have fueled my passion for solving complex challenges and helping businesses thrive in the digital age. Outside of work, you'll find me running, skiing or playing with my two daughters.