Google Compute Engine Reviews & Product Details

I really like how I can start a Google Compute Engine instance by clicking on the SSH button that is available on the portal, so I don't need to hassle with installing gcloud or SSH to my own terminal. It is a really handy feature. I really like its user interface—everything I need is visible. For example, if I need an external IP, it's already visible in the table at the top. It doesn't waste time; everything is just right there available on the screen. The initial setup was really smooth because of the SSH button. Review collected by and hosted on G2.com.

I can't see or update anything of backend files if the person responsible is on leave or if the backend suddenly breaks. I wish there was a way to control access so any developer can decide who else can access their folder. Review collected by and hosted on G2.com.

One thing I really like about Google Compute Engine is how much control it gives you over your virtual machines. You can choose the exact CPU, memory, disk type, and networking configuration depending on what you need. It’s very flexible compared to some higher-level managed services.

In my day-to-day work I also appreciate how well it fits into automation workflows. Spinning up instances, attaching disks, or modifying infrastructure is easy through APIs or infrastructure-as-code tools. Once everything is defined properly, you can recreate environments pretty quickly.

Overall it feels reliable and production-ready, which is really important when you’re running backend services. Review collected by and hosted on G2.com.

The pricing model can take some time to fully understand. There are a lot of small components that affect the final cost, like machine types, storage, and network traffic. If you’re not paying attention it’s easy to underestimate how much something will cost.

Another minor downside is the cloud console. It’s powerful but sometimes feels cluttered, especially when you’re trying to find a specific configuration option. After using it for a while it becomes easier, but the first impression can be a bit overwhelming. Review collected by and hosted on G2.com.

I used Google Compute Engine to run microservices and test new ones. I really like that you can manage infrastructure with code, which allows for quick automation of server deployments and configurations through scripts. Scaling the infrastructure as the load increases is extremely convenient. I especially appreciate the flexible orchestration in GKE and the quick configuration of VM Instances. The monitoring also impressed me. Integration with the load balancer and the ability to quickly deploy a virtual server without setting up physical hardware greatly facilitate the work. Review collected by and hosted on G2.com.

Sometimes the interface can be a bit complex for beginners because there are many configurations and components. Also, the pricing model is not always immediately clear, as the cost is composed of several parameters - computing resources, network, disks, and other services. It would be nice to have a simpler and more visual cost calculation for new users, and of course, discounts for different categories of users. Review collected by and hosted on G2.com.

One of the things I like most i like about Google Engine is that I can create highly customized machine types with exact CPU and memory specs based on the specific workloads. The live migration is also a massive advantage, as it keeps my instances running smoothly even during backend maintenance. Review collected by and hosted on G2.com.

While the core features are robust, I find that some of the documentation are outdated or lacks in-depth troubleshooting steps. I also find that the premium tech support cost can be quite high for smaller organizations that aren't on an enterprise level. Review collected by and hosted on G2.com.

I find Google Compute Engine easy to manage, thanks to the gcli, and it's easy to scale both vertically and horizontally without much hassle. It's simpler than AWS since I don't need to stop and resize for maintenance. I appreciate the tight integration with IAM, which helps with permission provisioning. The setup was pretty easy and fast. Review collected by and hosted on G2.com.

What I don't really like is the division of the projects around it. I need a project on GCP for everything, and having it all divided can be quite a handful to manage with all my Google Compute Engine having to switch this up. Review collected by and hosted on G2.com.

I like the scaling of resources on Google Compute Engine. It's great that I can go from using 2 T4 to 6 H100 quickly. Once the permission is granted from my organization, I can get 6 more resources on the same day. We can turn off the training, change the config, add extra resources, and restart training all in one day. The initial setup was easy, especially because I've used AWS before, and this felt easier once the credentials were set. Review collected by and hosted on G2.com.

I mean BigQuery is still separate and sometimes when running a large SQL query, it takes time. I wish they could add compute there as well. Review collected by and hosted on G2.com.

What I like best about Google Compute Engine (GCE) is how cleanly it balances power, flexibility, and performance without getting in your way.

A few standout things for me:

• Custom machine types – you’re not forced into rigid instance sizes. Need more CPU but less RAM? Easy. That level of control saves money and fits real workloads better.

• Strong performance & networking – Google’s global network is genuinely fast and reliable. Low latency, solid bandwidth, and great for distributed systems.

• Live migration – VMs can be moved during host maintenance with little to no downtime. That’s huge for production stability.

• Deep integration with the Google ecosystem – works seamlessly with GKE, Cloud Storage, BigQuery, IAM, and monitoring. Everything feels designed to work together.

• Pricing transparency – per-second billing and sustained-use discounts kick in automatically, which is refreshing.

Overall, GCE feels very engineer-friendly: fewer arbitrary limits, more control, and infrastructure that just quietly does its job. Review collected by and hosted on G2.com.

What I dislike about Google Compute Engine (being honest, not dramatic 😅):

• Steep learning curve – GCE assumes you’re already comfortable with cloud concepts. IAM, networking, projects, VPCs… powerful, but not very beginner-friendly.

• IAM can feel overcomplicated – super granular (which is good), but figuring out the right role vs too much access can be frustrating.

• Console UX is cluttered – the web console tries to do everything at once. Simple tasks sometimes take more clicks than they should.

• Support isn’t great unless you pay – community docs are solid, but real-time help or faster resolution usually means a paid support plan.

• Unexpected costs if you’re not careful – things like egress traffic, static IPs, and idle resources can quietly add up.

• Smaller ecosystem compared to AWS – fewer third-party tools, tutorials, and “copy-paste” solutions in some niches.

In short: GCE is excellent for engineers who know what they’re doing, but it can feel intimidating and a bit unforgiving if you’re new or moving fast without guardrails. Review collected by and hosted on G2.com.

I like that GCP feels like a modern solution where everything is fairly simple and clear. Unlike some other providers with a lot of legacy stuff, where you can spend hours trying to find what you need, GCP feels intuitive. Most of the time, I can figure out what I need and get it done within a few minutes without having to dig through a lot of documentation. Review collected by and hosted on G2.com.

I can’t really think of anything I genuinely dislike. Maybe the documentation could be better in some cases Review collected by and hosted on G2.com.

I really like the compatibility of Google Compute Engine with CI/CD pipelines and the capability of creating custom machines. It's great that I can choose my monitoring tools. It's very easy to create scalable and maintainable virtual machine clusters. The solution is also easy to use, which is why we chose it. Google Compute Engine has made efforts in security, making it easier to detect bot attacks and provide monitoring tools for GCP users. Review collected by and hosted on G2.com.

I think the protection against bots, especially cryptobots, could be improved. It has been an issue over the last year. I feel like GCP could do a better job with tools to protect against these threats. Review collected by and hosted on G2.com.

What I appreciate most about GCE is its emphasis on performance and the flexibility of its customizable VM architecture. One standout feature is live migration, which allows virtual machines to remain operational during maintenance, minimizing downtime—a capability that's uncommon among major providers. The per-second billing model ensures you only pay for the resources you actually use, adding to its cost-effectiveness. With custom machine types, you can precisely adjust CPU and RAM allocations rather than being limited to preset VM sizes. Networking performance is robust, which is particularly beneficial for distributed workloads and global deployments. Additionally, GCE offers deep integration with Google services, making it especially advantageous for AI/ML workloads, Kubernetes, and BigQuery. Altogether, these features make GCE a highly cost-efficient and reliable choice for production workloads, particularly for organizations that require flexibility and high uptime. Review collected by and hosted on G2.com.

One aspect I find frustrating about Google Compute Engine is the complexity of its pricing structure. The costs for custom machine types and network egress can accumulate quickly, and if you’re not keeping a close eye on your usage, it’s easy to end up spending more than intended. Review collected by and hosted on G2.com.