Outside of ease of use and configuration, I haven't experienced performance problems across any of the cloud services that I have used.
Google Compute Engine is an infrastructure-as-a-service (IaaS) product from Google Cloud. It provides virtual machines with carbon-neutral infrastructure which run on the same data centers that Google itself uses.
Predefined Machine Types
Virtual machine configurations for micro instances to instances with up to 160 vCPUs and 3.75 TB of memory
Custom Machine Types
Or create and customized, virtual machines with the shape (i.e. vCPU and memory) needed for each project
Network storage, up to 64 TB in size, can be attached to VMs as persistent disks. Create persistent disks in HDD or SSD formats. If a VM instance is terminated, its persistent disk retains data and can be attached to another instance. Take snapshots create new persistent disks from that snapshot.
Always-encrypted local solid-state drive (SSD) block storage. Local SSDs are physically attached to the server hosting the virtual machine instance. Local SSD sizes up to 3 TB are available for any VM with at least 1 vCPU.
Global Load Balancing
Distribute incoming requests across pools of instances across multiple regions.
Cost effectively run large compute and batch jobs using Preemptible VMs. Fixed pricing and no contracts or reservations make it easy: simply check a box when you create the VM and turn them off when the work is done.
Run, manage, and orchestrate Docker containers directly on Compute Engine VMs or with Google Kubernetes Engine.