Google Compute Engine ( GCE ) - Infrastructure as a Service
Google Compute Engine is the Infrastructure As A Service (IaaS) component of the Google Cloud Platform that facilitates the following.
- Cost effective, scalable virtual machines running on Google’s data centers and worldwide fiber network / Infrastructure
- Choice of predefined virtual machines as well as custom virtual machines.
- Tools and Workflows to enable scaling from single instance to global, load-balanced cloud computing.
- Everything you need for your computing infrastructure
Compute Engine’s VMs boot quickly, come with persistent disk storage, and deliver consistent performance. Cloud Compute virtual servers are available in many configurations, including predefined sizes. Or the option to create custom machine types optimized for specific needs.
Industry-leading price and performance
Compute Engine VMs boot quickly and are consistently high performance. Compute Engine also offers industry-leading local SSD performance. Due to Google’s fiber-optic network that transfers the data to the nearest destination, it is much faster in performance.
Low cost, automatic discounts
Google bills in second-level increments, so you only pay for the compute time you use. Google provides specialized discounts called sustained-use discounts, that automatically give discounted prices for long-running workloads with no up-front commitment required.
Most of the existing cloud infrastructure providers give discounts if you commit for longer-term but Google committed use discounts are without any upfront commitment period.
Fast and efficient networking
Create large compute clusters that benefit from strong and consistent cross-machine bandwidth. Connect to machines in other data centers and to other Google services using Google’s private global fiber network. Create an instance, check the network configs, and run some tests.
Flexibility for Every Workload
Google cloud compute tools help resize your clusters, create machine images, virtualize the network. Cloud Compute tools facilitate preemptible VMs for batch workloads, and custom machine types to optimize specific computing needs. Google pricing model does not have lock-in periods to push you into using obsolete machine types.
Predefined machine types
Compute Engine offers predefined virtual machine configurations for every need from micro instances to instances with up to 160 vCPUs and 3.75 TB of memory. You can use these virtual machines to host websites, web applications, databases, and any other Line of Business (LOB) applications. These predefined machines are defined in the following categories,
- General Purpose Machine Types
- Memory Optimised Machine Types
- Compute Optimised Machine Types
- Shared Core Machine Types
Based on the usage customers can choose the right virtual machine type and quickly start hosting or migration.
Custom machine types
Create virtual machines with the shape (i.e. vCPU and memory) that is right for your workloads. By tailoring a Custom Machine Type to your specific needs you can realize significant savings. With custom machine types, we can create virtual machines with the optimal amount of CPU and memory for the workloads planned. Compute Engine offers a start / stop feature that enables us to configure multiple custom machines and start the one that is suitable for the expected load and stop when the load changes.
Persistent disks are the network storage devices that virtual machines and instances can access like a physical disk of a desktop or server. The data on each persistent disk is distributed across several physical disks. Compute Engine manages the physical disks and their data distribution for optimal performance. Google Compute Engine supports
- Standard Persistent Disks supported by Standard Hard Disk Drives (HDD)
- SSD Persistent Disks supported by Solid State Drives (SDD)
Network storage, up to 64 TB in size, can be attached to VMs as persistent disks. You can 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. You can also take snapshots of your persistent disk and create new persistent disks from that snapshot.
Compute Engine offers always-encrypted local solid-state drive (SSD) block storage. Unlike standard persistent disks, local SSDs are physically attached to the server hosting the virtual machine instance offering very high input/output operations per second (IOPS) and very low latency compared to persistent disks. Local SSD sizes up to 3 TB are available for any VM with at least 1 vCPU.
The data that is stored in local SSD persists only till the instance is active. It gets deleted the moment the instance is stopped or deleted.
Our innovative datacenters and live migration technology enable proactive infrastructure maintenance, improving reliability and security. Your live VMs are automatically moved to nearby hosts, even if your VMs are under extreme load while underlying host machines undergo maintenance. You won’t have to reboot your VMs due to host software updates or even some classes of detectable hardware failure.
Global load balancing
Global load-balancing technology helps you distribute incoming requests across pools of instances across multiple regions, so you can achieve maximum performance, throughput, and availability at low cost.
Linux and Windows support
Run your choice of OS, including Debian, CentOS, CoreOS, SUSE, Ubuntu, Red Hat Enterprise Linux, FreeBSD, or Windows Server 2008 R2, 2012 R2, and 2016. You can also use a shared image from the Google Cloud Platform community or bring your own.
Batch processing with preemptive VMs
Preemptible VMs are highly affordable, short-lived compute instances suitable for batch processing and fault-tolerant workloads. Preemptible VMs offer the types and options of the same machines as regular compute instances and last up to 24 hours.
You can simply check a box when you create the VM as preemptible and turn them off when the work is done.
Preemptible VMs are fixed pricing up to 80% off the standard virtual machines. So you save a lot of money for load spikes and batch processing.
Compliance and security
All data written to persistent disk in Compute Engine is encrypted on the fly and then transmitted and stored in encrypted form. Compute Engine has completed ISO 27001, SSAE-16, SOC 1, SOC 2, and SOC 3 certifications, demonstrating our commitment to information security.
Commitment savings / Committed-use discount
Compute Engine offers the ability to purchase committed use contracts in return for deeply discounted prices for the VM usage. These discounts are called committed-use discounts and can save up to 57% with no upfront costs or instance type lock-ins.
Committed-use discounts are ideal for workloads with predictable resource needs. with committed-use contracts, you can purchase a certain amount of vCPUs, memory, SSDs, etc at a discounted price in return for 1 year or 3-year commitment.