Five Measures of Google Cloud Platform’s Size

Google Cloud Platform is currently one of the industry’s leading cloud computing platforms. Google entered the cloud computing market back in 2008 with Google App Engine – a Platform as a Service (PaaS) environment. Google App Engine was followed by BigQuery, a data analytics service in April 2012, and Compute Engine, an Infrastructure as a Service (IaaS) component – becoming generally available in December of 2013. Since then, Google has been regularly launching new cloud services and features.

This post will focus on the five key elements of the Google Cloud Platform.

Google Cloud Platform Regions

Each region in Google’s cloud contains any number of zones. Zones can be considered as isolated geographic locations within a region. Internally, zones are part of low-latency and high bandwidth networks. A zone failure does not affect other zones in that region. For achieving high availability, it is always advisable to deploy your application across multiple zones.

Globally, Google currently has three regions, i.e., us-central1, europe-west1 and asia-east1. These, by region, are the available zones:

us-central1 us-central1-a, us-central1-b, us-central1-f
europe-west1 europe-west1-a (Deprecated), europe-west1-b, europe-west1-c
asia-east1 asia-east1-a, asia-east1-b, asia-east1-c

Each of these zones supports either Ivy Bridge or Sandy Bridge processors. Specifically, us-central1-f and europe-west1-c zones work with Ivy Bridge, while the others use Sandy Bridge.


Since 2008, Google has launched multiple services for its Google Cloud Platform:

Group Service Name Service Description
Compute Engine
  • Google’s Infrastructure as a Service
  • For large scale workloads
  • Launch virtual machines on-demand
App Engine
  • Google’s Platform as a Service
  • Develop and host web applications
  • Currently supported programming languages: Python, PHP, Java and Go
  • Provides multiple built-in services like Memcache, Task Queues, etc
Container Engine
  • Powered by open source technology Kubernetes, enabling you to run Docker containers on Google Cloud Platform’s virtual machines
  • Currently, it is in the alpha phase.
  • Easy to move applications between development machines, on-premise systems, and public cloud providers
Cloud SQL
  • Fully managed relational MySQL database
  • Offers automatic encryption of data, patch management, replication, and scheduled backups
  • Durable and highly available object storage service
  • Can serve static objects directly from Cloud Storage
  • Managed NoSQL database
  • Built-in query support, ACID transactions
  • Automatic scaling and high availability
Elastic Load Balancing
  • Load Balancing as a Service
  • Offers HTTP, TCP, and UDP based load balancing
  • Perform health checks on backend instances
  • Does not require pre-warming
  • Carrier Interconnect – connect your infrastructure to Google, via enterprise-grade connections (interconnect service providers) to Google’s network edge
  • Direct Peering – Connect your business network directly to Google at any Google edge location worldwide and exchange high-volume cloud traffic.
  • Connect your existing network to your Compute Engine network via an IPsec connection
Big Data
Big Query
  • Analyze BigData
  • Run fast, SQL-like queries against multi-terabyte datasets
  • Real-time insights about your data
  • A simple, flexible, and powerful system you can use to perform data processing tasks of any size
  • Currently, it is the alpha mode
  • Used for large scale data processing scenarios such as Extract, Transform, Load (ETL), analytics, real-time computation, and process orchestration.
  • provide reliable, many-to-many, asynchronous messaging between applications
  • Publisher applications can send messages to a “topic” and other applications can subscribe to that topic to receive the messages
  • Currently, it is in alpha mode
Deployment Manager
  • Infrastructure management service that makes it simple to create, deploy, and manage Google Cloud Platform resources
  • Create a static or dynamic template that describes the configuration of your Google Cloud environment.
  • Currently, it is in alpha mode
  • Monitoring solution from Google Compute Platform
  • Provides dashboards and alerts for your cloud-powered applications
  • Currently, it is in beta mode and powered by Stackdriver

Compute Capacity

Google Compute Engine offers a wide range of computation capacity for above-described services. Apart from the wide pool of instance types, Google Compute Engine offers performance tuned instances for different types of workloads i.e., High CPU instance types for CPU intensive applications, High Memory instance types for memory intensive applications, etc.

Google Compute Engine instances are available in four categories: Standard, High CPU, High Memory, and Shared Core. The Standard includes five instance types built within a range between 1VirtualCore, 3.75GBMemory to 16VirtualCore, 60GBMemory. These instances are used for standard workloads. The High CPU category has four instance types with profiles lying between 2VirtualCore, 1.80GBMemory to 16VirtualCore, 14.40GBMemory. Using High Memory, you can choose between four instance types with compute capacities ranging from 2VirtualCore, 13GBMemory to 16VirtualCore, 60GBMemory. Shared Core category instances come in two flavors: f1.micro (1VirtualCore, 0.60GBMemory) and g1.small (1VirtualCore, 1.70GBMemory).

Big data

Google’s ability to deliver search results in milliseconds, serve six billion hours of YouTube video per month or successfully serve their 425 million Gmail users largely depends upon their interactive query service. BigQuery is a public implementation of their own internal query service ‘Dremel.’ It allows them to run SQL-like queries against very large datasets and fetch results in seconds.

According to Google BigData whitepaper, Dremel Can Scan 35 Billion Rows without an Index in Tens of Seconds. Dremel, the cloud-powered massively parallel query service, shares Google’s infrastructure, so it can parallelize each query and run it on tens of thousands of servers simultaneously”. Google uses Dremel for analyzing web documents, spam analysis, and monitoring data center performance.

Besides all this, there are box loads of Google-developed solutions to satisfy big data needs, like BigTable, Chubby, MapReduce, etc. Google also launched DataFlow, a managed data processing service, which will help you to ingest, transform, and analyze data in both batch and streaming modes by creating data pipelines.

Media Management

Google has developed solutions to tackle a wide range of problems and these solutions are massively used by people around the globe. Google search is one of the most popular, followed by Google Mail, YouTube, Google Drive, Google Maps, Adwords, and Google Docs. These serve millions of users on daily basis and themselves effectively run on Google Cloud Platform. With their many services, Google has built an ecosystem which allows you to integrate multiple solutions. For example, an attachment shared on Gmail account can be directly saved into your Google drive, automatically syncing to your mobile device. Or you can seamlessly share your favorite YouTube video directly to your Google+ profile. These integrations make the lives of millions of user a lot easier and flexible. But they can also provide you with powerful cloud computing tools.

To conclude, Google Cloud Platform is an enormous platform used, possibly, by most people currently living on our planet (and a handful in space). Enterprises, SMBs or Start-ups can leverage multiple Google Cloud Platform’s offerings, while individuals use many of the very Google solutions in their daily lives.

Cloud Academy