Which cloud is best? GCP vs Azure
It’s easy to think you’ll be getting the same services and quality when you’re choosing between two of the largest cloud providers. But this isn’t necessarily the case.
Increasing competition means cloud providers like Google Cloud Platform (GCP) and Microsoft Azure are having to offer the best they possibly can to attract new clients and keep old ones. However, while they all aim for feature parity there are still some differences in their products.
These differences could be the reason one cloud is better for you than another. So to complete the set we now need to take a look at what GCP and Azure can offer, and decide – which is the best cloud?
Just a side note before we begin – this is a pretty long article, but it’s still only a brief overview of GCP and Azure. There’s no way we can go into huge levels of detail, or it would take days to read.
So if you do want to find out more, all you need to do is give us a call! That out of the way let’s get started.
File Storage Solutions
Good file storage solutions are vital to all businesses, especially anyone considering moving their business onto the cloud. Transferring on-premise data to a third party can be daunting. So how do GCP and Microsoft Azure make the transition easier?
Let’s start with Azure.
They have seven main storage solutions for various types of storage needs. We’ll give you a brief overview of these here to see if they would fit your needs. Firstly, we’ve got Azure Disk storage, where you can find scalable and secure storage for your virtual machines. Azure Blob storage is another scalable storage for unstructured data.
If you’ve got a selection of data that you don’t use very often, Azure Archive Storage can take care of this.
Next we’ve got Azure File storage – which is exactly what it sounds like. It’s a service that offers secure cloud based file sharing.
Avere vFXT for Azure is another storage service that lets you run high performance, file-based workloads in the cloud. A sixth storage option is Queue Storage, which offers storage for message-based communication between applications. And finally, we’ve got Data Box, an appliance for Azure data transfer.
So, Azure has quite a variety. Can GCP say the same?
There are six Google storage products we should consider when trying to decide whether GCP or Azure is best. Firstly, there’s GCP Cloud Storage.
This simply provides a highly available object storage in the cloud, as the name suggests.
Secondly, we’ve got Persistent Disk, which is a block storage suitable for any virtual machine or container.
Google Cloud Filestore is a managed network attached storage with Compute Engine and Google Kubernetes Engine instances.
You can store and serve your user-generated content in GCP’s Cloud Storage for Firebase.
Our fifth file storage option is Data transfer services, where you can move your structured and unstructured datasets into Cloud Storage, BigQuery or Cloud Dataproc, directly into Google Cloud of via the command line.
And finally, we’ve got Google Drive Enterprise. This is a shared service that allows you to edit, store, and share files.
So, it’s clear both GCP and Azure offer a great variety of storage options.
Choosing a winner is pretty tricky, but we’re gonna give it to Azure, as it has a slightly bigger range of services. However, this has been an incredibly brief overview. So if storage is a deciding factor for you, make sure to get in touch with us. We can give you a lot more detail about which of these services will suit you best.
Datacenters are places cloud providers are able to store your information. Both GCP and Azure have many datacenters spread over a number of regions where their services are available. So how do the two compare in terms of global coverage?
To be honest, it’s quite easy to choose a winner here. Microsoft have more regions than any other cloud provider, so they’re always going to come out on top. They currently have 54 regions, which help you secure your apps on a global scale. Azure is available in 140 countries too.
Despite this huge coverage, they are still looking to expand, and have announced plans for 10 more regions currently. Within each of their regions they have a number of availability zones. These are made up of data centers with independent power, cooling, and networking, to let you run your apps with high availability, disaster recovery, and low-latency replication.
The number of regions to choose from gives you more options, especially if you have specific needs in terms of compliance and data residency. Like most cloud providers, you’ll need to check each of the services you need to use are available in the region you’re choosing. But either way, with a variety like this, you’re bound to find what you need.
So how does GCP compare?
They are quite a way behind Azure in terms of the number of regions they currently have. As this is written, they have 61 availability zones that span across 20 regions.
But GCP itself is available in over 200 countries.
Plus, they have plans to expand into Seoul, Salt Lake City, and Jakarta. So, although they currently have fewer data centers and availability zones for your data, they are constantly expanding.
Plus, the fact their services are available over such a huge span of countries is always going to be a bonus. Like Azure, you’ll need to check the region you want has the services you are going to require. If other factors are making you lean towards GCP, their coverage certainly isn’t bad. But Azure does have quite a lot more.
Autoscaling services are necessary if you have big fluctuations in demands and traffic. Most cloud providers have services that help your applications cope with this. But which is best out of Azure and GCP?
Azure Autoscale offers a variety of functions within one service. It maximises the responsiveness of your applications, and can scale them by any metric. But on top of this, it is able to anticipate loads with different schedules, and save you money in the long run by not wasting servers.
Azure’s Autoscale feature is built into their Cloud Services, Mobile Services, Virtual Machines and Websites. Microsoft knows that performance will be different in various applications. So, as well as coping with apps that are CPU-bound or memory-bound, it also allows you to scale by a custom metric, to suit whatever needs you may have.
Their scheduled auto-scale helps you by anticipating traffic before it happens. For instance, if you know there is an upcoming event that will cause a large peak, or you know your traffic is largest at a certain time, you can adjust your auto-scaling targets to meet this demand.
On the other hand, Azure’s AutoScale will scale down your applications at quieter times, to ensure you aren’t wasting money that you could use elsewhere.
A final benefit to Azure’s AutoScale is alerts. It will monitor your performance metrics, and alert you to any changes.
But what about GCP?
Google offers load balancing and autoscaling services for groups of instances. These capabilities allow you to automatically add or delete instances from a managed instance group based on variations in load.
GCP’s autoscaling will help your applications handle increases in traffic, but, like Azure, will reduce costs when your needs are lower. All you need to do is define the autoscaling policy, and let Google’s autoscaler handle things from there.
You can choose to scale using either: average CPU utilization, HTTP load balancing serving capacity, or Stackdriver Monitoring metrics. You can also choose a target utilization level, at which your virtual machine instances will be maintained. GCP autoscaling and load balancing will automatically detect and remove unhealthy virtual machine instances with its health checks. But instances that become healthy again get re-added. So, GCP also has a good autoscaling service, but Azure’s service is clearly more mature and comprehensive.
Ethics and Sustainability
Ethics and sustainability are increasingly important in the modern age. Customers are increasingly choosing their services based on the most ethically conscious providers. If this is what you’re looking for in a cloud provider, we can help you choose! Both Microsoft and Google are fully committed to tackling environmental issues. So let’s take a look at what they do to help.
Microsoft are targeting areas in which they can have the most positive impact. For instance, they’ve made a pledge to lower carbon emissions by 75% with carbon neutrality, renewable energy commitments, and investments into energy efficiency.
They’ve managed to achieve carbon neutrality each year since 2012. They’ve purchased over 1 gigawatt of renewable energy and have projects for renewable energy on three continents. Plus, they have dedicated over $50 million over a period of five years to get cloud and AI tools into the hands of the people who are best at solving global environmental challenges. Their work is focused in three main areas – carbon and energy, water and ecosystems, and waste minimization.
In 2018 they also set to redesign their datacenters by using fuel cells as their primary power source in a Washington datacenter. This could double their datacenter energy efficiency. If you’re keen to find out more about the actions Microsoft are taking to become more green, they publish environmental reports on their website available for public use. They also have a green blog, with updates, news and opinions about their recent environmental progress!
Google are just as committed as Microsoft when it comes to sustainability. They have a major commitment to renewable energy, and match 100% of energy used by their global operations with renewable energy.
The cumulative corporate renewable energy they purchased in October 2018 alone was significantly higher than Microsoft. Google is actually the largest corporate buyer of renewable energy globally. They buy through power purchase agreements, and aim to buy from projects that are new to the grid. They buy a surplus of energy needed to ensure that there is a constant supply of this clean energy. This is to help reduce and replace the level of non-renewable energy on the grid.
They also have a continued commitment to carbon neutrality. They aim to minimise waste by ensuring their services are as efficient as possible. They’ve managed to reduce their overhead energy to 11%. Google Cloud has received an A rating from Greenpeace – the report is available online if you want to read it!
Like Microsoft, they are constantly striving to become more green, and make sure there are constant updates printed online.
So this one’s pretty tough to choose a winner, but we’re going to give the win to Google for their commitment to renewable energy. But, no matter whether you choose GCP or Azure, both cloud providers are doing massive things to make the cloud a greener place. Either would be a great choice.
If you’re looking for a way to develop your business further, you might be looking for a cloud provider that offers machine learning and AI services. Or perhaps you already use them and want services that can help you develop your skills and applications. Either way both Azure and GCP have machine learning options. Let’s take a look at the two to see which provider offers the best.
Starting with Azure. Microsoft Azure has a Machine Learning Studio full of solutions and services. To summarise it, their Machine Learning Studio is a browser-based, visual drag-and-drop environment that doesn’t even need coding. It’s a really simple and easy way to get what you need. They have a selection of tutorials and guides on their website, that not only help you get to grips with their services, but also give you advice on how to best utilise machine learning in general.
On top of this, they have a forum of engineers available for further questions. If you are totally new to machine learning, Azure also has a gallery of sample experiments that you can browse and use. They have built-in packages and support for custom code if you’re a current R or Python user.
Azure provides a variety of machine learning and AI services that can power the applications you want to build through APIs. The services, infrastructure and tools they offer are easy to use and scale. So it doesn’t sound like it can get much better! But how does GCP compare?
There are three main areas you’ll want to learn about if you’re planning on utilising the machine learning and AI services GCP offers. These are the GCP AI Hub, AI building blocks, and the AI platform. Here’s a brief overview of what these services offer.
The AI hub is a place to explore and experiment with machine learning technology within your organisation. It is basically a repository of plug-and-play AI components that you can view, use, and adjust to find what’s right for your needs. This service has sharing capabilities so you can privately host AI content to encourage reuse and collaboration between your developers.
AI building blocks is a service that helps developers use sight, language, conversation, and structured data within their applications.
And the AI Platform is a code-based data science development environment where machine learning developers can take their ideas from theory to reality. On top of this, GCP also provides consultants who are available to give you technical advice on every stage of this process. Plus, like Azure, they offer a repository of pre-trained AI solutions that you can use, which are really helpful if you’re new to machine learning and AI solutions.
Both Azure and GCP have a vast number of machine learning services and are constantly developing more, but again the greater maturity of the Azure platform gives it the win here.
Another factor that commonly influences what provider people choose is cost. Obviously you want to pay for quality services, but this doesn’t have to break the bank! Let’s compare the pricing of GCP and Azure to see which one helps you get the most out of your money.
First up is Azure.
The pricing of Azure services do vary from product to product. Thy have a handy pricing calculator that helps you find out exactly how much you’re going to be spending before you actually pay anything.
The majority of their services as pay-as-you-go, or pay as you need, such as their storage solutions and their virtual machines services. However the time-span of these payments can vary. For instance, some services may be pay-per-minute of use, whereas others may be a fixed payment each month. So, you’ll want to look at the specific services you’re planning on using the most when you’re properly comparing Azure and GCP.
But, there are some more perks to signing up for Azure.
An Azure account is free to make in the first place. But when you sign up, you also receive 12 months of free services, including compute, storage, network and database, you get over 25 “always free” services, including serverless, containers, and artificial intelligence, plus, you get £150 credit to explore other Azure services for 30 days!
So, you get a great opportunity to try before you buy.
But how does GCP compare?
Like Azure, GCP is free to sign up for, and has no upfront costs for their services. You also pay for services as you need them, so you’re not wasting money on products you aren’t actually using. Plus, if you do ever decide you don’t want to use a service any more, there are no termination fees.
On top of this easy pricing, Google offer a variety of discounts, including automatic sustained use discounts of 30% for workloads that run a significant amount of the billing month, and up to 79% off workloads that can be interrupted.
They also offer sizing recommendations that are based on compute usage, time savings, and money management, to help you save as much as possible. GCP does also offer an online calculator to help you figure out exactly how much you’re going to be spending.
Another similarity Azure and GCP have is the option to try it out for free first. GCP offers a 12 month free trial, that comes with $300 free credit to explore their services. This trial has access to all Cloud Platform products, including the most popular services, although some will have monthly usage limits.
Plus, you get free technical support for up to 12 months to help you understand exactly how to use the services.
I struggle to make an accurate decision as to which is the cheapest as the services they offer differ, but these guys https://kinsta.com/blog/google-cloud-hosting/#pricing are very clear that GCP is the cheapest option and they’ve done all sorts of research, and so if you are able to do all the work setting it up yourself, and are wanting to buy on price GCP wins this round.
A final influencing factor we’re going to look at today is security. Choosing a secure provider is always important when you’re handing over a lot of private information and data. So what steps do Azure and GCP take to protect it?
Let’s take a look at Azure first.
Azure has multi-layered security over their physical data centers, infrastructure, and general operations. To start things off, they have a team of over 3,500 cybersecurity experts that help safeguard your data.
Azure Security Center provides you with deeper insights into your protection with built-in controls across identity, data, networking, and apps. It offers a real time cybersecurity intelligence that can identify threats with actionable insights. These are achieved with machine learning, behavioral analytics, and application based intelligence.
There are several other steps Microsoft take to protect your information on Azure, so let’s summarise a few.
They allow you to manage and control user identity and access through multi-factor authentication.
Azure encrypts communications and operation processes, protecting your data in transit and at rest.
Azure Key Vault lets users safeguard and control crypto keys and other information used by cloud applications and services. Plus, Microsoft continuously monitors servers, networks, and applications to detect any threats. This is just a very brief overview of some of the security services Azure offers. If you want more information, you can give us a call.
Now, what about GCP?
Google Cloud was named a leader in Forrester Research’s Public Cloud Platform Native Security report, 2018. So what steps do they take to keep our data safe?
Their cloud infrastructure relies on multiple technologies for security for defense depth. They have an operations team that detects and responds to threats to infrastructure 24/7, every day of the year.
Online communications are encrypted, as is data at rest and in transit. Plus, identities, users, and services are strongly authenticated, and access to sensitive data can be protected by tools like security keys.
As well as all of these services to secure your data in the cloud, Google’s data centers also feature this multi-layered security. They have custom-designed electronic access cards, perimeter fencing, metal detectors, and even laser beam intrusion detection! They regularly undergo third-party audits and independent security verification, which has earned them a number of compliance certifications available to view online.
Overall, it’s clear to see that both Microsoft Azure and GCP offer some really great services. Azure is obviously a more mature and developed set of services than GCP, but it is fair to say that they don’t win everything, and that certain projects or sites are best suited to GCP.
There’s a lot to consider, even this list doesn’t have everything. So, if you want to talk through your specific needs, give us a call. We can help you make the decision, and we have great coffee.