3 Cloud Computing Services
My own cloud computing crash course.
Someone once told me: “If you still download your data, you’re not in 2020.”
For those who don’t know, this means that we are now in the age of cloud computing. As a data scientist, not knowing cloud computing is a sin (at least I have begun to think so).
What is Cloud Computing?
By Wikipedia, cloud computing is “the on-demand availability of computer system resources, especially data storage (cloud storage) and computing power, without direct active management by the user,” which means that you can store data elsewhere than your computer, still has access to it but with the security that it is not shared with everyone.
More practically, these are all the cloud services you probably have used already if you own a smartphone.
and also this
also Apple Cloud, just to name a few.
Essentially, if you can access your information and data from a different device with an online account, you are using the cloud service.
I still remember the day when my mom’s phone broke down, and we lost all the photos of my precious baby sister. That was before the cloud, and everything was stored on her iPhone. Now, if her phone breaks down, she can buy a new one and download all the data from her old phone from her cloud account. My company also stores everything on Google Drive with the enterprise account so that we can share and collaborate on documents online without having to send excel sheets over to one another as I used to at a different company.
Cloud computing relieves you from waiting time before someone takes a look at your documents and then sending it back and forth. It also allows the computer to work faster on bigger datasets because it does not have to store and load this data on its own. There’s also that back up and security component that takes away the stress and financial burden when the hardware is broken or lost (building caught on fire? iPhone dropped down the drain?).
Types of cloud services
When it comes to users and customers of cloud computing services, we have a beautiful image to help visualize the different tiers.
First, there is the SaaS, then the PaaS, and lastly, the IaaS.
SaaS is the host of end-user tools and services that I showed above, with Google Drive, MS Office, Apple iCloud, and everything of the kind.
PaaS is typically for developers who design websites, develop apps, and manage databases. Data scientists can optimize the cloud to manage their business database as I do with Google Sheets. However, when it comes to database cloud services, Google Sheets is not the best for big datasets because it’s limited to only 1M rows, and it takes quite long to load if your database structure is complicated. More on that below.
IaaS is a storage infrastructure, allowing companies to rent out the CPUs at data centers to store their data. Simply speaking, they rent out the computers, while PaaS lend out the time you can store data on their CPUs + some cool tools, and with SaaS you are just buying some time to store your data on their hard drive and not yours (like the iCloud storage you pay annually).
This image also helps to show what you manage and what your service provider manages for each service.
Cloud service providers
Now we are getting to the service providers, our household names are Amazon Web Services, Microsoft Azure, and Google Cloud Platform. These three companies are the most popular and seemingly coming on top of the race as well.
Amazon Web Services
AWS is the oldest cloud services since 2006 so it is no surprise that it is leading the way in the cloud computing sector. They have the highest number of data centers and has the most varied range of services. Notable clients of AWS are Netflix, LinkedIn, Facebook, Baidu, and ESPN.
Microsoft Azure
Launched in 2010 and is the second most common service for cloud computing. Microsoft alone takes up 20% market share (AWS takes up about 40%) and also offers many services, While AWS remains the most popular with user-friendly features, MS Azure is great for companies that are already committed to MS products and services as the data migration process to the cloud would be more seamless. Their notable clients are eBay, Boeing, Samsung, and BMW.
Google Cloud Platform
Launched in 2011 with significantly fewer data centers globally compared to the other two, Google is catching up with 10% of market share. Initially, GCP was built to enhance Google’s own products, including Google Services and Youtube, but has since successfully expanded their service capability. GCP also has proven to have incredible processing speed compared to its two competitors. Many big companies also vouches for its potential, including HTC, Sony Music, Coca Cola, Evernote, and more.
Conclusion
While there are many other cloud services for different types of users, many still prefer AWS for being the trailblazer with the expertise in cloud computing. But with the current trend of everything being moved to the cloud, more companies might be able to bring much surprises.
Either way, cloud computing is the future and it is just as essential to learn as Microsoft Excel a few decades ago. So I’d say that — yes, it is time to kickstart some Cloud computing crash course and this is just the beginning.
