Microsoft, Amazon, Google and Oracle offer infrastructure services whose names change, but the concepts remain the same. It may be that you are interested in the cloud too, but it’s still not clear because you find much information all over and it’s hard to make decisions. Isn’t it?
Clouds infrastructure provide traditional data centers, nothing new, but instead of provisioning your machines and your connections by hand in your air-conditioned room, you do everything by mouse clicks (or lines of commands).
Imagine your company launches a new product on the market and you must implement a web space from where customers can buy that new product. Or maybe, you have to support your organization with a collaborative platform from which your colleague can work more efficiently. How would you start in the cloud?
The first thing you’ll have to do is to choose the right location to provide your services. Here is the list of available countries for the Oracle Cloud:
- United Kingdom
- United States
- South Korea
- Switzerland (September 2019)
Then you need one or more servers to run your platform. Once logged in your cloud customer portal, you can choose physical machines or virtual machines. Then, those physical and virtual machines are then declined in “shapes” (template):
- Standard shapes: A standard for most applications
- DenseIO shapes: Made for high data workloads
- GPU shapes: Made to use graphics card processor (Calculation of Bitcoins, algorthym to win the game of Go, …)
- High performance computing (HPC) shapes: For physical machine only. Made for massive CPU computing needs in parallel
Few minutes after, you have your server, and you probably need to store data for that platform. Does it make sense?
The Autonomous Database service
Oh no! Maybe you remember that to get a database, you must request it and usually, it lasts weeks. Let’s see how Cloud providers address that issue. Still, from your cloud customer portal, you can request an Autonomous database. It’s a service managed by the cloud provider which serves two types of workloads:
- The standard application workload (very frequent short requests)
- The workload for data warehouses (long unit processing)
What’s still on your own? You configure the number of CPUs and the storage capacity, which is simple and done without service interruption or performance degradation. Then, the Oracle handles all the following operations automatically:
- Creating the database
- Backing up the database
- Patching the database
- Upgrading the database
- Tuning the database
Note: It is possible to keep existing licenses for the cloud or to use the licenses included in the service. A short and proper audit of the cost will tell you which option is the best.
Now, how would you connect your application server, your database server, and make the application available to your customers and your colleagues? The connection possibilities are limitless because you can do what is possible in On-premises. It is possible to:
- Connect your Cloud Infrastructure with the Internet and / or your data center
- It is possible to connect the Oracle Cloud with another Cloud (Amazon AWS for example)
- It is possible to connect a cloud space with your local infrastructure with a “FastConnect” link providing a connection with a bandwidth of at least 10Gb.
Those connectivity options allow you to cover all scenarios and avoid being locked-in in the Oracle Cloud for instance. Which may not be a confortable situation in the long-term, right? It also gives you ways to transform your Cloud space into an extension of your Data Center (FastConnect):
Ok, you’ve spent the day to configure the perfect architecture for the new company’s project, and it’s time to estimate the budget upfront for your financial department. I’m personally more comfortable knowing upfront my budget and hate lousy surprise. Aren’t you?
The Oracle cloud gives you two options to pay: the “Pay as You Go” or “Pay Monthly” subscriptions. You can forecast precisely the price in the Oracle price estimator
Before buying, take time to compare what would be the price from other Cloud providers. Each one of them has its online price calculator:
- Microsoft Azure: https://azure.microsoft.com/en-in/pricing/calculator
- Amazon AWS: https://calculator.s3.amazonaws.com/index.html
- Google Cloud: https://cloud.google.com/products/calculator/?hl=fr
For example for a virtual server will pay monthly (pay-as-you-go):
- Microsoft Azure (DS13v2 | 8 vCPU | 56GB ): 576,83 $
- Oracle Cloud (B88514 | 8 vCPU | 120GB ): 396.00 $
- Amazon AWS (f1.x2large | 8 vCPU | 122GB): 1207.00 $
- Google Cloud (n1-highmem-8 | 8 vCPU | 52GB ): 241.81 $
If you already know upfront the hours and days when your platform is going to run, cloud providers propose a monthly subscription. It makes life much more easier and they will reward you with more or less 30% discount. For instance, Oracle describes that here.
You made it happen!
After a day of provisioning, you have now your infrastructure, the connectivity, and you know the price. The platform is ready, and you can delight the business department with a brand new platform ready to support the launch of the new product. Great work, right? Now the product has been sold for months, and the business tells you to remove the platform. Again from your cloud customer portal, you can destroy all the resources which immediately stops the billing.
The cloud paradigm provides the fastest way to provide digital experience and reduce the time to market. Because of the physical abstraction and the automation it provides, it becomes simplier to create digital experience than ever before. But bad advice and misinterpretation could become costly, whereas the good choices generate incomes for your business. When you need acceleration to move into the cloud properly, we will be here to support you with our experience working with different cloud providers.
Have a question? Please comment below, and we’ll answer that shortly.