The cloud is a service that provides access to computing resources (servers, storage, and networking) that can be used remotely for specific tasks.
There are three main types of cloud services: IaaS (Infrastructure-as-a-Service), PaaS (Platform-as-a-Service), and SaaS (Software-as-a-Service). Most large companies are already using them for their needs, gradually moving their business processes to the cloud.
Data should be migrated carefully, as it is one of the most valuable assets of any company. In this article, we will explore the process of data migration to the cloud, without losses, and with minimal downtime.
Why business migrate to the cloud?
In general migrating to the cloud gives access to virtually limitless computer resources, improves performance, and security. But there are also some specific reasons:
- Migration due to regulatory requirements. For example, if the law prohibits the cross-border transfer of personal data.
- Flexible business scaling. Migrating to the cloud is always about growth. A business that is growing and needs more resources requires an environment where those resources are provided on-demand and as quickly as possible.
- A radical change of storage architecture. For example, changing the type of DBMS.
- Gaining benefits from new technologies. For example, migrating the DBMS to cloud databases with the connection of automatic backups and disaster recovery systems – to take storage security and usability to a new level.
- Improving the speed of storage systems and/or reducing the cost of storage maintenance by implementing preconfigured cloud solutions.
Ways of migration
There are two main ways to migrate data and services to a cloud provider.
Full migration: In this case, the entire IT infrastructure is migrated to the cloud provider: data, applications with all their settings. The customer receives terminal clients, through which the employees access services.
Partial migration: A hybrid model, meaning that a part of information systems and resources are moved to the cloud, and the remaining data is kept on the company’s computing capacities for security, legal or other reasons.
Partial migration also refers to the gradual migration of certain elements of the IT infrastructure to a virtual platform. This model is suitable for companies with a geographically distributed structure (a large number of remote offices). In this case, a detailed migration roadmap is prepared and services are migrated consistently – so that the execution of business processes in the organization remains uninterrupted. This is the so-called seamless migration, which is possible due to clear planning and consistent execution of predefined steps.
Procedure of migration
A company that is planning to migrate all or part of its IT infrastructure to the cloud can develop its migration plan or use a generally accepted scheme. In any case, it must be adjusted to the specifics of the business, the volume of data, the geographically distributed network of offices, and other factors. The generally accepted scheme can be divided into several typical steps.
Helps to assess the current IT-infrastructure of the company, identify priority components, and establish dependencies between them, which should be taken into account during migration. In the process of inventory, it is possible to discover problem areas, which require maximum attention during the migration of data and services.
Selecting a cloud model
There are three types of cloud deployment models
Public cloud: the hosting provider delivers cloud services to a large number of customers, compute, networking, and storage resources are housed in one large (or multiple) data center. A public cloud is necessary for flexibility and automation of routine tasks. it is popular among website and online service owners, companies that develop and test software;
Private cloud: a company builds its own cloud infrastructure and the resources which will be used to run services in the cloud belong only to the company. The greatest benefit from a private cloud is gained by large companies with an extensive IT infrastructure and which work with sensitive data or state secrets;
Hybrid cloud: A symbiosis of a private cloud and a public cloud, where the exchange of data between them is strictly controlled by the company. Hybrid cloud is advantageous where the business faces sudden bursts of activity (for example, in days of sales or advertising campaigns), where complex technological solutions need to be implemented after preliminary testing, where elements of the IT infrastructure need to be tested, or where new software products are being developed.
After selecting the appropriate model, calculate the required resources: processing power, type, and storage capacity, the width of the Internet channel, the number of IP-addresses, etc.
Choosing a cloud provider
This is one of the most important tasks when preparing for cloud migration. You need to ensure that the provider is reliable, offers an appropriate infrastructure, and provides exactly the services and tools that your business needs. Ideally, the provider offers a test period. This will help you evaluate the potential of the site on real or test projects.
Conclude a Service Level Agreement (SLA). In the document, the provider specifies the time during which customer services may not be available and assumes the issues related to technology and software failures in the data center.
Creating a roadmap
A correct roadmap will allow you to plan your migration and avoid critical mistakes. It is important to describe not only the sequence of actions but also the deadlines for completing each task. Critical services are migrated when they are least utilized (e.g. at night). The ideal result is to make the migration seamless so that your customers will not notice it.
To sum up, migration consists of the following steps:
- Inventory of stored data and software to process it by a third party.
- Assessing the possibility of data processing by a third party (you can choose a hybrid model and keep some data on your server).
- Selecting a cloud provider.
- Making a plan for the step-by-step migration of non-critical information to a third-party cloud provider.
- Migration and testing of services in new environments, as well as integration with the data that remains on your side.