How to choose a cloud service provider?
The cloud services market is growing rapidly, and Russia is no exception. Users of public clouds can easily get the necessary IT resources and even a whole virtual data center without serious costs, paying only for what they really use. The cloud model allows you to save not only on capital investments, but also on support, since all non-core business tasks are outsourced. Cloud solutions do not require traditional IT costs, hardware and software.
The possibility of cost savings, or, to put it more correctly, the transfer of CAPEX to OPEX, is especially relevant in times of crisis. Many companies are actively optimizing IT budgets and infrastructure, and the cloud model has good potential in this regard. And for small and medium businesses, cloud services provide opportunities that were previously available only to large companies with a solid IT budget.
Not surprisingly, more and more organizations are moving their applications to the cloud. According to the Oracle research , cloud technologies help Russian companies to introduce innovations faster (96%), retain existing (80%) and attract new customers (92%). More than half of the companies (55%) say that they will reach “cloud maturity” in the next two years.
SAP and the National Agency for Financial Research (NAFI) presented the results of a study of cloud technologies in the banking sector, according to which Russian banks are both technologically and economically ready for a cloud boom. And despite the fact that cloud technologies are traditionally associated with such public services as mail, data storage or platform infrastructures, for almost a third (32%) of cloud experts are business applications.
')
As a result, more and more companies are starting to offer cloud services. Today, there is a fairly large number of offers from cloud service providers and owners of commercial data centers. Experts believe that the Russian market of IaaS, SaaS and PaaS will only grow in the next 3-4 years. The cloud market in recent years has shown an increase of about 30% per year. So the IaaS segment is growing by almost 50% every year, and within three years it will increase by another 50-60%.
The main drivers of growth are the worsening economic situation and the need to save it, optimize IT infrastructure costs and software licensing, the high cost of purchasing new equipment due to changes in exchange rates. IT budgets will be reduced, so moving to the cloud will be one of the key savings factors.
In the current economic situation in Russia, most companies are moving from long-term planning to short-term and medium-term. They strive to reduce costs and better and more efficient use of existing resources. In this sense, cloud solutions, virtualization, and virtual data processing centers (VDC) provide the most efficient use of existing infrastructure. They allow you to more accurately assess the return on investment and significantly reduce the cost of maintaining the IT infrastructure.
Benefits of a public cloud include:
Of course, this is not a complete list. The main thing is that the cloud does not always provide cost savings (rather, we can talk about their restructuring), especially over a long time interval, but it allows you to become more flexible and get better service for the same money.
The use of cloud solutions for crisis management purposes is relevant both for the commercial sector - companies from different industries, and for government organizations. The competence of customers in relation to cloud services over the past year or two has grown significantly: companies are well aware of the benefits of cloud services, most of them know exactly what they are willing and able to get for their money by choosing the optimal ratio of price and quality of services.
However, the choice of provider is a responsible process. Finding the “optimal” cloud provider for a particular application is becoming increasingly difficult. And an attempt to introduce a cloud service, which actually does not suit your company, can take a lot of time and nerves.
And the proposals themselves are becoming more diverse - from low-cost standardized, “bundled” services to high-level and complex services with additional redundancy, up to and including disaster recovery. Often, companies try to optimize staff, shifting most of the routine tasks and operations to an outsourcing company, and providers often offer their services along with support for business services. However, the main requirements remain reliability, safety, high performance and affordability of services at a price.
So what should we look for when choosing a public cloud service provider? Of course, everything depends on the requirements of the business, the nature and criticality of the tasks to be solved. Nevertheless, we will try to give general recommendations.
When choosing an external service provider for most companies, the key criterion is compliance with the maximum levels of reliability. A very important indicator is the reliability of the entire data center, the site on which the cloud platform is located. Ultimately, it determines the availability of cloud services. Many cloud providers are ready to show their sites to potential customers so that they can see firsthand their reliability and the availability of all the subsystems needed in a high-end data center.
One way to check the reliability of the service provider’s data center will be to check for Uptime Institute certificates . Best of all, when a company owns not one data center, but a network of data centers built in accordance with the requirements of TIA-942 (TIER III) international standards. Along with such a formal criterion as the level of TIER, it is worthwhile to focus on the experience of the smooth operation of the provider - how long its data center works without failures, whether there were serious accidents.
The performance and reliability of a cloud platform largely depends on what IT systems the provider uses. It is important what equipment the service provider cloud is built on (servers, storage, network equipment). It is believed that the use of low-end servers by a cloud provider increases the likelihood of their failure, reduces reliability and affects service continuity.
The smooth operation of the data center requires the use of modern IT equipment. As a rule, the higher the class of equipment, the more functional and productive the cloud platform of the provider. Reliability in it is increased due to duplication, redundancy and the lack of a single point of failure. The same applies to storage systems. All network equipment must also be duplicated. For uninterrupted operation, it is necessary to reserve all engineering systems, monitor and maintain all life support systems 24x7x365. As a result, the failure of any single piece of equipment does not affect the performance of the complex as a whole.
What virtualization software are cloud services based on? It is worth paying attention to the partner statuses of the provider: it is good if its activity is carried out in close cooperation with world leaders in the IT industry. The presence of high status (for example, VMware) indicates how experienced a service provider is in this area. Using the cloud provider of industry-leading virtualization solutions improves the reliability, security and stability of the cloud.
For companies that already use virtualization and want to bring their applications to the cloud, it is important that the hypervisor used by the cloud provider is the same as the one already used in the company. This will simplify the migration process, facilitate the construction of a hybrid cloud, if the need arises, and also allows you to maximize the functionality of integrating your own server capacity with the facilities leased in the cloud.
To say which of the clouds will allow you to achieve the best performance is not easy. You can focus on the specifications provided by the providers themselves, but usually they are rather fuzzy and do not cover all aspects of the cloud.
Meanwhile, cloud performance is not only computing speed, but also I / O performance when accessing storage systems (it is this that has a greater effect on the effective operation of the application), bandwidth of communication channels when accessing services, latency parameters and much more. Performance also depends on the load created by other cloud users. In the best case, the cloud provider guarantees I / O performance indicators (IOPS and latency parameters).
Accidents are unavoidable in any technically complex systems, but they can be made not critical for business. To prevent such situations, disaster-resistant systems are being created — reserve capacity is being deployed at a geographically remote site.
An important business task is to improve the quality of service due to the fault tolerance of cloud solutions. The availability of a backup site, the ability to stretch the cloud into two data centers allows the provider to provide disaster protection and disaster recovery services (Disaster Recovery as a Service, DRaaS) and even a whole range of design and backup services for the customer’s primary data center. The main data center of the customer can be either a physical infrastructure located on the customer’s site or a virtual IT infrastructure.
DRaaS implies redundancy at multiple sites connected by optical channels. In case of failure of individual pieces of equipment, automatic restarting of systems on healthy servers begins. And as a result, the customer system in the cloud continues to operate as before.
What regulations does the provider use, how is the documentation organized, what is the responsibility of the service provider? It is advisable to get answers to these questions.
A cloud provider can provide high availability only with well-organized and formalized processes for maintaining and maintaining its cloud infrastructure, including managing service requests, incidents and changes. This will minimize the impact of incidents on the client’s business and recovery time.
The support service should provide for continuous availability in 24/7/365 mode, receiving calls through different channels, monitoring and adherence to key metrics - response time to a call, the maximum time to solve an incident, and to solve a typical request. The provider must also be ready to help solve problems that go beyond the contract.
The package of documents may include documents that the provider uses in its work, provides the customer, documents for the person responsible for security and the administrator of the VM, documents for the user.
The agreed service level is fixed in the SLA - Service Level Agreement. For violations of the SLA parameters, the cloud provider must be financially responsible. In addition, the provider must be prepared to discuss quality parameters.
At the site of the data center, specialists should be on duty, always ready to carry out maintenance, configuration and configuration of the customer’s systems. Availability of a 24-hour cloud support engineer (24x7x365) is required. Typical service requests are usually specified in the service description, and the target parameters of the solution time are recorded in the SLA. For non-standard requests, the planned solution time is also determined.
A reliable provider infrastructure should provide for the rapid deployment of the cloud, flexible pricing and service management. Therefore, the presence of a self-service portal (personal account), debugged and certified billing, flexibility of tariff plans and the ability to quickly increase or decrease the pool of IT resources are required.
With the right approach to building a line of services, complementing them with valuable services for the customer, ensuring the necessary level of security and reliability with acceptable pricing policies, cloud services will allow companies to develop dynamically, increase productivity, adapt services in accordance with the needs and capabilities of their organization, choose the optimal tariff plan and a set of additional options.
This is just a brief list of requirements for a cloud provider. In reality, the choice has to take into account much more nuances, and the requirements of each customer are individual. And, of course, you should pay attention to the experience of the provider in the market and the feedback from other customers.
The Russian cloud services market is developing dynamically, but there are still relatively few providers that meet all the listed criteria. In the general case, it is recommended, first of all, to identify those providers that will be able to solve the client’s tasks with greater efficiency and less risk to the business. Some providers have been working in the market for more than ten years and have managed to form a team of specialists aimed at maximally meeting customer needs and provide all the necessary IT resources to solve business problems of any complexity and orientation.
With the growth of the cloud services market, more and more data centers are starting to provide virtual infrastructure rental services in addition to equipment placement services. A cloud provider, by definition, provides not only such basic data center services as colocation, equipment rental and IT systems (dedicated), and room rental, but also cloud solutions .
This could be the creation of an IT infrastructure based on a “virtual platform”, a virtual data center (including meeting the requirements of the Federal Law 152 ), rental of virtual workplaces ( VDI ), backup data center services, SaaS services (e-mail, etc.). Of course, the list and service names of different providers differ, but the complexity of the proposals is of great importance.
Sometimes these services complement telecommunications services, the provision of Internet access, the organization of L2 channels over a data network, the rental services of fiber-optic lines, and even managed CDNs.
What about protection against DDoS attacks ? In recent years, they have reached the first place in the list of network threats for corporate clients and telecom operators. Protection against them is an actual problem not only for every Internet business, but for many other companies using web resources. As a rule, such attacks are aimed at disabling critical applications and resources of the owner. Today, the means to protect customers from such attacks is a necessary element in the arsenal of any large data center operator.
Of particular value are packages of services: firstly, such complex services are often integrated, which gives a synergistic effect, and the customer can receive them at a discount. A significant advantage will be additional services (VAD), such as migration to the cloud environment , building a private cloud at the customer or provider site, as well as various services related to the deployment and operation of hybrid clouds. Having additional services, such as backing up virtual machines, is another plus.
SAFEDATA company offers a range of services based on its own data centers. The company is the only player in the market that combines businesses in three main segments - data center, traffic exchange and content delivery.
According to research , the majority of Russian customers are planning to expand the range of cloud services used. Therefore, the ability to integrate new services with the already sought-after ones, a convenient payment scheme for expanding services and complex offers become additional factors that increase the attractiveness of solutions developed by providers.
Choosing the right cloud provider will enable you to fully realize the potential of efficiency built into the cloud ecosystem, achieve economies of scale and outsource IT administration, allow flexible utilization of the computing power of cloud servers at geographically distributed sites, minimize risks due to reliable Tier III data center infrastructure, the use of equipment and software by the leading vendors, high qualifications and experience of its specialist ialistov.
Crisis as a growth factor
The possibility of cost savings, or, to put it more correctly, the transfer of CAPEX to OPEX, is especially relevant in times of crisis. Many companies are actively optimizing IT budgets and infrastructure, and the cloud model has good potential in this regard. And for small and medium businesses, cloud services provide opportunities that were previously available only to large companies with a solid IT budget.
Not surprisingly, more and more organizations are moving their applications to the cloud. According to the Oracle research , cloud technologies help Russian companies to introduce innovations faster (96%), retain existing (80%) and attract new customers (92%). More than half of the companies (55%) say that they will reach “cloud maturity” in the next two years.
SAP and the National Agency for Financial Research (NAFI) presented the results of a study of cloud technologies in the banking sector, according to which Russian banks are both technologically and economically ready for a cloud boom. And despite the fact that cloud technologies are traditionally associated with such public services as mail, data storage or platform infrastructures, for almost a third (32%) of cloud experts are business applications.
')
As a result, more and more companies are starting to offer cloud services. Today, there is a fairly large number of offers from cloud service providers and owners of commercial data centers. Experts believe that the Russian market of IaaS, SaaS and PaaS will only grow in the next 3-4 years. The cloud market in recent years has shown an increase of about 30% per year. So the IaaS segment is growing by almost 50% every year, and within three years it will increase by another 50-60%.
The main drivers of growth are the worsening economic situation and the need to save it, optimize IT infrastructure costs and software licensing, the high cost of purchasing new equipment due to changes in exchange rates. IT budgets will be reduced, so moving to the cloud will be one of the key savings factors.
In the current economic situation in Russia, most companies are moving from long-term planning to short-term and medium-term. They strive to reduce costs and better and more efficient use of existing resources. In this sense, cloud solutions, virtualization, and virtual data processing centers (VDC) provide the most efficient use of existing infrastructure. They allow you to more accurately assess the return on investment and significantly reduce the cost of maintaining the IT infrastructure.
Benefits of a public cloud include:
- Cost optimization and effective cost control.
- Getting access to IT resources in a short time.
- Releasing your own resources.
- Increase business efficiency and focus on customer service.
- Accelerating the reorganization of the company.
- Reduced project time.
- Gaining access to IT infrastructure management expertise in the absence of in-house IT staff.
- Quick access to new markets.
Of course, this is not a complete list. The main thing is that the cloud does not always provide cost savings (rather, we can talk about their restructuring), especially over a long time interval, but it allows you to become more flexible and get better service for the same money.
Who and how to choose?
The use of cloud solutions for crisis management purposes is relevant both for the commercial sector - companies from different industries, and for government organizations. The competence of customers in relation to cloud services over the past year or two has grown significantly: companies are well aware of the benefits of cloud services, most of them know exactly what they are willing and able to get for their money by choosing the optimal ratio of price and quality of services.
However, the choice of provider is a responsible process. Finding the “optimal” cloud provider for a particular application is becoming increasingly difficult. And an attempt to introduce a cloud service, which actually does not suit your company, can take a lot of time and nerves.
And the proposals themselves are becoming more diverse - from low-cost standardized, “bundled” services to high-level and complex services with additional redundancy, up to and including disaster recovery. Often, companies try to optimize staff, shifting most of the routine tasks and operations to an outsourcing company, and providers often offer their services along with support for business services. However, the main requirements remain reliability, safety, high performance and affordability of services at a price.
So what should we look for when choosing a public cloud service provider? Of course, everything depends on the requirements of the business, the nature and criticality of the tasks to be solved. Nevertheless, we will try to give general recommendations.
1. Reliability data center
When choosing an external service provider for most companies, the key criterion is compliance with the maximum levels of reliability. A very important indicator is the reliability of the entire data center, the site on which the cloud platform is located. Ultimately, it determines the availability of cloud services. Many cloud providers are ready to show their sites to potential customers so that they can see firsthand their reliability and the availability of all the subsystems needed in a high-end data center.
One way to check the reliability of the service provider’s data center will be to check for Uptime Institute certificates . Best of all, when a company owns not one data center, but a network of data centers built in accordance with the requirements of TIA-942 (TIER III) international standards. Along with such a formal criterion as the level of TIER, it is worthwhile to focus on the experience of the smooth operation of the provider - how long its data center works without failures, whether there were serious accidents.
2. Hardware platform
The performance and reliability of a cloud platform largely depends on what IT systems the provider uses. It is important what equipment the service provider cloud is built on (servers, storage, network equipment). It is believed that the use of low-end servers by a cloud provider increases the likelihood of their failure, reduces reliability and affects service continuity.
The smooth operation of the data center requires the use of modern IT equipment. As a rule, the higher the class of equipment, the more functional and productive the cloud platform of the provider. Reliability in it is increased due to duplication, redundancy and the lack of a single point of failure. The same applies to storage systems. All network equipment must also be duplicated. For uninterrupted operation, it is necessary to reserve all engineering systems, monitor and maintain all life support systems 24x7x365. As a result, the failure of any single piece of equipment does not affect the performance of the complex as a whole.
3. Virtualization Software and Hypervisor
What virtualization software are cloud services based on? It is worth paying attention to the partner statuses of the provider: it is good if its activity is carried out in close cooperation with world leaders in the IT industry. The presence of high status (for example, VMware) indicates how experienced a service provider is in this area. Using the cloud provider of industry-leading virtualization solutions improves the reliability, security and stability of the cloud.
For companies that already use virtualization and want to bring their applications to the cloud, it is important that the hypervisor used by the cloud provider is the same as the one already used in the company. This will simplify the migration process, facilitate the construction of a hybrid cloud, if the need arises, and also allows you to maximize the functionality of integrating your own server capacity with the facilities leased in the cloud.
4. Performance
To say which of the clouds will allow you to achieve the best performance is not easy. You can focus on the specifications provided by the providers themselves, but usually they are rather fuzzy and do not cover all aspects of the cloud.
Meanwhile, cloud performance is not only computing speed, but also I / O performance when accessing storage systems (it is this that has a greater effect on the effective operation of the application), bandwidth of communication channels when accessing services, latency parameters and much more. Performance also depends on the load created by other cloud users. In the best case, the cloud provider guarantees I / O performance indicators (IOPS and latency parameters).
5. Resilience assurance
Accidents are unavoidable in any technically complex systems, but they can be made not critical for business. To prevent such situations, disaster-resistant systems are being created — reserve capacity is being deployed at a geographically remote site.
An important business task is to improve the quality of service due to the fault tolerance of cloud solutions. The availability of a backup site, the ability to stretch the cloud into two data centers allows the provider to provide disaster protection and disaster recovery services (Disaster Recovery as a Service, DRaaS) and even a whole range of design and backup services for the customer’s primary data center. The main data center of the customer can be either a physical infrastructure located on the customer’s site or a virtual IT infrastructure.
DRaaS implies redundancy at multiple sites connected by optical channels. In case of failure of individual pieces of equipment, automatic restarting of systems on healthy servers begins. And as a result, the customer system in the cloud continues to operate as before.
6. Customer Service and SLA
What regulations does the provider use, how is the documentation organized, what is the responsibility of the service provider? It is advisable to get answers to these questions.
A cloud provider can provide high availability only with well-organized and formalized processes for maintaining and maintaining its cloud infrastructure, including managing service requests, incidents and changes. This will minimize the impact of incidents on the client’s business and recovery time.
The support service should provide for continuous availability in 24/7/365 mode, receiving calls through different channels, monitoring and adherence to key metrics - response time to a call, the maximum time to solve an incident, and to solve a typical request. The provider must also be ready to help solve problems that go beyond the contract.
The package of documents may include documents that the provider uses in its work, provides the customer, documents for the person responsible for security and the administrator of the VM, documents for the user.
The agreed service level is fixed in the SLA - Service Level Agreement. For violations of the SLA parameters, the cloud provider must be financially responsible. In addition, the provider must be prepared to discuss quality parameters.
At the site of the data center, specialists should be on duty, always ready to carry out maintenance, configuration and configuration of the customer’s systems. Availability of a 24-hour cloud support engineer (24x7x365) is required. Typical service requests are usually specified in the service description, and the target parameters of the solution time are recorded in the SLA. For non-standard requests, the planned solution time is also determined.
7. Self-service portal, tariffing and billing
A reliable provider infrastructure should provide for the rapid deployment of the cloud, flexible pricing and service management. Therefore, the presence of a self-service portal (personal account), debugged and certified billing, flexibility of tariff plans and the ability to quickly increase or decrease the pool of IT resources are required.
With the right approach to building a line of services, complementing them with valuable services for the customer, ensuring the necessary level of security and reliability with acceptable pricing policies, cloud services will allow companies to develop dynamically, increase productivity, adapt services in accordance with the needs and capabilities of their organization, choose the optimal tariff plan and a set of additional options.
8. Experience and reputation of the provider
This is just a brief list of requirements for a cloud provider. In reality, the choice has to take into account much more nuances, and the requirements of each customer are individual. And, of course, you should pay attention to the experience of the provider in the market and the feedback from other customers.
The Russian cloud services market is developing dynamically, but there are still relatively few providers that meet all the listed criteria. In the general case, it is recommended, first of all, to identify those providers that will be able to solve the client’s tasks with greater efficiency and less risk to the business. Some providers have been working in the market for more than ten years and have managed to form a team of specialists aimed at maximally meeting customer needs and provide all the necessary IT resources to solve business problems of any complexity and orientation.
What to choose?
With the growth of the cloud services market, more and more data centers are starting to provide virtual infrastructure rental services in addition to equipment placement services. A cloud provider, by definition, provides not only such basic data center services as colocation, equipment rental and IT systems (dedicated), and room rental, but also cloud solutions .
This could be the creation of an IT infrastructure based on a “virtual platform”, a virtual data center (including meeting the requirements of the Federal Law 152 ), rental of virtual workplaces ( VDI ), backup data center services, SaaS services (e-mail, etc.). Of course, the list and service names of different providers differ, but the complexity of the proposals is of great importance.
Sometimes these services complement telecommunications services, the provision of Internet access, the organization of L2 channels over a data network, the rental services of fiber-optic lines, and even managed CDNs.
What about protection against DDoS attacks ? In recent years, they have reached the first place in the list of network threats for corporate clients and telecom operators. Protection against them is an actual problem not only for every Internet business, but for many other companies using web resources. As a rule, such attacks are aimed at disabling critical applications and resources of the owner. Today, the means to protect customers from such attacks is a necessary element in the arsenal of any large data center operator.
Of particular value are packages of services: firstly, such complex services are often integrated, which gives a synergistic effect, and the customer can receive them at a discount. A significant advantage will be additional services (VAD), such as migration to the cloud environment , building a private cloud at the customer or provider site, as well as various services related to the deployment and operation of hybrid clouds. Having additional services, such as backing up virtual machines, is another plus.
SAFEDATA company offers a range of services based on its own data centers. The company is the only player in the market that combines businesses in three main segments - data center, traffic exchange and content delivery.
According to research , the majority of Russian customers are planning to expand the range of cloud services used. Therefore, the ability to integrate new services with the already sought-after ones, a convenient payment scheme for expanding services and complex offers become additional factors that increase the attractiveness of solutions developed by providers.
Choosing the right cloud provider will enable you to fully realize the potential of efficiency built into the cloud ecosystem, achieve economies of scale and outsource IT administration, allow flexible utilization of the computing power of cloud servers at geographically distributed sites, minimize risks due to reliable Tier III data center infrastructure, the use of equipment and software by the leading vendors, high qualifications and experience of its specialist ialistov.
Source: https://habr.com/ru/post/301714/
All Articles