Countries, regions, availability zones and pools
With Servercore's products and services, you can create an infrastructure with all fault tolerance and security requirements taken into account.
Servercore's infrastructure is deployed in four countries and is divided into 6 regions, 9 availability zones and 22 pools.
To see the availability of Servercore products and services in different regions, please refer to the availability matrices.
Country
A country (example: Russia) is one or more regions that are located within a single state.
In most cases, the division into countries does not affect anything and the main division of infrastructure is at the regional level.In some cases, regions in one country are separated from regions in other countries:
- When using a global router, you cannot configure network connectivity between products that are located in different countries. You can only configure product and service connectivity between different regions within the same country;
- the object store uses different domains for each country, and TLS (SSL) certificates are tied to the country — the certificate works only for containers hosted in the region of the certificate's country.
Region
A region (example: St. Petersburg) is a number of data centers, Servercore and partner data centers, which are located within the agglomeration of a large city. Data centers in one region are connected by a fiber-optic line with data redundancy, and minimal signal delay is ensured between data centers within the region.
Each region is completely isolated from the others: separate connection to power lines, autonomous power and cooling sources, dedicated communication channels.
Servercore provides connectivity between regions through redundant high-capacity communication channels.
To host your application, we recommend choosing the region closest to your users as the main region. You can deploy resources in different regions — then the infrastructure will be able to survive possible ISP failures, such as internet outages at the border level, natural disasters and emergencies with minimal down time.
See the Servercore Infrastructure table for a list of regions.
Accessibility zone
An Availability Zone (example: Availability Zone 1) is one or more data centers within the same region. Each data center is equipped with autonomous power and cooling sources and its own redundant ultra-low latency link.
Servercore backs up all critical elements of the system in the availability zone: power supply, air conditioning, network infrastructure. Each availability zone is provided with round-the-clock physical security and monitoring. Equipment maintenance operations are not performed simultaneously in multiple Availability Zones in the same region.
We recommend reserving critical parts of your infrastructure in multiple Availability Zones. Availability zones do not have a single point of failure — only the point of failure for the entire region. This will protect your system in the event of a failure in one of the zones, such as a power outage in the data center, fire, or weather events.
See the Servercore Infrastructure table for a list of availability zones.
Poole
A pool (example: ru-9, SPB-4) is a part of infrastructure in one data center. There are several pools located in each data center and, accordingly, in the availability zone. Each pool is isolated from hardware and software failures in other pools.
Network connectivity within a pool is provided by an L2 connection. A Servercore Global Router can be used to configure connectivity between pools.
Placing infrastructure in multiple pools protects against network hardware and software failures (e.g., network hardware failure or power failure), reduces the likelihood of data loss and service downtime. The single point of failure of pools is failures in the availability zone.
Pools are divided into pool segments (example: ru-2a, ru-8a) — groups of equipment in different racks within one data center.
See the Servercore Infrastructure table for a list of pools and segments.
Increase fault tolerance
You can improve infrastructure resiliency at the region, availability zone, pool, and additionally at the server level:
- For a dedicated server, move the server to another rack;
- cloud server — place servers on different hosts in a placement group.
Servercore Infrastructure
Russia
Uzbekistan
Kazakhstan
Kenya
St. Petersburg
Moscow
Novosibirsk
Tashkent
Almaty
Nairobi