Connectivity is the basis. If the network is not stable, it directly affects your application environment. Availability, performance and security are criteria we used to design and build our network. Everything is redundant with so much capacity available that you do not have to worry about it.
The advantages we offer with networking are:
Network function virtualization
High capacity with burst capability
Orchestration as a Service (HEAT)
Load Balancing as a Service
Within our OpenStack, the network is powered by Neutron, and we have used OpenContrail to provide full network function virtualization, secure and with many functionalities for optimal flexibility.
The high availability of our network is realized by making all components within our OpenStack platform redundant. We guarantee performance through a huge network capacity to the Internet. Within our availability zones, all compute nodes are equipped with two redundant 10Gbps links to a redundant switch environment with a 4x40Gbps backbone.
You can segment networks in OpenStack and determine which instances you want to connect to. You do this manually or automatically via the OpenStack API or Heat Orchestration as a Service. Furthermore, you do not need to implement load balancing yourself as you can use the Load Balancing as a Service (LBaaS).
Our two availability zones are interconnected and you decide how to distribute your application. Traffic that occurs between the instances, both within the availability zone and in between, will not be charged. Even traffic from and to the Internet will not be charged. We use a fair use policy instead.
The traffic between your instances, both within and between the availability zones, is unlimited and based on a fair use policy. The same applies to traffic to and from the Internet.
As fair use, we use a maximum consumption of 4x the average consumption of customers with the same flavor.
You can assign an IP address to each instance you start. However, you may want to reserve and continue using certain IP addresses, because they are used, for example, in the DNS of for the connection from users.
Within OpenStack, you can reserve IP addresses so that they are always available when you need them. You can assign these floating IPs to an instance. If you do not need the floating IP temporarily, but you want to keep it reserved, then this is possible. At any time, you can use the floating IP again for the same or another instance.
You have full control over the network through the Horizon dashboard and the OpenStack API. You can easily configure the network and ensure that your instances have Internet access, or fully segment groups of instances. For example, you can make your web servers publicly available with a private network segment where you place your database and data instances.
By default, we protect your environment by blocking full access. Security groups allow you to decide which ports should be open for one or more instances and from where. You can set this up for both inbound and outbound traffic. Additionally, you can set network policies for networks. This gives you even more control over the traffic you may or may not allow. You have the ability to control traffic based on the type of traffic, protocol used, traffic ports, and traffic direction through network policy rules.
Security groups and network policies can easily be created via the Horizon dashboard or automated with the OpenStack API.
When you add multiple networks, routers and subnets it can make things unclear. Within the Horizon dashboard, we provide you with a schematic and interactive overview. You can see exactly how the various components are connected and which instances you have linked.