Using dynamic policies for flexible address selection
Publication Date: 2017-May-25
The IP.com Prior Art Database
Using Dynamic Policies for Flexible Address Selection
In DSN memory, each device is identified by one or more internet protocol network address. The configuration information for DSN memory provides the network addresses of ds unit as unordered set. The ds processing unit treats all addresses in the set equally. It then creates connection on each address of ds unit to send data. Having multiple connections between ds processing unit and ds unit increases the throughput and make the communication more resilient. However having a set of addresses provides very little flexibility for ds processing unit to select a particular address. There are various scenarios, such as following, in which the ds processing unit may want to prefer one address over another: - If more than one address of ds unit belongs to a same network interface then it might be unnecessary to open connections to all addresses. Additionally in some cases it may affect the throughput adversely. In this case ds processing unit needs to distinguish between addresses belonging to unique network interfaces. - Reports indicate that biasing towards IPv6 is now beneficial for customers; IPv6 is now mainstream instead of being an exception, there are less broken IPv6 tunnels, IPv4 carrier-grade NATs are increasing in numbers, and throughput may even be better on average over IPv6. The ds processing unit may want to select IPv6 address only. In this case ds processing needs more information about the network protocol information about each address. In order to provide the flexibility in address selection the configuration information of DSN memory provides additional information about each address of ds unit. The addresses can be grouped based on the network interfaces and ordered by the p...