Method of creating source IP address bunches in a router, and router comprising a network routing system comprising source IP address bunches

A method of creating a scalable source IP address bunch in a router, the method comprising:

• constituting a source IP address block in the router;
• associating the source IP address block using an association function to form a source IP address bunch in the router;
• and
• allocating, splitting, merging, and aggregating the source IP address bunch in the router to reduce a storage size of a routing table stored in the router, and releasing storage space in the router for storing IP information of additional network devices;
• wherein
• the source IP address block is a sequence of continuous or discontinuous internet IP addresses having a size of n or 2k (k=log n), n IP address sequences being a1, a2, an, where ai+1=ai+1, i=1, 2, n−1, a1 to an are divisible by 2;
• the source IP address bunch means a set of r continuous or discontinuous source IP address blocks in the router, a first source IP address block S0 is called a root block of the source IP address bunch, and a handle of the root block is a handle of the source IP address bunch;
• associating the source IP address block through the association function to form the source IP address bunch in the router comprises linking the r continuous or discontinuous source IP address blocks to each other and forming the source IP address bunch using the association function in the router;
• the association function comprises a linear association and an exponential association;
• the linear association refers to linking the r continuous or discontinuous source IP address blocks of a size n through a linear relationship;
• the exponential association refers to linking the r continuous or discontinuous source IP address blocks of the size n through an exponential relationship;
• a rule of allocating the source IP address bunch comprises:
• 1) any root block of two source IP address bunches are not the same;
• 2) overlap is allowed between domains of a linear association of the source IP address bunch and an exponential association of the source IP address bunch;
• 3) no overlap between domains of two linear associations of the source IP address bunch is allowed;
• 4) overlaps between domains of two exponential associations of the source IP address bunch are allowed;
• 5) domains of different sizes should be interleaved;
• and 6) domains of the same size can overlap;
• allocating the source IP address bunch is in accordance with a best matching rule, which refers to minimizing the difference between an actually allocated amount and a requested amount;
• splitting the source IP address bunch refers to splitting a large source IP address bunch into a number of small source IP address bunches;
• merging the source IP address bunch refers to merging a number of the small source IP address bunches into the large source IP address bunch;
• and
• an algorithm of aggregating the source IP address bunch comprises a first algorithm using an association A, an exponential base b, a filter symbol M, and an offset O to express a relationship between a specific IP address D and a source IP address bunch containing the specific IP address D, and a second algorithm using the specific IP address D, the association A, the exponential base b, the filter symbol M, and the offset O to calculate the handle of the source IP address bunch comprising the specific IP address D and to get a handle Hs of the source IP address block comprising the specific IP address D.