Approximated per-flow rate limiting

A computer system for communications, wherein said computer system comprises:

• a buffer manager receiving a plurality of input flows, each input flow comprising a plurality of packets;
• a controller coupled to said buffer manager and configured to,
• map each of said input flows into a flow table, said flow table comprising a plurality of entries each containing at least a field identifying a credit value wherein said controller to map each of said input flows into a flow table further comprises at least one of, a controller to hash data of said plurality of input flows, and a controller to concatenate data of said plurality of input flows;
• compare said credit value to a quantity signifying insufficient credits;
• process or drop a packet based on said comparison;
• and
• periodically increment said credit value by an increment.

A computer system for communications, wherein said computer system comprises computer instructions for:

• parsing an incoming packet comprising a plurality of fields to generate an extracted plurality of fields;
• transforming said extracted plurality of fields according to an algorithm to generate a result wherein said transforming comprises at least one of, hashing data of said extracted plurality of fields, and concatenating data of said extracted plurality of fields;
• mapping said result according to a mapping function wherein the output of said mapping function is an index corresponding to a flow table entry;
• reading the flow table entry identified by said index, said flow table entry containing at least a field identifying a credit value;
• comparing said credit value to a quantity signifying insufficient credits;
• processing or dropping the packet based on said comparing;
• and
• periodically incrementing said credit value by an increment.

A computer readable storage medium comprising computer instructions for:

• parsing an incoming packet comprising a plurality of fields to generate an extracted plurality of fields;
• transforming said extracted plurality of fields according to an algorithm to generate a result wherein said transforming comprises at least one of, hashing data of said extracted plurality of fields, and concatenating data of said extracted plurality of fields;
• mapping said result according to a mapping function wherein the output of said mapping function is an index corresponding to a flow table entry;
• reading the flow table entry identified by said index, said flow table entry containing at least a field identifying a credit value;
• comparing said credit value to a quantity signifying insufficient credits;
• processing or dropping the packet based on said comparing;
• and
• periodically incrementing said credit value by an increment.