Method and apparatus for multiset membership testing using combinatorial bloom filters
View Patent ↗A method and apparatus providing improved set membership determination and group membership identification of candidate data elements using a single Bloom filter programmed to provide a plurality of non-zero f-bit binary vectors, where each of the f-bit binary vectors is associated with a respective group. The Bloom filter is programmed using one or more (but not all) of a plurality of hash filter sets.
1. A method for identifying group membership of a candidate data element in a data processing apparatus including at least a memory circuit, the method comprising:
providing in the memory circuit a plurality of hash sets operative to program a single Bloom filter with a plurality of non-zero f-bit binary vectors, each of the f-bit binary vectors being associated with a respective group indicative of group IDs for one or more candidate data elements;
performing group membership probes into the Bloom filter using candidate data elements to generate thereby f-bit binary vector output signal indicative of a matching function.
2. The method of claim 1 , wherein group membership probes into the Bloom filter are performed by hashing candidate data elements according to each of h hashing functions of each off sets of hashing functions.
3. The method of claim 2 , wherein the Bloom filter is programmed by hashing each data element of a group with each of the h hashing functions of a subset θ of the f sets of hashing functions.
4. The method of claim 3 , wherein a group membership probe resulting in an f-bit binary vector output signal having fewer than θ bits indicates a no match condition.
5. The method of claim 3 , wherein a group membership probe resulting in an f-bit binary vector output signal having more than θ bits indicates a misclassification error.
6. The method of claim 1 , wherein the candidate data element comprises a media access control (MAC) address associated with a data frame, and the group comprises an output port of a switching device associated with the MAC address.
7. The method of claim 6 , wherein the switching device comprises a Layer-2 switch.
8. The method of claim 1 , wherein the candidate data element comprises a data packet and the group comprises a classification associated with the packet.
9. The method of claim 1 , wherein the group comprises a pattern associated with the data element.
10. The method of claim 9 , wherein the data element comprises one of a data packet and a data frame.
11. The method of claim 1 , further comprising:
error correcting the f-bit binary vector output signal using any of Hamming codes, forward error correction (FEC), checksums and parity bits.
12. An apparatus for identifying group membership of a candidate data element, comprising:
a first memory, for storing a plurality of hash sets operative to program a single Bloom filter with a plurality of non-zero f-bit binary vectors, each of the f-bit binary vectors being associated with a respective group indicative of group IDs for one or more candidate data elements; and
a second memory, for storing f sets of h hashing functions;
wherein group membership probes into the Bloom filter are performed by hashing candidate data elements according to each of the h hashing functions of each of the f sets of hashing functions.
13. The apparatus of claim 12 , wherein the Bloom filter is programmed by hashing each data element of a group with each of the h hashing functions of a subset of the f sets of hashing functions.
14. The apparatus of claim 12 , wherein said apparatus further comprises a decoder circuit adapted to decode f-bit binary vectors provided by the first memory to determine thereby a group associated with a candidate data element.
15. The apparatus of claim 12 , wherein the Bloom filter is programmed by hashing each data element of a group with each of the h hashing functions of a subset θ of the f sets of hashing functions.
16. The apparatus of claim 15 , wherein a group membership probe resulting in a binary vector having fewer than θ bits indicates a no match condition, and a binary vector having more than θ bits indicates a misclassification error.
17. The apparatus of claim 14 , wherein the first memory, second memory and decoder circuit are implemented in a programmable logic device.
18. The apparatus of claim 17 , further comprising a controller, for selectively placing the apparatus in one of a programming operational state and a testing operational state.
19. A computer-readable non-transitory medium storing a software program, that, when executed by a computer, causes the computer to perform a method for identifying group membership of a candidate data element, the method comprising:
providing in the memory circuit a plurality of hash sets operative to program a single Bloom filter with a plurality of non-zero f-bit binary vectors, each of the f-bit binary vectors being associated with a respective group indicative of group IDs for one or more candidate data elements;
performing group membership probes into the Bloom filter using candidate data elements.
20. The computer-readable medium of claim 19 , wherein group membership probes into the Bloom filter are performed by hashing candidate data elements according to each of h hashing functions of each off sets of hashing functions.
21. The computer-readable medium of claim 20 , wherein the Bloom filter is programmed by hashing each data element of a group with each of the h hashing functions of a subset of the f sets of hashing functions.