Scalable architectures for reference signature matching and updating

1. An apparatus, comprising:

a first compute node associated with a first subset of reference signatures, the first compute node to:

determine a signature node number for a first site signature of a plurality of site signatures representative of monitored media;

determine that a reference signature corresponding to the first site signature is stored at the first compute node based on a comparison of the signature node number with a first node number of the first compute node; and

in response to the determination that the reference signature is stored at the first compute node, compare a neighborhood of site signatures including the first site signature to reference signatures in the first subset of reference signatures to identify the monitored media, and determine matching information, the matching information including media identification information that identifies reference media corresponding to a matching reference signature; and

a first memory partition of a computer system associated with the first compute node, the first memory partition to store the first subset of reference signatures.

2. The apparatus as defined in claim 1 , wherein the reference signature is a first reference signature, and further including:

a second compute node associated with a second subset of reference signatures different than the first subset of reference signatures, the second compute node to:

determine that a second reference signature corresponding to the first site signature is stored in the second compute node; and

in response to the determination that the second reference signature is stored at the second compute node, compare the neighborhood of site signatures including the first site signature to reference signatures in the second subset of reference signatures; and

a second memory partition of the computer system, different than the first memory partition, the second memory partition to store the second subset of reference signatures.

3. The apparatus as defined in claim 2 , wherein the first compute node and the second compute node operate in parallel to determine the first reference signature is stored at the first compute node and the second reference signature is stored at the second compute node.

4. The apparatus as defined in claim 1 , wherein the first compute node further includes:

a linear list of the first subset of reference signatures; and

a nested distributed map (NDMap), the first compute node to query the NDMap using the first site signature as a key to retrieve an inner map, the inner map including locations of reference signatures corresponding to the first site signature in the linear list, the first compute node to compare the first site signature to the reference signatures corresponding to the first site signature in the linear list.

5. The apparatus as defined in claim 4 , further including a first graphic processing unit (GPU), the first compute node to send the first site signature and the inner map to the GPU, the GPU to compare the first site signature to the reference signatures corresponding to the first site signature in the linear list.

6. The apparatus as defined in claim 4 , wherein the NDMap is a reverse index map.

7. The apparatus as defined in claim 1 , wherein the matching information further includes a matching score.

8. The apparatus as defined in claim 1 , wherein the first compute node is to determine the signature node number for the first site signature based on (1) a number associated with the first site signature and (2) a total number of compute nodes.

9. The apparatus as defined in claim 1 , wherein the first compute node is to perform a first modulo operation based on the first site signature and a total number of compute nodes to determine the signature node number.

10. A tangible machine readable storage medium comprising machine readable instructions which, when executed, cause a first compute node to at least:

determine a signature node number for a first site signature of a plurality of site signatures representative of monitored media;

determine that a reference signature corresponding to the first site signature is stored at the first compute node based on a comparison of the signature node number with a first node number of a first compute node; and

in response to the determination that the reference signature is stored at the first compute node, compare a neighborhood of site signatures including the first site signature to reference signatures in a first subset of reference signatures to identify the monitored media, and determine matching information, the matching information including media identification information that identifies reference media corresponding to a matching reference signature, the first subset of reference signatures stored in a first memory partition of a computer system associated with the first compute node.

11. The tangible machine readable storage medium as defined in claim 10 , wherein the reference signature is a first reference signature, and the machine readable instructions, when executed, further to cause the machine to:

determine that a second reference signature corresponding to the first site signature is stored in a second compute node; and

in response to the determination that the second reference signature is stored at the second compute node, compare the neighborhood of site signatures including the first site signature to reference signatures in a second subset of reference signatures different than the first subset of reference signatures, the second subset of reference signatures stored in a second memory partition of the computer system, different than the first memory partition.

12. The tangible machine readable storage medium as defined in claim 10 , wherein the machine readable instructions, when executed, further cause the machine to:

query a nested distributed map (NDMap) to compare the first site signature to the reference signatures in the first subset of reference signatures, the query to use the first site signature as a key to retrieve an inner map, the inner map including locations of reference signatures corresponding to the first site signature in a linear list of the first subset of reference signatures; and

compare the first site signature to the reference signatures corresponding to the first site signature in the linear list.

13. The tangible machine readable storage medium as defined in claim 10 , wherein the machine readable instructions, when executed, cause the machine to at least determine the signature node number for the first site signature based on (1) a number associated with the first site signature and (2) a total number of compute nodes.

14. A method, comprising:

determining, by executing an instruction with a processor, a signature node number for a first site signature of a plurality of site signatures representative of monitored media;

determining, by executing an instruction with the processor, that a reference signature corresponding to the first site signature is stored at a first compute node based on a comparison of the signature node number with a first node number of the first compute node; and

in response to the determination that the reference signature is stored at the first compute node, comparing, by executing an instruction with the processor, a neighborhood of site signatures including the first site signature to reference signatures in a first subset of reference signatures to identify the monitored media, and determining matching information, the matching information including media identification information that identifies reference media corresponding to a matching reference signature, the first subset of reference signatures stored in a first memory partition of a computer system associated with the first compute node.

15. The method as defined in claim 14 , wherein the reference signature is a first reference signature, and further including:

determining that a second reference signature corresponding to the first site signature is stored in a second compute node; and

in response to the determination that the second reference signature is stored in the second compute node, comparing the neighborhood of site signatures including the first site signature to reference signatures in a second subset of reference signatures different than the first subset of reference signatures, the second subset of reference signatures stored in a second memory partition of the computer system, different than the first memory partition.

16. The method as defined in claim 14 , further including:

querying a nested distributed map (NDMap) to compare the first site signature to reference signatures in the first subset of reference signatures, the querying of the NDMap using the first site signature as a key to retrieve an inner map, the inner map including locations of references signatures corresponding to the first site signature in a linear list of the first subset of reference signatures; and

comparing the first site signature to the reference signatures corresponding to the first site signature in the linear list.

17. The method as defined in claim 16 , further including:

sending the first site signature and the inner map to a first graphic processing unit (GPU), the first GPU to compare the first site signature to the reference signatures corresponding to the first site signature in the linear list.

18. The method as defined in claim 16 , wherein the NDMap is a reverse index map.

19. The method as defined in claim 14 , wherein the matching information further includes a matching score.

20. The method as defined in claim 14 , further including determining the signature node number for the first site signature based on (1) a number associated with the first site signature and (2) a total number of compute nodes.