Method and a system for network-on-chip arbitration

1. A computer-implemented method ( 100 ) for network-on-chip arbitration, the method ( 100 ) comprising the steps of:

receiving input from a user via a user interface; and

selecting a plurality of flits from a plurality of ingress into a plurality of virtual channels followed by selecting the flits from the virtual channels into a plurality of egress based on the input from the user;

wherein selecting the flits into the virtual channels and the egress characterized by the steps of:

computing default and elevated bandwidths of the virtual channels;

computing default and elevated weights of the virtual channels based on the default and elevated bandwidths; and

generating a weightage lookup table using the default and elevated weights to perform arbitration weightage lookup for the flits with default and elevated priority levels for selecting the flits into the virtual channels and the egress;

wherein the flits are from different ingress and comprise different default and elevated weight, and

wherein the method ( 100 ) comprises deficit interleaved weighted round-robin arbitration.

2. The computer-implemented method ( 100 ) for network-on-chip arbitration as claimed in claim 1 , wherein the input from the user comprises node data bandwidth, node bandwidth allocation and direction, node priority level allocation, priority virtual channel mapping, priority level multiplier and maximum weight.

3. The computer-implemented method ( 100 ) for network-on-chip arbitration as claimed in claimed 1 , wherein the method ( 100 ) comprises at least one priority level.

4. The computer-implemented method ( 100 ) for network-on-chip arbitration as claimed in claim 1 , wherein the weightage lookup table is used in a weight counter refresh process.

5. A computer program product comprising a non-transitory computer readable medium having a computer readable program embodied therein, wherein the computer readable program, when executed on a computing device, causes the computing device to:

receiving input from a user via a user interface;

computing default and elevated bandwidths of a plurality of virtual channels;

computing default and elevated weights of the virtual channels based on the default and elevated bandwidths; and

generating a weightage lookup table using the default and elevated weights to perform arbitration weightage lookup for a plurality of flits with default and elevated priority levels for selecting the flits into the virtual channels and a plurality of egress,

wherein the flits are from different ingress and comprise different default and elevated weight, and

wherein the computing device utilizes deficit interleaved weighted round-robin arbitration.

6. A system ( 200 ) for arbitrating network-on-chip comprising:

a processor in communication with a memory storing instructions that when executed by the processor provide:

a global order buffer module ( 202 ) for storing a plurality of flits from a plurality of ingress, tracking the flits destination in a plurality of virtual channels and a plurality of egress, and buffering the flits;

a linked-list managing module ( 204 ) for managing a linked-list tracking of the egress, wherein the flits going into the same virtual channel of the same egress are stitched in the same linked-list; and

at least one priority tracker module ( 206 ) for recording priority levels of the flits and assigning the flits from the same linked-list to exit the global order buffer module ( 202 ) into the virtual channels and the egress;

wherein the global order buffer module ( 202 ), the linked-list managing module ( 204 ) and the priority tracker module ( 206 ) are configured in a router,

wherein the flits are from different ingress and comprise different default and elevated weight, and

wherein the system ( 200 ) utilizes deficit interleaved weighted round-robin arbitration.

7. The system ( 200 ) for network-on-chip arbitration as claimed in claim 6 , wherein the system ( 200 ) further comprises a tracker module ( 208 ) configured in a node for ensuring the flits of a same transaction ordering are send through the same virtual channels regardless of the priority level.