| CPC G06F 13/1642 (2013.01) [G06F 13/1673 (2013.01)] | 20 Claims |

|
1. A request processing method, applied to a current accelerator card, comprising:
acquiring, from a local first input first output (FIFO) queue, an address of a current request to be processed, and based on the address of the current request to be processed, acquiring the current request to be processed from a preset request buffer area;
parsing the current request to be processed, and based on a parsing result, acquiring corresponding current source data to be processed and a serial address item for recording an address of a next request to be processed;
processing the current source data to be processed to obtain currently processed target data, writing the currently processed target data into a preset data buffer area, and determining whether the serial address item is empty;
notifying a host to read all processed target data from the preset data buffer area in response to the serial address item being empty, and sending an address recorded in the serial address item to a FIFO queue of a next accelerator card for saving in response to the serial address item being non-empty;
wherein the parsing the current request to be processed, and based on the parsing result, acquiring the corresponding current source data to be processed and the serial address item for recording the address of the next request to be processed comprises:
parsing the current request to be processed to obtain a corresponding parsing result, wherein the parsing result comprises an address of processed target data corresponding to a previous processed request, an address of currently processed target data, and a serial address item for recording an address of a next request to be processed; acquiring corresponding data from the preset data buffer area based on the address of the processed target data corresponding to the previous processed request to obtain the current source data to be processed; and
wherein the writing the currently processed target data into the preset data buffer area comprises:
writing the currently processed target data into a corresponding position of the preset data buffer area based on an address of the currently processed target data carried in the parsing result.
|