Memory component with input/output data rate alignment
First data is read out of a core storage array of a memory component over a first time interval constrained by data output bandwidth of the core storage array. After read out from the core storage array, the first data is output from the memory component over a second time interval that is shorter than the first time interval and that corresponds to a data transfer bandwidth greater than the data output bandwidth of the core storage array.
1. A stacked-die memory component to be coupled to a processing component, the stacked-die memory component comprising:
a plurality of memory dies disposed in a stack; and
a base die having (i) a first interface coupled to receive, from any one of the memory dies, successive sets of N data bits over successive memory-die readout intervals, respectively, N being a nonzero integer, and (ii) a second interface to output the successive sets of N data bits to the processing component over successive data-output intervals, respectively, each one of the data-output intervals being briefer than any one of the memory-die readout intervals.
2. The stacked-die memory component of claim 1 wherein the second interface to output the successive sets of N data bits to the processing component comprises an M-bit wide synchronous signaling interface to output the successive sets of N data bits at a first data rate, M being a nonzero integer less than N.
3. The stacked-die memory component of claim 2 wherein the one of the memory dies comprises an N-bit wide data interface to output, to the first interface of the base die, the successive sets of N data bits at a second data rate that is lower than the first data rate.
4. The stacked-die memory component of claim 3 wherein a multiplication product of M and the first data rate is greater than a multiplication product of N and the second data rate.
5. The stacked-die memory component of claim 1 wherein the base die comprises a command/address interfaces to receive command/address values transmitted by the processing component.
6. The stacked-die memory component of claim 1 wherein the first interface comprises a first data interface coupled to respective counterpart data interfaces within the memory dies by through-silicon vias.
7. The stacked-die memory component of claim 1 wherein each of the memory dies comprises a plurality of separately-addressable dynamic random access memory (DRAM) storage banks.
8. The stacked-die memory component of claim 1 wherein the base die comprises a first command/address interface to receive command/address values transmitted by the processing component and a second command/address interface to transmit the command/address values to the plurality of memory dies.
9. The stacked-die memory component of claim 1 wherein the base die comprises a first command/address interface to receive a column read command from the processing component and a second command/address interface to output the column read command to the one of the memory dies, and wherein the one of the memory dies comprises circuitry to transmit, as the successive sets of N data bits, a column of data to the first interface of the base die in response to the column read command.
10. The stacked-die memory component of claim 1 wherein the successive sets of N data bits received via the first interface of the base die comprise successive sets of N read data bits, and wherein the first interface of the base die comprises circuitry to transmit successive sets of N write-data bits to the one of the memory dies for storage within a core storage array of the one of the memory dies.
11. A method of operation within a stacked-die memory component that is coupled to a processing component, the method comprising:
receiving, via a first interface of a base die within the first stacked-die memory component over successive memory-die readout intervals, successive sets of N data bits, respectively, from one of a plurality of memory dies disposed in a stack within the stacked-die memory component, N being a nonzero integer; and
outputting, via a second interface of the base die, the successive sets of N data bits to the processing component over successive data-output intervals, respectively, each one of the data-output intervals being briefer than any one of the memory-die readout intervals.
12. The method of claim 11 wherein outputting the successive sets of N data bits to the processing component via the second interface comprises outputting the successive sets of N data bits at a first data rate via an M-bit wide synchronous signaling interface, M being a nonzero integer less than N.
13. The method of claim 12 wherein receiving the successive sets of N data bits from the one of the memory dies via the first interface comprises receiving the successive sets of N data bits via an N-bit wide data interface at a second data rate that is lower than the first data rate.
14. The method of claim 13 wherein a multiplication product of M and the first data rate is greater than a multiplication product of N and the second data rate.
15. The method of claim 13 wherein receiving the successive sets of N data bits via an N-bit wide data interface at the second data rate comprises receiving data bits conducted to the N-bit wide data interface by through-silicon vias coupled between the base die and the plurality of memory dies.
16. The method of claim 11 wherein each of the memory dies comprises a plurality of separately-addressable dynamic random access memory (DRAM) storage banks.
17. The method of claim 11 further comprising:
receiving, via a first command/address interface of the base die, command/address values transmitted by the processing component; and
transmitting the command/address values to the plurality of memory dies via a second command/address interface of the base die.
18. The method of claim 11 further comprising:
receiving a column read command from the processing component via a first command/address interface of the base die;
outputting the column read command to the one of the memory dies via a second command/address interface base die; and
transmitting, as the successive sets of N data bits, a column of data from the one of the memory dies to the first interface of the base die in response to the column read command.
19. The method of claim 11 wherein receiving the successive sets of N data bits via the first interface of the base die comprises receiving successive sets of N read data bits, the method further comprising transmitting, via the first interface of the base die successive sets of N write-data bits to the one of the memory dies for storage within a core storage array of the one of the memory dies.
20. A stacked-die memory component to be coupled to a processing component, the stacked-die memory component comprising:
a plurality of memory dies disposed in a stack; and
a base die having:
means for receiving, from any one of the memory dies, successive sets of N data bits over successive memory-die readout intervals, respectively, N being a nonzero integer; and
means for outputting the successive sets of N data bits to the processing component over successive data-output intervals, respectively, each one of the data-output intervals being briefer than any one of the memory-die readout intervals.