Memory controller and request scheduling method using the same
AssigneeSK hynix Inc.
InventorsYoung-Suk MOON, Hong-Sik KIM
Life of this patent
Estimating the term…
Abstract
A memory controller includes a plurality of request queues suitable for storing requests transmitted from corresponding host devices among a plurality of host devices; a token information generation unit suitable for generating informations on the numbers of first and second tokens corresponding to the plurality of respective host devices; and a request scheduler suitable for selecting repeatedly and sequentially the plurality of request queues, and outputting requests stored in a selected request queue, by using first and second tokens, wherein the request scheduler outputs one request per one first token and, when first tokens are consumed all, outputs one request per one second token.
Claims20
A memory controller comprising:
- a plurality of request queues, each request queue being suitable for storing requests received from a corresponding host device among a plurality of host devices;
- a token information generation unit suitable for generating information on the numbers of at least one first and second tokens corresponding to each of the plurality of host devices;
- and
- a request scheduler suitable for selecting repeatedly and sequentially one among the plurality of request queues, and outputting requests stored in a selected request queue, by using first and second tokens, wherein the request scheduler outputs one request per one first token and, when all the first tokens are consumed, outputs one request per one second token.
A request scheduling method of a memory controller for outputting requests receiving from a plurality of host devices, to a memory device, comprising:
- allocating at least one first token and at least one second token to each of the plurality of host devices;
- storing requests received from the plurality of host devices, in a plurality of request queues respectively corresponding to the plurality of host devices;
- selecting one request queue among the plurality of request queues;
- outputting each of requests stored in the selected request queue, per one first token;
- and
- outputting, when all the first tokens are consumed, each of requests stored and remained in the selected request queue, per one second token.
Dependencies are inferred from the claim text.