Memory controller and memory system including the same

1. A memory system comprising:

a memory device including a plurality of memory banks; and

a memory controller suitable for monitoring a workload of the memory device, and applying one of a per-bank refresh command and an all-bank refresh command to the memory device according to a result of the monitoring,

wherein only one bank of the memory device is refreshed when the per-bank refresh command is applied to the memory device and all banks of the memory device are refreshed when the all-bank refresh command is applied to the memory device,

wherein the memory controller applies the per-bank refresh command to the memory device when the workload is heavy, whereas the memory controller applies the all-bank refresh command to the memory device when the workload is light, and

wherein the memory controller comprises:

a workload monitoring unit suitable for monitoring the workload of the memory device;

a scheduler suitable for determining an operation order of the memory device based on a provision order of requests provided from one or more hosts, wherein the operation order is different from the provision order of requests provided from the one or more hosts so that when a first-requesting-host provides a read request to the memory device before a second-requesting-host provides a write request to the memory device, the scheduler determines the operation order to be that the write request from the second-requesting-host is performed prior to the read request from the first-requesting-host; and

a command generator suitable for generating a command to be applied to the memory device according to the determined operation order.

2. The memory system according to claim 1 , wherein the workload monitoring unit counts a number of the requests, and determines the workload of the memory device using a result of the counting.

3. The memory system according to claim 2 , wherein, if a total number of the requests for a predetermined time is equal to or greater than a first threshold value, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

4. The memory system according to claim 2 ,

wherein the one or more hosts are classified into first hosts providing performance-sensitive requests, and second hosts providing requests other than the performance-sensitive requests, and

wherein, if a ratio of a number of requests from the first hosts over a number of requests from the second hosts for a predetermined time is equal to or greater than a second threshold value, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

5. The memory system according to claim 4 ,

wherein the first hosts include at least one or more of a central processing unit (CPU) and a graphic processing unit (GPU), and

wherein the second hosts include one or more of a communication unit, a display controller and a video codec.

6. The memory system according to claim 2 ,

wherein the one or more hosts are classified into first hosts providing performance-sensitive requests, and second hosts providing requests other than the performance-sensitive requests, and

wherein, if a number of requests from the first hosts for a predetermined time is a third threshold value or more, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

7. The memory system according to claim 1 , wherein the first-requesting-host is a central processing unit (CPU) and the second-requesting-host is a graphic processing unit (GPU).

8. A memory controller comprising:

a workload monitoring unit suitable for monitoring a workload of a memory device;

a scheduler suitable for determining an operation order of the memory device based on a provision order of requests provided from one or more hosts, wherein the operation order is different from the provision order of requests provided from the one or more hosts so that when a first-requesting-host provides a read request to the memory device before a second-requesting-host provides a write request to the memory device, the scheduler determines the operation order to be that the write request from the second-requesting-host is performed prior to the read request from the first-requesting-host; and

a command generator suitable for generating a command to be applied to the memory device according to the determined operation order,

wherein the command generator generates one of a per-bank refresh command and an all-bank refresh command depending on the workload monitored by the workload monitoring unit,

wherein only one bank of the memory device is refreshed when the per-bank refresh command is applied to the memory device and all banks of the memory device are refreshed when the all-bank refresh command is applied to the memory device,

wherein, in the case where the workload is heavy, the command generator generates the per-bank refresh command, and

wherein, in the case where the workload is light, the command generator generates the all-bank refresh command.

9. The memory controller according to claim 8 , wherein the workload monitoring unit counts a number of the requests, and determines the workload of the memory device using a result of the counting.

10. The memory controller according to claim 9 , wherein, if a total number of the requests for a predetermined time is equal to or greater than a first threshold value, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

11. The memory controller according to claim 9 ,

wherein the one or more hosts are classified into first hosts providing performance-sensitive requests, and second hosts providing requests other than the performance-sensitive requests, and

wherein, if a ratio of a number of requests from the first hosts over a number of requests from the second hosts for a predetermined time is equal to or greater than a second threshold value, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

12. The memory controller according to claim 11 ,

wherein the first hosts include at least one or more of a central processing unit (CPU) and a graphic processing unit (GPU), and

wherein the second hosts include one or more of a communication unit, a display controller and a video codec.

13. The memory controller according to claim 9 ,

wherein the one or more hosts are classified into first hosts providing performance-sensitive requests, and second hosts providing requests other than the performance-sensitive requests, and

wherein, if a number of requests from the first hosts for a predetermined time is a third threshold value or more, the workload monitoring unit determines the workload as being heavy, and, otherwise, the workload monitoring unit determines the workload as being light.

14. The memory controller according to claim 8 , wherein the first-requesting-host is a central processing unit (CPU) and the second-requesting-host is a graphic processing unit (GPU).