IP Library Granted Patent US 10,860,498
Granted Patent B2
US 10,860,498 · App. 16/288,015 · Granted Dec 8, 2020

Data processing system

Inventor: Seung Gyu Jeong (Gwangmyeong, KR)
Assignee: SK hynix Inc.
G06F12/123G06F12/0804G06F12/0871G06F12/126
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,860,498
App. No.
16/288,015
Granted
Dec 8, 2020
Kind
B2
Abstract

A data processing system is disclosed, which relates to a technology for implementing a convergence memory system provided with a plurality of memories. The data processing system includes a compute blade configured to generate a write command to store data and a read command to read the data, and a memory blade configured to selectively performed read and write operations in response to the read and write commands in a plurality of memories. The compute blade has a memory that stores information about performance characteristics of each of the plurality of memories, and is configured to determine priority information through which eviction of a cache line is carried out based on the stored information.

Claims (43)

1. A data processing system comprising:

a compute blade configured to generate a write command to store data and a read command to read the data; and

a memory blade configured to selectively perform read and write operations in response to the read and write commands in a plurality of memories,

wherein the compute blade has a cache memory that stores information about performance characteristics of each of the plurality of memories, and is configured to determine priority information of a memory to which an eviction operation of a cache line is performed based on the performance characteristics,

wherein the cache memory includes:

a way block configured to store cached data for a central processing unit (CPU);

a data processor configured to process the cached data to perform the eviction operation of the cache line;

a characteristic storage circuit configured to store the performance characteristics; and

an eviction circuit configured to determine the priority information of the memory based on the performance characteristics, and to perform the eviction operation on the way block corresponding to the memory according to the determined priority information,

wherein the system determines the priority information of the memory after the data processor processes the cached data that is to perform the eviction operation, and

wherein, when memories having a slow data access speed are respectively determined to be a first-priority memory and a second-priority memory, the eviction circuit allocates a second priority to a memory having a fast data access speed such that the memory having the fast data access speed is changed to the second-priority memory.

2. The data processing system according to claim 1 , further comprising:

an interconnect blade configured to communicate data between the compute blade and the memory blade.

3. The data processing system according to claim 1 , wherein the memory blade includes:

the plurality of memories in which the read operation and the write operation are performed; and

a controller configured to selectively access the plurality of memories in response to a request received from the compute blade.

4. The data processing system according to claim 3 , wherein the controller includes:

a data controller configured to control data that is communicated between the compute blade and the plurality of memories;

a plurality of memory controllers configured to control communication of the data between the data controller and the plurality of memories; and

an input/output (I/O) interface that is an interface between the data controller and the compute blade.

5. The data processing system according to claim 4 , wherein:

in response to the write command, the data controller receives data to be written from the compute blade and writes the received data in a first memory from among the plurality of memories; and

in response to the read command, the data controller reads data stored in a second memory from among the plurality of memories and outputs the read data to the compute blade.

6. The data processing system according to claim 1 , wherein the plurality of memories includes:

a first-group memory in a first temperature environment;

a second-group memory in a second temperature environment; and

a third-group memory in a third temperature environment,

wherein the second temperature is lower than the first temperature and the third temperature is lower than the second temperature.

7. The data processing system according to claim 6 , wherein the first-group memory, the second-group memory, and the third-group memory are the same kind of memories.

8. The data processing system according to claim 6 , wherein the plurality of memories have different performance characteristics.

9. The data processing system according to claim 6 , wherein:

the first-group memory has a first access speed;

the second-group memory has a second access speed faster than the first access speed; and

the third-group has a third access speed faster than the second access speed.

10. The data processing system according to claim 1 , wherein the compute blade includes:

the central processing unit (CPU) configured to generate a command to control the memory blade;

the cache memory configured to store the cached data for the central processing unit (CPU) in a cache line; and

an input/output (I/O) interface between the central processing unit (CPU) and the memory blade.

11. The data processing system according to claim 1 , wherein the cache memory further includes:

an address decoder configured to decode an address received from the central processing unit (CPU).

12. The data processing system according to claim 1 , wherein the data processor replaces the cached data using a Least Recently Used (LRU) operation.

13. The data processing system according to claim 1 , wherein the data processor replaces the cached data using a First-In First-Out (FIFO) operation.

14. The data processing system according to claim 1 , wherein the performance characteristics include any one of a data access speed, a refresh period, and an on-chip leakage amount.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 28, 2019
From: JEONG, SEUNG-GYU
To: SK HYNIX INC.
Reel/Frame 048466/0082 →
Priority Claims (1)
KR 10-2018-0144288 · Nov 21, 2018 · national
Continuity (1)
Related Publication 20200159671A1 · May 21, 2020
Cited By (2)
US 12,346,260 US 12,717,579