Method of improving access to a main memory of a computer system, a corresponding computer system and a computer program product
View Patent ↗A method of improving access to a main memory of a computer system includes providing a main memory including at least a first memory of a first memory configuration and a second memory of a second memory configuration, and providing a choice between the first memory and the second memory for access to the main memory, wherein access to the first memory and access to the second memory can be distinguished by the choice depending on the respective access behavior of the first memory and of the second memory, and the choice is provided by a running application program.
1. A method of improving access to a main memory of a computer system comprising:
providing a main memory comprising at least a first memory of a first memory configuration and a second memory of a second memory configuration, and
providing a choice between the first memory and the second memory for access to the main memory,
wherein access to the first memory and access to the second memory can be distinguished by the choice depending on the respective access behavior of the first memory and of the second memory,
the choice is implemented in a running application program, in a user layer of a hierarchical computer system architecture, the user layer being different from an operating system layer of the hierarchical computer system architecture, and
the application program by the choice decides in advance which of the first memory or the second memory of the main memory it intends to access.
2. The method according to claim 1 , wherein at least two program interfaces are provided in the application program and the first memory is addressed via the first program interface and the second memory is addressed via the second program interface.
3. The method according to claim 1 , wherein program data of the application program running in the computer system are stored either in the first memory or in the second memory by the choice depending on the respective access behavior of the first memory and the second memory or depending on the access frequency of the respective program data.
4. The method according to claim 1 , wherein the respective access behavior of the first memory and the second memory is determined at least by the following properties:
memory access time,
memory cycle time,
latency time, and
data throughput.
5. The method according to claim 1 , wherein the first memory is provided as a volatile memory and the second memory is provided as a volatile or non-volatile memory.
6. The method according to claim 1 , wherein the second memory is provided as a remote memory via network by one or multiple remote computer systems.
7. The method according to claim 1 , wherein access to the main memory is effected byte by byte or word by word.
8. A computer system comprising: a main memory comprising at least a first memory of a first memory configuration and a second memory of a second memory configuration,
wherein a choice between the first memory and the second memory for access to the main memory is set up such that access to the first memory and access to the second memory can be distinguished by the choice depending on the respective access behavior of the first memory and the second memory,
the choice is implemented in an application program in a user layer of a hierarchical computer system architecture, the user layer being different from an operating system layer of the hierarchical computer system architecture, and
the application program is configured to decide in advance by the choice which of the first memory or the second memory of the main memory it intends to access.
9. The computer system according to claim 8 , wherein at least two program interfaces are implemented in the application program for access to the main memory and the first memory is addressable via the first program interface and the second memory is addressable via the second program interface.
10. The computer system according to claim 8 , configured to perform a method comprising:
providing a main memory comprising at least a first memory of a first memory configuration and a second memory of a second memory configuration, and
providing a choice between the first memory and the second memory for access to the main memory,
wherein access to the first memory and access to the second memory can be distinguished by the choice depending on the respective access behavior of the first memory and of the second memory, and
the choice is provided by a running application program.
11. A computer program product configured to be executed on a computer system and performs a method according to claim 1 when being executed.
12. The computer system according to claim 9 , configured to perform a method comprising:
providing a main memory comprising at least a first memory of a first memory configuration and a second memory of a second memory configuration, and
providing a choice between the first memory and the second memory for access to the main memory,
wherein access to the first memory and access to the second memory can be distinguished by the choice depending on the respective access behavior of the first memory and of the second memory, and
the choice is provided by a running application program.
13. The method according to claim 2 , wherein program data of the application program running in the computer system are stored either in the first memory or in the second memory by the choice depending on the respective access behavior of the first memory and the second memory or depending on the access frequency of the respective program data.
14. The method according to claim 2 , wherein the respective access behavior of the first memory and the second memory is determined at least by the following properties:
memory access time,
memory cycle time,
latency time, and
data throughput.
15. The method according to claim 3 , wherein the respective access behavior of the first memory and the second memory is determined at least by the following properties:
memory access time,
memory cycle time,
latency time, and
data throughput.
16. The method according to claim 2 , wherein the first memory is provided as a volatile memory and the second memory is provided as a volatile or non-volatile memory.
17. The method according to claim 3 , wherein the first memory is provided as a volatile memory and the second memory is provided as a volatile or non-volatile memory.
18. The method according to claim 4 , wherein the first memory is provided as a volatile memory and the second memory is provided as a volatile or non-volatile memory.
19. The method according to claim 2 , wherein the second memory is provided as a remote memory via network by one or multiple remote computer systems.
20. The method according to claim 3 , wherein the second memory is provided as a remote memory via network by one or multiple remote computer systems.