Memory management method based on compressed memory and apparatus using the same
A memory management method based on a compressed memory and an apparatus using the same are provided. A memory device includes a memory and a near memory processing unit. The memory includes a normal memory area for storing uncompressed data and a compressed memory area for storing compressed data. The near memory processing unit is configured to receive a first command to frontswap-store a huge page, which is stored in the normal memory area, in the compressed memory area; identify addresses of sub-pages of the huge page based on an address of the huge page; compress the sub-pages using the addresses of the sub-pages to generate compressed sub-pages; and store the compressed sub-pages in the compressed memory area. A size of the huge page is larger than a regular page.
1. A memory device comprising:
a memory comprising a normal memory area for storing uncompressed data and a compressed memory area for storing compressed data; and
a near memory processing unit configured to:
receive a first command to frontswap-store a huge page, which is stored in the normal memory area, in the compressed memory area, wherein a size of the huge page is larger than a regular page,
identify addresses of sub-pages of the huge page based on an address of the huge page,
compress the sub-pages using the addresses of the sub-pages to generate compressed sub-pages, and
store the compressed sub-pages in the compressed memory area.
2. The memory device of claim 1 , wherein the near memory processing unit is configured to identify the addresses of the sub-pages based on whether the addresses of the sub-pages are consecutive.
3. The memory device of claim 1 , wherein the near memory processing unit is configured to, when the sub-pages are stored in a consecutive address space, determine an address of a following sub-page of the sub-pages by adding a sub-page size to an address of a current sub-page of the sub-pages.
4. The memory device of claim 1 , wherein the near memory processing unit is configured to:
when the sub-pages are stored in a non-consecutive address space, obtain a page frame number of a following sub-page of the sub-pages by performing a page table walk, and
determine an address of the following sub-page by adding a page offset of the following sub-page to the page frame number of the following sub-page.
5. The memory device of claim 1 , wherein the near memory processing unit is configured to:
receive a second command to request a target sub-page of the huge page after the huge page is frontswap-stored in the compressed memory area,
determine a target frame number based on a difference between the address of the huge page and an address of the target sub-page,
determine an address of a compressed target sub-page in the compressed memory area based on the target frame number,
decompress the compressed target sub-page using the address of the compressed target sub-page to generate a decompressed target sub-page, and
store the decompressed target sub-page in the normal memory area.
6. The memory device of claim 5 , wherein decompression of the compressed target sub-page is performed prior to decompression of remaining compressed sub-pages of the huge page.
7. The memory device of claim 1 , wherein the memory comprises a first memory channel and a second memory channel.
8. The memory device of claim 7 , wherein the sub-pages of the huge page are distributed across the first memory channel and the second memory channel, and
the near memory processing unit is configured to compress the sub-pages by parallel processing the sub-pages distributed across the first memory channel and the second memory channel.
9. The memory device of claim 7 , wherein the near memory processing unit comprises a first processing unit disposed to communicate with the first memory channel and a second processing unit disposed to communicate with the second memory channel.
10. The memory device of claim 9 , wherein, when a sub-page of the sub-pages is comprised in the first memory channel, the sub-page is compressed by the first processing unit, and
when a sub-page of the sub-pages is comprised in the second memory channel, the sub-page is compressed by the second processing unit.
11. The memory device of claim 9 , wherein the first processing unit is configured to compress a portion of the sub-pages stored in the first memory channel,
the second processing unit is configured to compress the other portion of the sub-pages stored in the second memory channel, and
a compression operation of the portion of the sub-pages by the first processing unit is performed in parallel with a compression operation of the other portion of the sub-pages by the second processing unit.
12. A memory device comprising:
a memory comprising a first memory channel comprising a first normal memory area and a first compressed memory area and a second memory channel comprising a second normal memory area and a second compressed memory area;
a first processing unit disposed to communicate with the first memory channel; and
a second processing unit disposed to communicate with the second memory channel,
wherein sub-pages of a huge page having a size larger than a regular page are distributed across the first normal memory area and the second normal memory area, and
in response to a first command to frontswap-store the huge page in the compressed memory area, the first processing unit and the second processing unit are configured to compress the sub-pages by parallel processing the sub-pages.
13. The memory device of claim 12 , wherein, based on addresses of the sub-pages, when a sub-page of the sub-pages is comprised in the first memory channel, the sub-page is compressed by the first processing unit, and
when a sub-page of the sub-pages is comprised in the second memory channel, the sub-page is compressed by the second processing unit.
14. The memory device of claim 12 , wherein the first processing unit is configured to compress a portion of the sub-pages stored in the first memory channel,
the second processing unit is configured to compress the other portion of the sub-pages stored in the second memory channel, and
a compression operation of the portion of the sub-pages by the first processing unit is performed in parallel with a compression operation of the other portion of the sub-pages by the second processing unit.
15. A memory management method of a memory device, the memory management method comprising:
receiving a first command to frontswap-store a huge page, which is stored in a normal memory area of a memory, in a compressed memory area of the memory, wherein a size of the huge page is larger than a regular page;
identifying addresses of sub-pages of the huge page based on an address of the huge page;
compressing the sub-pages using the addresses of the sub-pages to generate compressed sub-pages; and
storing the compressed sub-pages in the compressed memory area.
16. The memory management method of claim 15 , wherein the identifying of the addresses of the sub-pages comprises, when the sub-pages are stored in a consecutive address space, determining an address of a following sub-page of the sub-pages by adding a sub-page size to an address of a current sub-page of the sub-pages.
17. The memory management method of claim 15 , wherein the identifying of the addresses of the sub-pages comprises:
when the sub-pages are stored in a non-consecutive address space, obtaining a page frame number of a following sub-page of the sub-pages by performing a page table walk; and
determining an address of the following sub-page by adding a page offset of the following sub-page to the page frame number of the following sub-page.
18. The memory management method of claim 15 , further comprising:
receiving a second command to request a target sub-page of the huge page after the huge page is frontswap-stored in the compressed memory area;
determining a target frame number based on a difference between the address of the huge page and an address of the target sub-page;
determining an address of a compressed target sub-page in the compressed memory area based on the target frame number;
decompressing the compressed target sub-page using the address of the compressed target sub-page to generate a decompressed target sub-page; and
storing the decompressed target sub-page in the normal memory area.
19. The memory management method of claim 15 , wherein the memory comprises a first memory channel and a second memory channel,
the sub-pages of the huge page is distributed across the first memory channel and the second memory channel, and
the compressing of the sub-pages comprises compressing the sub-pages by parallel processing the sub-pages distributed across the first memory channel and the second memory channel.
20. The memory management method of claim 19 , further comprising:
identifying memory channels comprising the sub-pages based on the addresses of the sub-pages;
when a sub-page of the sub-pages is comprised in the first memory channel, compressing the sub-page using a first processing unit disposed to communicate with the first memory channel; and
when the sub-page is comprised in the second memory channel, compressing the sub-page using a second processing unit disposed to communicate with the second memory channel.