Optimizing memory usage in physical memory
A computer-implemented method for optimizing memory usage. A processor set classifies data stored in virtual memory into hot data and cold data based on frequency of access for the data stored in the virtual memory. The processor set segregates the hot data and the cold data into different memory regions in the virtual memory to generate hot memory regions comprising the hot data and cold memory regions comprising the cold data. The processor set maps the hot memory regions to large physical pages in memory pages of physical memory and the cold memory regions to small physical pages in the memory pages. The processor set disclaims data in the cold memory regions from the memory pages of physical memory to reduce memory space used in the physical memory.
1 . A computer implemented method for optimizing memory usage, the computer implemented method comprising:
classifying, by a processor set, data stored in virtual memory into hot data and cold data based on frequency of access for the data stored in the virtual memory, wherein metadata from the data are classified into the cold data when compiled codes associated with the metadata are already classified into the cold data, and wherein the compiled codes associated with the metadata comprise unresolved references associated with codes that did not run before the compilation;
segregating, by the processor set, the hot data and the cold data into different memory regions in the virtual memory to generate hot memory regions comprising the hot data and cold memory regions comprising the cold data;
mapping, by the processor set, the hot memory regions to large physical pages in memory pages of physical memory and the cold memory regions to small physical pages in the memory pages of the physical memory; and
disclaiming, by the processor set, data in the cold memory regions from the memory pages of the physical memory to reduce memory space used in the physical memory.
2 . The computer implemented method of claim 1 , wherein classifying, by the processor set, the data stored in the virtual memory into the hot data and the cold data based on the frequency of access for the data stored in the virtual memory comprises:
estimating, by the processor set, the frequency of access for the data stored in the virtual memory;
determining, by the processor set, whether the frequency of access for the data stored in the virtual memory exceeds a predefined threshold;
in response to determining that the frequency of access for the data stored in the virtual memory exceeds the predefined threshold, classifying, by the processor set, the data stored in the virtual memory into the hot data; and
in response to determining that the frequency of access for the data stored in the virtual memory does not exceed the predefined threshold, classifying, by the processor set, the data stored in the virtual memory into the cold data.
3 . The computer implemented method of claim 1 , wherein disclaiming, by the processor set, the data in the cold memory regions from the memory pages of physical memory to reduce the memory space used in the physical memory comprises:
disclaiming, by the processor set, the memory pages backing the data in the cold memory regions after an application accessing data in the cold memory regions experiences a phase change.
4 . The computer implemented method of claim 1 , wherein disclaiming, by the processor set, the data in the cold memory regions from the memory pages of physical memory to reduce the memory space used in the physical memory comprises:
disclaiming, by the processor set, the memory pages backing data in a cold memory region after the cold memory region becomes full.
5 . The computer implemented method of claim 1 , wherein disclaiming, by the processor set, the data in the cold memory regions from the memory pages of physical memory to reduce the memory space used in the physical memory comprises:
disclaiming, by the processor set, the memory pages backing the data in the cold memory regions periodically in a predetermined time interval.
6 . The computer implemented method of claim 1 , wherein the compiled codes are associated with slow paths for speculative optimization generated by a compiler.
7 . The computer implemented method of claim 1 , wherein basic blocks of the compiled codes are dominated or post dominated by other basic blocks that are classified into the cold data.
8 . A computer system comprising:
a processor set;
a set of one or more computer-readable storage media; and
program instructions, collectively stored in the set of one or more storage media, for causing the processor set to perform the following computer operations:
classify data stored in virtual memory into hot data and cold data based on frequency of access for the data stored in the virtual memory, wherein metadata from the data are classified into the cold data when compiled codes associated with the metadata are already classified into the cold data, and wherein the compiled codes associated with the metadata comprise unresolved references associated with codes that did not run before the compilation;
segregate the hot data and the cold data into different memory regions in the virtual memory to generate hot memory regions comprising the hot data and cold memory regions comprising the cold data;
map the hot memory regions to large physical pages in memory pages of physical memory and the cold memory regions to small physical pages in the memory pages of the physical memory; and
disclaim data in the cold memory regions from the memory pages of the physical memory to reduce memory space used in the physical memory.
9 . The computer system of claim 8 , wherein as part of classifying the data stored in the virtual memory into the hot data and the cold data based on the frequency of access for the data stored in the virtual memory, the program instructions, collectively stored in the set of one or more storage media, cause the processor set to perform the following computer operations:
estimate the frequency of access for the data stored in the virtual memory;
determine whether the frequency of access for the data stored in the virtual memory exceeds a predefined threshold;
in response to determining that the frequency of access for the data stored in the virtual memory exceeds the predefined threshold, classify the data stored in the virtual memory into the hot data; and
in response to determining that the frequency of access for the data stored in the virtual memory does not exceed the predefined threshold, classify the data stored in the virtual memory into the cold data.
10 . The computer system of claim 8 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the program instructions, collectively stored in the set of one or more storage media, cause the processor set to perform the following computer operations:
disclaim the memory pages backing the data in the cold memory regions after an application accessing the data in the cold memory regions experiences a phase change.
11 . The computer system of claim 8 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the program instructions, collectively stored in the set of one or more storage media, cause the processor set to perform the following computer operations:
disclaim the memory pages backing data in a cold memory region after the cold memory region becomes full.
12 . The computer system of claim 8 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the program instructions, collectively stored in the set of one or more storage media, cause the processor set to perform the following computer operations:
disclaim the memory pages backing the data in the cold memory regions periodically in a predetermined time interval.
13 . The computer system of claim 8 , wherein the compiled codes are associated with slow paths for speculative optimization generated by a compiler.
14 . The computer system of claim 8 , wherein basic blocks of the compiled codes are dominated or post dominated by other basic blocks that are classified into the cold data.
15 . A computer program product for optimizing memory usage, the computer program product comprising:
a set of one or more computer-readable storage media;
program instructions, collectively stored in the set of one or more storage media, cause a processor set to perform the following computer operations:
classify data stored in virtual memory into hot data and cold data based on frequency of access for the data stored in the virtual memory, wherein metadata from the data are classified into the cold data when compiled codes associated with the metadata are already classified into the cold data, and wherein the compiled codes associated with the metadata comprise unresolved references associated with codes that did not run before the compilation;
segregate the hot data and the cold data into different memory regions in the virtual memory to generate hot memory regions comprising the hot data and cold memory regions comprising the cold data;
map the hot memory regions to large physical pages in memory pages of physical memory and the cold memory regions to small physical pages in the memory pages of the physical memory; and
disclaim data in the cold memory regions from the memory pages of the physical memory to reduce memory space used in the physical memory.
16 . The computer program product of claim 15 , wherein as part of classifying the data stored in the virtual memory into the hot data and the cold data based on the frequency of access for the data stored in the virtual memory, the operation performed by the processor set comprises:
estimate the frequency of access for the data stored in the virtual memory;
determine whether the frequency of access for the data stored in the virtual memory exceeds a predefined threshold;
in response to determining that the frequency of access for the data stored in the virtual memory exceeds the predefined threshold, classify the data stored in the virtual memory into the hot data; and
in response to determining that the frequency of access for the data stored in the virtual memory does not exceed the predefined threshold, classify the data stored in the virtual memory into the cold data.
17 . The computer program product of claim 15 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the operation performed by the processor set comprises:
disclaim the memory pages containing the data in the cold memory regions after an application accessing the data in the cold memory regions experiences a phase change.
18 . The computer program product of claim 15 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the operation performed by the processor set comprises:
disclaim the memory pages backing data in a cold memory region after the cold memory region becomes full.
19 . The computer program product of claim 15 , wherein as part of disclaiming the data in the cold memory regions from the memory pages of the physical memory to reduce the memory space used in the physical memory, the operation performed by the processor set comprises:
disclaim the memory pages backing the data in the cold memory regions periodically in a predetermined time interval.
20 . The computer program product of claim 15 , wherein the compiled codes are associated with slow paths for speculative optimization generated by a compiler.