memmapfile
Create memory map to a file
Description
maps
an existing file, m = memmapfile(filename)filename, to memory and returns
the memory map, m.
Memory-mapping is a mechanism that maps a portion of a file, or an entire file, on disk to a range of memory addresses within the MATLAB® address space. Then, MATLAB can access files on disk in the same way it accesses dynamic memory, accelerating file reading and writing. Memory-mapping allows you to work with data in a file as if it were a MATLAB array.
specifies
the properties of m = memmapfile(filename,Name,Value)m using one or more name-value
pair arguments. For example, you can specify the format of the data
in the file.
Examples
At the command prompt, create a sample file in your current
folder called records.dat, containing 10 uint8 values.
myData = uint8(1:10)'; fileID = fopen('records.dat','w'); fwrite(fileID, myData,'uint8'); fclose(fileID);
Create a map for records.dat. When
using memmapfile, the default data format is uint8 so
the file name is the only required input argument in this case.
m = memmapfile('records.dat')m =
Filename: 'd:\matlab\records.dat'
Writable: false
Offset: 0
Format: 'uint8'
Repeat: Inf
Data: 10x1 uint8 arrayMATLAB maps the entire records.dat file
to memory, setting all properties of the memory map to their default
values. The memory map is assigned to the variable, m.
In this example, the command maps the entire file as a sequence of
unsigned 8-bit integers and gives the caller read-only access to its
contents.
View the mapped data by accessing the Data property
of m.
m.Data
ans =
1
2
3
4
5
6
7
8
9
10Create a memory map for double-precision data. The syntax is similar when specifying other data types.
At the command prompt, create a sample file in your current folder called records.dat, containing 10 double values.
myData = (1:10)'; fileID = fopen('records.dat','w'); fwrite(fileID,myData,'double'); fclose(fileID);
Create a memory map for records.dat, and set the Format property for the output to 'double'.
m = memmapfile('records.dat','Format','double') ;
The memmapfile, m, contains the following properties: Filename, Writable, Offset, Format, Repeat, and Data. To display any one property, for example Format, type m.Format in the command window.
m.Format
ans = 'double'
The Data property contains the 10 double-precision values in records.dat.
Create a memory map for a large array of int32 data.
Specify write access, and nondefault Format and Offset values.
At the command prompt, create a sample file in your current
folder called records.dat, containing 10,000 int32 values.
myData = int32([1:10000]); fileID = fopen('records.dat','w'); fwrite(fileID,myData,'int32'); fclose(fileID);
Create a memory map for records.dat,
and set the Format property for the output to int32.
Also, set the Offset property to disregard the
first 9000 bytes in the file, and the Writable property
to permit write access.
m = memmapfile('records.dat',... 'Offset',9000,... 'Format','int32',... 'Writable',true);
An Offset value of 9000 indicates
that the first 9000 bytes of records.dat are not
mapped.
Type the name of the memory map to see the current settings for all properties.
m
m =
Filename: 'd:\matlab\records.dat'
Writable: true
Offset: 9000
Format: 'int32'
Repeat: Inf
Data: 7750x1 int32 arrayThe Format property indicates that any read
or write operation made via the memory map reads and writes the file
contents as a sequence of signed 32-bit integers. The Data property
contains only 7750 elements because the first 9000 bytes of records.dat,
representing the first 2250 values in the file, are not mapped.
View the first five elements of the mapped data by accessing
the Data property of m.
m.Data(1:5)
ans =
2251
2252
2253
2254
2255Create a memory map for a region of a file containing 100 double-precision values.
At the command prompt, create a sample file in your current
folder called mybinary.bin, containing 100 double-precision
values.
rng('default') randData = rand([100,1]); fileID = fopen('mybinary.bin','w'); fwrite(fileID,randData,'double'); fclose(fileID);
Map the first 75 values in mybinary.bin to
a 5-by-5-by-3 array of double-precision values that can be referenced
in the structure of the memory map using the field name x.
Specify these parameters with the Format name-value
pair argument.
m = memmapfile('mybinary.bin',... 'Format',{'double',[5 5 3],'x'})
m =
Filename: 'd:\matlab\mybinary.bin'
Writable: false
Offset: 0
Format: {'double' [5 5 3] 'x'}
Repeat: Inf
Data: 1x1 struct array with fields:
xThe Data property is a structure array that
contains the mapped values in the field, x.
Assign the mapped data to a variable, A.
Because the Data property is a structure array,
you must index into the field, x, to access the
data.
A = m.Data.x;
View information about A.
whos AName Size Bytes Class Attributes A 5x5x3 600 double
Map segments of a file with different array shapes and data types to memory.
At the command prompt, create a sample file in your current
folder called mybinary.bin. Write uint16 data
and double-precision data representing sample pressure, temperature,
and volume values into the file. In this case, each of the uint16 arrays
are 50-by-1 and the double-precision arrays are 5-by-10. k is
a sample scaling factor.
rng('default') k = 8.21; pres1 = randi([1,300],[50,1],'uint16'); temp1 = randi([1,300],[50,1],'uint16'); vol1 = double(reshape(k*temp1./pres1,5,10)); pres2 = randi([5,500],[50,1],'uint16'); temp2 = randi([5,500],[50,1],'uint16'); vol2 = double(reshape(k*temp2./pres2,5,10)); fileID = fopen('mybinary.bin','w'); fwrite(fileID,pres1,'uint16'); fwrite(fileID,temp1,'uint16'); fwrite(fileID,vol1,'double'); fwrite(fileID,pres2,'uint16'); fwrite(fileID,temp2,'uint16'); fwrite(fileID,vol2,'double'); fclose(fileID);
Map the file to arrays accessible by unique names. Define
a field, pressure, containing a 50-by-1 array of uint16 values,
followed by a field, temperature, containing 50-by-1 uint16 values.
Define a field, volume, containing a 5-by-10 array
of double-precision values. Use a cell array to define the format
of the mapped region and repeat the pattern twice.
m = memmapfile('mybinary.bin',... 'Format',{'uint16',[50 1],'pressure';... 'uint16',[50,1],'temperature';... 'double',[5,10],'volume'},'Repeat',2)
m =
Filename: 'd:\matlab\mybinary.bin'
Writable: false
Offset: 0
Format: {'uint16' [50 1] 'pressure'
'uint16' [50 1] 'temperature'
'double' [5 10] 'volume'}
Repeat: 2
Data: 2x1 struct array with fields:
pressure
temperature
volumeThe Data property of the memory map, m,
is a 2-by-1 structure array because the Format is
applied twice.
Copy the Data property to a variable, A.
Then, view the last block of double data, which
you can access using the field name, volume.
A = m.Data; myVolume = A(2).volume
myVolume =
2 13 32 5 5 16 4 22 3 8
2 9 53 38 13 19 23 85 2 120
29 10 6 1 2 5 6 58 20 11
7 15 4 1 5 18 1 4 14 8
9 8 4 2 0 9 8 6 3 3Input Arguments
Name of the file to map including the file extension, specified as a character vector or
string scalar. The filename argument cannot include any wildcard
characters (for example, * or ?).
Example: 'myFile.dat'
Data Types: char | string
Name-Value Arguments
Specify optional pairs of arguments as
Name1=Value1,...,NameN=ValueN, where Name is
the argument name and Value is the corresponding value.
Name-value arguments must appear after other arguments, but the order of the
pairs does not matter.
Before R2021a, use commas to separate each name and value, and enclose
Name in quotes.
Example: m = memmapfile('myFile.dat','Format','int32','Offset',255) maps int32 data
in the file, myFile.dat, to memory starting from
the 256th byte.
Write permission, specified as true or
false. If you specify false, the mapped region
is read-only. If you specify true, the mapped region has both read
and write permissions.
Example: 'Writable',true
Data Types: logical
Distance from start of file to start of mapped region, specified as a nonnegative integer. This value is zero-based, a value of 0 represents the start of the file.
Example: 'Offset',1024
Data Types: double
Format of the mapped region, specified as a character vector, string scalar, or
n-by-3 cell array.
If the region you are mapping contains data of only one type, specify the Format
value as a character vector or string scalar identifying the type. For example, if
your data consists of only 16-bit signed integers, specify 'int16'.
You can use any of the following data types when you specify a
Format value:
'int8''int16''int32''int64''uint8''uint16''uint32''uint64''single''double'
If the region you are mapping requires you to specify an array shape to the data
in the mapped file, as well as a field name to reference this array, specify the
Format value as a 1-by-3 cell array.
The first cell value, specified as a character vector or string scalar, assigns the data type to apply to the mapped region.
The second cell value, specified as a 1-by-n array, assigns the array dimensions to apply to the mapped region.
The third cell value, specified as a character vector or string scalar, assigns the field name to use in the
Datastructure array of the memory map. For example,{'uint64',[30 4 10],'x'}
If the region you are mapping is composed of segments of varying data types or
array shapes, you can specify each segment's format using the rows of an n-by-3 cell
array. For example, {'uint64',[30 4 10],'x'; 'uint32',[30 4
6],'y'}
Data Types: char | string | cell
Number of times to apply the Format parameter to the mapped region,
specified as Inf or a positive integer. If you specify
Inf, memmapfile applies the
Format parameter until the end of the file.
Example: 'Repeat',2000
Data Types: double
Output Arguments
Memory map, returned as a memmapfile object
with the following properties.
| Property | Description |
|---|---|
| Path and name of the mapped file |
| Type of access allowed to the mapped region |
| Number of bytes from the start of the file to the start of the mapped region |
| Format of the contents of the mapped region, including data type, array size, and field name by which to access the data |
| Number of times to apply the pattern specified by the Format property
to the mapped region of the file |
| Memory-mapped data from the file. Data can
be a numeric array or a structure array with field names specified
in the Format property |
The values for any property (except for Data)
are set at the time you call memmapfile, using
name-value pair arguments.
Access any property of m with dot notation
similar to accessing fields of a structure array. For example, to
access the memory-mapped data in the Data property,
do one of the following:
If
Datais a numeric array, callm.Data.If
Datais a scalar (1-by-1) structure array, callm.Data., wherefieldnamefieldnameis the name of a field.If
Datais a nonscalar structure array, callm.Data(whereindex).fieldnameindexis the index for the element in the structure array, andfieldnameis the name of a field. For example, to access the file data in thetemperaturefield of the first element ofData, callm.Data(1).temperature.
After you create a memory map, m, you can
change the value of any of its properties, except for Data.
To assign a new value, use dot notation. For example, to set a new Offset value
for m, type:
m.Offset = 2048;
Tips
You can map only an existing file. You cannot create a new file and map that file to memory in one operation. Use the MATLAB file I/O functions to create the file before attempting to map it to memory.
After
memmapfilelocates the file, MATLAB stores the file’s absolute pathname internally, and then uses this stored path to locate the file from that point on. As a result, you can work in other directories outside your current work directory and retain access to the mapped file.memmapfiledoes not expand or append to a mapped file. Use instead standard file I/O functions likefopenandfwrite.
Algorithms
The actual mapping of a file to the MATLAB address space
does not take place when you construct a memmapfile object.
A memory map, based on the information currently stored in the mapped
object, is generated the first time you reference or modify the Data property
for that object.
Extended Capabilities
This function fully supports thread-based environments. For more information, see Run MATLAB Functions in Thread-Based Environment.
Version History
Introduced before R2006aThis function supports thread-based environments.
See Also
MATLAB Command
You clicked a link that corresponds to this MATLAB command:
Run the command by entering it in the MATLAB Command Window. Web browsers do not support MATLAB commands.
Select a Web Site
Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .
You can also select a web site from the following list
How to Get Best Site Performance
Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.
Americas
- América Latina (Español)
- Canada (English)
- United States (English)
Europe
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)