hdf5/doxygen/dox/TechnicalNotes.dox

/** \page TN Technical Notes

\li \ref api-compat-macros
\li \ref APPDBG
\li \ref FMTDISC
\li \ref FILEIMGOPS
\li \ref subsubsec_dataset_transfer_filter
\li \ref IOFLOW
\li \ref TNMDC
\li \ref thread-safe-lib
\li \ref SWMR
\li \ref VDS
\li \ref RELVERSION
\li \ref VFL
\li <a href="https://github.com/HDFGroup/arch-doc/blob/main/An_Overview_of_the_HDF5_Library_Architecture.v2.pdf">HDF5 Library Architecture Overview</a>
\li \ref VOL_Connector

 */

/** \page IOFLOW HDF5 Raw I/O Flow Notes

\htmlinclude IOFlow.html

*/

/** \page RELVERSION HDF5 Library Release Version Numbers

\htmlinclude LibraryReleaseVersionNumbers.html

*/

/** \page VFL HDF5 Virtual File Layer

\htmlinclude VFL.html

*/

/** \page FMTDISC HDF5 File Format Discussion

\htmlinclude FileFormat.html

*/

/** \page FILEIMGOPS HDF5 File Image Operations

\htmlinclude FileImageOps.html

*/

/** \page APPDBG Debugging HDF5 Applications

\htmlinclude DebuggingHDF5Applications.html

*/

/** \page SWMR Introduction to Single-Writer/Multiple-Reader (SWMR)

\section sec_swmr_intro Introduction to SWMR
The Single-Writer / Multiple-Reader (SWMR) feature enables multiple processes to read an HDF5 file
while it is being written to (by a single process) without using locks or requiring communication between processes.
<img src=tutr-swmr1.png alt="tutr-swmr1.png" width=500>

All communication between processes must be performed via the HDF5 file. The HDF5 file under SWMR access must
reside on a system that complies with POSIX write() semantics.

The basic engineering challenge for this to work was to ensure that the readers of an HDF5 file always
see a coherent (though possibly not up to date) HDF5 file.

The issue is that when writing data there is information in the metadata cache in addition to the physical file on disk:
<img src=tutr-swmr2.png alt="tutr-swmr2.png" width=500>

However, the readers can only see the state contained in the physical file:
<img src=tutr-swmr3.png alt="tutr-swmr3.png" width=500>

The SWMR solution implements dependencies on when the metadata can be flushed to the file. This ensures that metadata cache
flush operations occur in the proper order, so that there will never be internal file pointers in the physical file
that point to invalid (unflushed) file addresses.

A beneficial side effect of using SWMR access is better fault tolerance. It is more difficult to corrupt a file when using SWMR.

\subsection subsec_swmr_doc Documentation
\subsubsection subsubsec_swmr_doc_guide User Guide
<a href="https://docs.hdfgroup.org/documentation/HDF5/features/SWMR/HDF5_SWMR_Users_Guide.pdf">SWMR User Guide</a>

\subsubsection subsubsec_swmr_doc_apis HDF5 Library APIs
<ul>
<li>#H5Fstart_swmr_write — Enables SWMR writing mode for a file</li>
<li>#H5DOappend — Appends data to a dataset along a specified dimension</li>
<li>#H5Pset_object_flush_cb — Sets a callback function to invoke when an object flush occurs in the file</li>
<li>#H5Pget_object_flush_cb — Retrieves the object flush property values from the file access property list</li>
<li>#H5Odisable_mdc_flushes — Prevents metadata entries for an HDF5 object from being flushed from the metadata cache to storage</li>
<li>#H5Oenable_mdc_flushes — Enables flushing of dirty metadata entries from a file’s metadata cache</li>
<li>#H5Oare_mdc_flushes_disabled — Determines if an HDF5 object has had flushes of metadata entries disabled</li>
</ul>

\subsubsection subsubsec_swmr_doc_tools Tools
\li h5watch — Outputs new records appended to a dataset as the dataset grows
\li h5format_convert — Converts the layout format version and chunked indexing types of datasets created with
HDF5-1.10 so that applications built with HDF5-1.8 can access them
\li h5clear — Clears superblock status_flags field, removes metadata cache image, prints EOA and EOF, or sets EOA of a file

\subsubsection subsubsec_swmr_doc_design Design Documents

\subsection subsec_swmr_model Programming Model
Please be aware that the SWMR feature requires that an HDF5 file be created with the latest file format. See
#H5Pset_libver_bounds for more information.

To use SWMR follow the the general programming model for creating and accessing HDF5 files and objects along with the steps described below.

\subsubsection subsubsec_swmr_model_writer SWMR Writer
The SWMR writer either opens an existing file and objects or creates them as follows.

Open an existing file:
Call #H5Fopen using the #H5F_ACC_SWMR_WRITE flag.
Begin writing datasets.
Periodically flush data.

Create a new file:
Call #H5Fcreate using the latest file format.
Create groups, datasets and attributes, and then close the attributes.
Call #H5Fstart_swmr_write to start SWMR access to the file.
Periodically flush data.

<h4 id="example-code">Example Code:</h4>
Create the file using the latest file format property:
\code
   fapl = H5Pcreate (H5P_FILE_ACCESS);
   status = H5Pset_libver_bounds (fapl, H5F_LIBVER_LATEST, H5F_LIBVER_LATEST);
   fid = H5Fcreate (filename, H5F_ACC_TRUNC, H5P_DEFAULT, fapl);
   // Create objects (files, datasets, ...).
   // Close any attributes and named datatype objects.
   // Groups and datasets may remain open before starting SWMR access to them.

   // Start SWMR access to the file:
   status = H5Fstart_swmr_write (fid);

   // Reopen the datasets and then start writing, periodically flushing data:
   status = H5Dwrite (dset_id, ...);
   status = H5Dflush (dset_id);
\endcode   

\subsubsection subsubsec_swmr_model_reader SWMR Reader
The SWMR reader must continually poll for new data:

Call #H5Fopen using the #H5F_ACC_SWMR_READ flag.
Poll, checking the size of the dataset to see if there is new data available for reading.
Read new data, if any.

<h4 id="example-code-1">Example Code:</h4>
\code
   // Open the file using the SWMR read flag:
   fid = H5Fopen (filename, H5F_ACC_RDONLY | H5F_ACC_SWMR_READ, H5P_DEFAULT);
   // Open the dataset and then repeatedly poll the dataset, by getting the dimensions, reading new data, and refreshing:
   dset_id = H5Dopen (...);
   space_id = H5Dget_space (...);
   while (...) {
      status = H5Dread (dset_id, ...);
      status = H5Drefresh (dset_id);
      space_id = H5Dget_space (...);
   }
\endcode

\subsection subsec_swmr_scope Limitations and Scope
An HDF5 file under SWMR access must reside on a system that complies with POSIX write()
semantics. It is also limited in scope as follows.

The writer process is only allowed to modify raw data of existing datasets by;
Appending data along any unlimited dimension.
Modifying existing data
The following operations are not allowed (and the corresponding HDF5 files will fail)
\li The writer cannot add new objects to the file.
\li The writer cannot delete objects in the file.
\li The writer cannot modify or append data with variable length, string or region reference datatypes.
\li File space recycling is not allowed. As a result the size of a file modified by a SWMR writer may be larger than a file modified by a non-SWMR writer.</p>

\subsection subsec_swmr_tools Tools for Working with SWMR
Two new tools, h5watch and h5clear, are available for use with SWMR. The other HDF5 utilities have also been modified to recognize SWMR
\li The h5watch tool allows a user to monitor the growth of a dataset.
\li The h5clear tool clears the status flags in the superblock of an HDF5 file.
\li The rest of the HDF5 tools will exit gracefully but not work with SWMR otherwise.

\subsection subsec_swmr_example Programming Example
A good example of using SWMR is included with the HDF5 tests in the source code. You can run it while reading
the file it creates. If you then interrupt the application and reader and look at the resulting file, you will
see that the file is still valid. Follow these steps:
\li Download the HDF5 source code to a local directory on a filesystem (that complies with POSIX write() semantics).
Build the software. No special configuration options are needed to use SWMR.
\li Invoke two command terminal windows. In one window go into the bin directory of the built binaries.
In the other window go into the test directory of the HDF5-1.10 source code that was just built.
\li In the window in the test directory compile and run use_append_chunk.c. The example writes a three
dimensional dataset by planes (with chunks of size 1 x 256 x 256).
\li In the other window (in the bin directory) run h5watch on the file created by
use_append_chunk.c (use_append_chunk.h5). It should be run while use_append_chunk is executing and you
will see valid data displayed with h5watch.
\li Interrupt use_append_chunk while it is running, and stop h5watch.
\li Use h5clear to clear the status flags in the superblock of the HDF5 file (use_append_chunk.h5).
\li View the file with h5dump. You will see that it is a valid file even though the application did not
close properly. It will contain data up to the point that it was interrupted.

*/

/** \page VDS Introduction to the Virtual Dataset - VDS

\section sec_vds_intro Introduction to VDS
The HDF5 Virtual Dataset (VDS) feature enables users to access data in a collection of HDF5 files as a
single HDF5 dataset and to use the HDF5 APIs to work with that dataset.

For example, your data may be collected into four files:
<img src="tutrvds-multimgs.png" alt="tutrvds-multimgs.png" width=750>

You can map the datasets in the four files into a single VDS that can be accessed just like any other dataset:
<img src="tutrvds-snglimg.png" alt="tutrvds-snglimg.png" width=500>

The mapping between a VDS and the HDF5 source datasets is persistent and transparent to an application. If a source
file is missing the fill value will be displayed.

See the Virtual (VDS) Documentation for complete details regarding the VDS feature.

The VDS feature was implemented using hyperslab selection (#H5Sselect_hyperslab). See the tutorial on
Reading From or Writing to a Subset of a Dataset for more information on selecting hyperslabs.

\subsection subsec_vds_intro_model Programming Model
To create a Virtual Dataset you simply follow the HDF5 programming model and add a few additional API calls
to map the source code datasets to the VDS.

Following are the steps for creating a Virtual Dataset:
\li Create the source datasets that will comprise the VDS
\li Create the VDS: ‐ Define a datatype and dataspace (can be unlimited)
\li Define the dataset creation property list (including fill value)
\li (Repeat for each source dataset) Map elements from the source dataset to elements of the VDS
\li Select elements in the source dataset (source selection)
\li Select elements in the virtual dataset (destination selection)
\li Map destination selections to source selections (see Functions for Working with a VDS)
\li Call H5Dcreate using the properties defined above
\li Access the VDS as a regular HDF5 dataset
\li Close the VDS when finished

<h4>Functions for Working with a VDS</h4>
The #H5Pset_virtual API sets the mapping between virtual and source datasets. This is a dataset creation property list.
Using this API will change the layout of the dataset to #H5D_VIRTUAL. As with specifying any dataset creation property
list, an instance of the property list is created, modified, passed into the dataset creation call and then closed:
\code
  dcpl = H5Pcreate (H5P_DATASET_CREATE);
  src_space = H5screate_simple ...
  status = H5Sselect_hyperslab (space, ...
  status = H5Pset_virtual (dcpl, space, SRC_FILE[i], SRC_DATASET[i], src_space);
  dset = H5Dcreate2 (file, DATASET, H5T_NATIVE_INT, space, H5P_DEFAULT, dcpl, H5P_DEFAULT);
  status = H5Pclose (dcpl);
\endcode

There are several other APIs introduced with Virtual Datasets, including query functions. For details
see the complete list of HDF5 library APIs that support Virtual Datasets.

<h4>Limitations</h4>
This feature was introduced in HDF5-1.10.

The number of source datasets is unlimited. However, there is a limit on the size of each source dataset.

\subsection subsec_vds_intro_examples Programming Examples
<em>Example 1</em>
This example creates three HDF5 files, each with a one-dimensional dataset of 6 elements. The datasets in these files
are the source datasets that are then used to create a 4 x 6 Virtual Dataset with a fill value of -1. The first three
rows of the VDS are mapped to the data from the three source datasets as shown below:
<img src="tutrvds-ex.png" alt="tutrvds-ex.png" width=500>

In this example the three source datasets are mapped to the VDS with this code:
\code>
src_space = H5Screate_simple (RANK1, dims, NULL);
for (i = 0; i &lt; 3; i++) {
    start[0] = (hsize_t)i;
    // Select i-th row in the virtual dataset; selection in the source datasets is the same.
    status = H5Sselect_hyperslab (space, H5S_SELECT_SET, start, NULL, count, block);
    status = H5Pset_virtual (dcpl, space, SRC_FILE[i], SRC_DATASET[i], src_space);
}
endcode>

After the VDS is created and closed, it is reopened. The property list is then queried to determine the
layout of the dataset and its mappings, and the data in the VDS is read and printed.

This example is in the HDF5 source code and can be obtained from here:
<h4>C Example</h4>
For details on compiling an HDF5 application: [ Compiling HDF5 Applications ]

<h4>Example 2</h4>
This example shows how to use a C-style printf statement for specifying multiple source datasets as one virtual
dataset. Only one mapping is required. In other words only one #H5Pset_virtual call is needed to map multiple datasets.
It creates a 2-dimensional unlimited VDS. Then it re-opens the file, makes queries, and reads the virtual dataset.

The source datasets are specified as A-0, A-1, A-2, and A-3. These are mapped to the virtual dataset with one call:
\code
status = H5Pset_virtual (dcpl, vspace, SRCFILE, "A-%b", src_space);
\endcode

The %b indicates that the block count of the selection in the dimension should be used.

<h4>C Example</h4>
For details on compiling an HDF5 application: [ Compiling HDF5 Applications ]

Using h5dump with a VDS
The h5dump utility can be used to view a VDS. The h5dump output for a VDS looks exactly like that for any other dataset.
If h5dump cannot find a source dataset then the fill value will be displayed.

You can determine that a dataset is a VDS by looking at its properties with
\code
 h5dump -p
\endcode
 It will display each source dataset mapping, beginning with Mapping 0. Below is an excerpt of the output of
\code
 h5dump -p
\endcode
on the vds.h5 file created in Example 1.You can see that the entire source file a.h5 is mapped to the first row of the VDS dataset.

<img src="tutrvds-map.png" alt="tutrvds-map.png" width=650>

*/