An implementation of block vectors based on the parallel vector class implemented in PETScWrappers. While the base class provides for most of the interface, this class handles the actual allocation of vectors and provides functions that are specific to the underlying vector type.
The model of distribution of data is such that each of the blocks is distributed across all MPI processes named in the MPI communicator. I.e. we don't just distribute the whole vector, but each component. In the constructors and reinit() functions, one therefore not only has to specify the sizes of the individual blocks, but also the number of elements of each of these blocks to be stored on the local process.
Declare a type that has holds real-valued numbers with the same precision as the template argument to this class. If the template argument of this class is a real data type, then real_type equals the template argument. If the template argument is a std::complex type then real_type equals the type underlying the complex numbers.
This alias is used to represent the return type of norms.
Update internal structures after resizing vectors. Whenever you reinited a block of a block vector, the internal data structures are corrupted. Therefore, you should call this function after all blocks got their new size.
Return a reference on the object that describes the mapping between block and global indices. The use of this function is highly deprecated and it should vanish in one of the next versions
Return an index set that describes which elements of this vector are owned by the current processor. Note that this index set does not include elements this vector may store locally as ghost elements but that are in fact owned by another processor. As a consequence, the index sets returned on different processors if this is a distributed vector will form disjoint sets that add up to the complete index set. Obviously, if a vector is created on only one processor, then the result would satisfy
Instead of getting individual elements of a vector via operator(), this function allows getting a whole set of elements at once. The indices of the elements to be read are stated in the first argument, the corresponding values are returned in the second.
If the current vector is called v, then this function is the equivalent to the code
for (unsignedint i=0; i<indices.size(); ++i)
values[i] = v[indices[i]];
Precondition
The sizes of the indices and values arrays must be identical.
Instead of getting individual elements of a vector via operator(), this function allows getting a whole set of elements at once. In contrast to the previous function, this function obtains the indices of the elements by dereferencing all elements of the iterator range provided by the first two arguments, and puts the vector values into memory locations obtained by dereferencing a range of iterators starting at the location pointed to by the third argument.
If the current vector is called v, then this function is the equivalent to the code
ForwardIterator indices_p = indices_begin;
OutputIterator values_p = values_begin;
while (indices_p != indices_end)
{
*values_p = v[*indices_p];
++indices_p;
++values_p;
}
Precondition
It must be possible to write into as many memory locations starting at values_begin as there are iterators between indices_begin and indices_end.
Performs a combined operation of a vector addition and a subsequent inner product, returning the value of the inner product. In other words, the result of this function is the same as if the user called
The reason this function exists is that this operation involves less memory transfer than calling the two functions separately on deal.II's vector classes (Vector<Number> and LinearAlgebra::distributed::Vector<double>). This method only needs to load three vectors, this, V, W, whereas calling separate methods means to load the calling vector this twice. Since most vector operations are memory transfer limited, this reduces the time by 25% (or 50% if W equals this).
For complex-valued vectors, the scalar product in the second step is implemented as .
Return whether the vector contains only elements with value zero. This function is mainly for internal consistency check and should seldom be used when not in debug mode since it uses quite some time.
Return true if the vector has no negative entries, i.e. all entries are zero or positive. This function is used, for example, to check whether refinement indicators are really all positive (or zero).
Take an address where n_elements are stored contiguously and add them into the vector. Handles all cases which are not covered by the other two add() functions above.