OpenVDB  3.2.0
Classes | Public Types | Public Member Functions | Static Public Member Functions | Friends | List of all members
PagedArray< ValueT, Log2PageSize > Class Template Reference

Concurrent page-based linear data structure with O(1) random access and std-compliant iterators. It is primarily intended for applications that involve multi-threading of dynamically growing linear arrays with fast random access. More...

#include <PagedArray.h>

Classes

class  ConstIterator
 
struct  Fill
 
class  Iterator
 
class  Page
 
class  ValueBuffer
 

Public Types

typedef ValueT ValueType
 

Public Member Functions

 PagedArray ()
 Default constructor. More...
 
 ~PagedArray ()
 Destructor removed all allocated pages. More...
 
size_t push_back (const ValueType &value)
 Thread safe insertion, adds a new element at the end and increases the container size by one. More...
 
size_t push_back_unsafe (const ValueType &value)
 Slightly faster then the thread-safe push_back above. More...
 
ValueType pop_back ()
 Returns the last element, decrements the size by one. More...
 
void shrink_to_fit ()
 Reduce the page table to fix the current size. More...
 
ValueTypeoperator[] (size_t i)
 Return a reference to the value at the specified offset. More...
 
const ValueTypeoperator[] (size_t i) const
 Return a const-reference to the value at the specified offset. More...
 
void fill (const ValueType &v)
 Set all elements to the specified value. More...
 
void resize (size_t size)
 Resize this array to the specified size. More...
 
void resize (size_t size, const ValueType &v)
 Resize this array to the specified size and set all elements to the specified value. More...
 
size_t size () const
 Return the number of elements in this array. More...
 
size_t capacity () const
 Return the maximum number of elements that this array can contain without allocating more memory pages. More...
 
size_t freeCount () const
 Return the number of additional elements that can be added to this array without allocating more memory pages. More...
 
size_t pageCount () const
 Return the number of allocated memory pages. More...
 
size_t memUsage () const
 Return the memory footprint of this array in bytes. More...
 
bool isEmpty () const
 Return true if the container contains no elements. More...
 
bool isPartiallyFull () const
 Return true if the page table is partially full, i.e. the last non-empty page contains less than pageSize() elements. More...
 
void clear ()
 Removes all elements from the array and delete all pages. More...
 
Iterator begin ()
 Return a non-const iterator pointing to the first element. More...
 
Iterator end ()
 Return a non-const iterator pointing to the past-the-last element. More...
 
ConstIterator cbegin () const
 Return a const iterator pointing to the first element. More...
 
ConstIterator cend () const
 Return a const iterator pointing to the past-the-last element. More...
 
void sort ()
 Parallel sort of all the elements in ascending order. More...
 
void invSort ()
 Parallel sort of all the elements in descending order. More...
 
template<typename Functor >
void sort ()
 Parallel sort of all the elements based on a custom functor with the api: More...
 
void merge (PagedArray &other)
 Transfer all the elements (and pages) from the other array to this array. More...
 
void print (std::ostream &os=std::cout) const
 Print information for debugging. More...
 

Static Public Member Functions

static size_t pageSize ()
 Return the number of elements per memory page. More...
 
static size_t log2PageSize ()
 Return log2 of the number of elements per memory page. More...
 

Friends

class ValueBuffer
 

Detailed Description

template<typename ValueT, size_t Log2PageSize = 10UL>
class openvdb::v3_2_0::util::PagedArray< ValueT, Log2PageSize >

Concurrent page-based linear data structure with O(1) random access and std-compliant iterators. It is primarily intended for applications that involve multi-threading of dynamically growing linear arrays with fast random access.

Note
Multiple threads can grow the page-table and push_back new elements concurrently. A ValueBuffer provides accelerated and threadsafe push_back at the cost of potentially re-ordering elements (when multiple instances are used).

This data structure employes contiguous pages of elements (like a std::deque) which avoids moving data when the capacity is out-grown and new pages are allocated. The size of the pages can be controlled with the Log2PageSize template parameter (defaults to 1024 elements of type ValueT).

There are three fundamentally different ways to insert elements to this container - each with different advanteges and disadvanteges.

The simplest way to insert elements is to use PagedArray::push_back e.g.

PagedArray<int> array;
for (int i=0; i<100000; ++i) array.push_back(i);

or with tbb task-based multi-threading

struct Functor1 {
Functor1(int n, PagedArray<int>& _array) : array(&_array) {
tbb::parallel_for(tbb::blocked_range<int>(0, n, PagedArray<int>::pageSize()), *this);
}
void operator()(const tbb::blocked_range<int>& r) const {
for (int i=r.begin(), n=r.end(); i!=n; ++i) array->push_back(i);
}
PagedArray<int>* array;
};
PagedArray<int> array;
Functor1 tmp(10000, array);

PagedArray::push_back has the advantage that it's thread-safe and preserves the ordering of the inserted elements. In fact it returns the linear offset to the added element which can then be used for fast O(1) random access. The disadvantage is it's the slowest of the three different ways of inserting elements.

The fastest way (by far) to insert elements is to use one (or more) instances of a PagedArray::ValueBuffer, e.g.

PagedArray<int> array;
for (int i=0; i<100000; ++i) buffer.push_back(i);
buffer.flush();

or

PagedArray<int> array;
{//local scope of a single thread
for (int i=0; i<100000; ++i) buffer.push_back(i);
}

or with tbb task-based multi-threading

struct Functor2 {
Functor2(int n, PagedArray<int>& array) : buffer(array) {
tbb::parallel_for(tbb::blocked_range<int>(0, n, PagedArray<int>::pageSize()), *this);
}
void operator()(const tbb::blocked_range<int>& r) const {
for (int i=r.begin(), n=r.end(); i!=n; ++i) buffer.push_back(i);
}
mutable typename PagedArray<int>::ValueBuffer buffer;
};
PagedArray<int> array;
Functor2 tmp(10000, array);

or with tbb Thread Local Storage for even better performance (due to fewer concurrent instantiations of partially full ValueBuffers)

struct Functor3 {
typedef tbb::enumerable_thread_specific<PagedArray<int>::ValueBuffer> PoolType;
Functor3(size_t n, PoolType& _pool) : pool(&_pool) {
tbb::parallel_for(tbb::blocked_range<int>(0, n, PagedArray<int>::pageSize()), *this);
}
void operator()(const tbb::blocked_range<int>& r) const {
PagedArray<int>::ValueBuffer& buffer = pool->local();
for (int i=r.begin(), n=r.end(); i!=n; ++i) buffer.push_back(i);
}
PoolType* pool;
};
PagedArray<int> array;
PagedArray<int>::ValueBuffer exemplar(array);//dummy used for initialization
Functor3::PoolType pool(exemplar);//thread local storage pool of ValueBuffers
Functor3 tmp(10000, pool);
for (Functor3::PoolType::iterator i=pool.begin(); i!=pool.end(); ++i) i->flush();

This technique generally outperforms PagedArray::push_back, std::vector::push_back, std::deque::push_back and even tbb::concurrent_vector::push_back. Additionally it is thread-safe as long as each thread has it's own instance of a PagedArray::ValueBuffer. The only disadvantage is the ordering of the elements is undefined if multiple instance of a PagedArray::ValueBuffer are employed. This is typically the case in the context of multi-threading, where the ordering of inserts are undefined anyway. Note that a local scope can be used to guarentee that the ValueBuffer has inerted all its elements by the time the scope ends. Alternatively the ValueBuffer can be explicitly flushed by calling ValueBuffer::flush.

The third way to insert elements is to resize the container and use random access, e.g.

PagedArray<int> array;
array.resize(100000);
for (int i=0; i<100000; ++i) array[i] = i;

or in terms of the random access iterator

PagedArray<int> array;
array.resize(100000);
for (PagedArray<int>::Iterator i=array.begin(); i!=array.end(); ++i) *i = i.pos();

While this approach is both fast and thread-safe it suffers from the major disadvantage that the problem size, i.e. number of elements, needs to be known in advance. If that's the case you might as well consider using std::vector or a raw c-style array! In other words the PagedArray is most useful in the context of applications that involve multi-threading of dynamically growing linear arrays that require fast random access.

Member Typedef Documentation

◆ ValueType

typedef ValueT ValueType

Constructor & Destructor Documentation

◆ PagedArray()

PagedArray ( )
inline

Default constructor.

◆ ~PagedArray()

~PagedArray ( )
inline

Destructor removed all allocated pages.

Member Function Documentation

◆ begin()

Iterator begin ( )
inline

Return a non-const iterator pointing to the first element.

◆ capacity()

size_t capacity ( ) const
inline

Return the maximum number of elements that this array can contain without allocating more memory pages.

◆ cbegin()

ConstIterator cbegin ( ) const
inline

Return a const iterator pointing to the first element.

◆ cend()

ConstIterator cend ( ) const
inline

Return a const iterator pointing to the past-the-last element.

Warning
Itrator does not point to a valid element and should not be dereferenced!

◆ clear()

void clear ( )
inline

Removes all elements from the array and delete all pages.

Warning
Not thread-safe!

◆ end()

Iterator end ( )
inline

Return a non-const iterator pointing to the past-the-last element.

Warning
Iterator does not point to a valid element and should not be dereferenced!

◆ fill()

void fill ( const ValueType v)
inline

Set all elements to the specified value.

◆ freeCount()

size_t freeCount ( ) const
inline

Return the number of additional elements that can be added to this array without allocating more memory pages.

◆ invSort()

void invSort ( )
inline

Parallel sort of all the elements in descending order.

◆ isEmpty()

bool isEmpty ( ) const
inline

Return true if the container contains no elements.

◆ isPartiallyFull()

bool isPartiallyFull ( ) const
inline

Return true if the page table is partially full, i.e. the last non-empty page contains less than pageSize() elements.

When the page table is partially full calling merge() or using a ValueBuffer will rearrange the ordering of existing elements.

◆ log2PageSize()

static size_t log2PageSize ( )
inlinestatic

Return log2 of the number of elements per memory page.

◆ memUsage()

size_t memUsage ( ) const
inline

Return the memory footprint of this array in bytes.

◆ merge()

void merge ( PagedArray< ValueT, Log2PageSize > &  other)

Transfer all the elements (and pages) from the other array to this array.

Note
The other PagedArray is empty on return.
Warning
The ordering of elements is undefined if this page table is partially full!

◆ operator[]() [1/2]

ValueType& operator[] ( size_t  i)
inline

Return a reference to the value at the specified offset.

Note
This random access has constant time complexity.
Warning
It is assumed that the i'th element is already allocated!

◆ operator[]() [2/2]

const ValueType& operator[] ( size_t  i) const
inline

Return a const-reference to the value at the specified offset.

Note
This random access has constant time complexity.
Warning
It is assumed that the i'th element is already allocated!

◆ pageCount()

size_t pageCount ( ) const
inline

Return the number of allocated memory pages.

◆ pageSize()

static size_t pageSize ( )
inlinestatic

Return the number of elements per memory page.

◆ pop_back()

ValueType pop_back ( )
inline

Returns the last element, decrements the size by one.

Consider subsequnetly calling shrink_to_fit to reduce the page table to match the new size.

Note
Calling this method on an empty containter is undefined (as is also the case for std containers).
Warning
If values were added to the container by means of multiple ValueBuffers the last value might not be what you expect since the ordering is generally not perserved. Only PagedArray::push_back preserves the ordering (or a single instance of a ValueBuffer).

◆ print()

void print ( std::ostream &  os = std::cout) const
inline

Print information for debugging.

◆ push_back()

size_t push_back ( const ValueType value)
inline

Thread safe insertion, adds a new element at the end and increases the container size by one.

Note
Constant time complexity. May allocate a new page.

◆ push_back_unsafe()

size_t push_back_unsafe ( const ValueType value)
inline

Slightly faster then the thread-safe push_back above.

Note
For best performance consider using the ValueBuffer!
Warning
Not thread-safe!

◆ resize() [1/2]

void resize ( size_t  size)
inline

Resize this array to the specified size.

Note
This will grow or shrink the page table.
Warning
Not thread-safe!

◆ resize() [2/2]

void resize ( size_t  size,
const ValueType v 
)
inline

Resize this array to the specified size and set all elements to the specified value.

Warning
Not thread-safe!

◆ shrink_to_fit()

void shrink_to_fit ( )

Reduce the page table to fix the current size.

Warning
Not thread-safe!

◆ size()

size_t size ( ) const
inline

Return the number of elements in this array.

◆ sort() [1/2]

void sort ( )
inline

Parallel sort of all the elements in ascending order.

◆ sort() [2/2]

void sort ( )
inline

Parallel sort of all the elements based on a custom functor with the api:

bool operator()(const ValueT& a, const ValueT& b)

which returns true if a comes before b.

Friends And Related Function Documentation

◆ ValueBuffer

friend class ValueBuffer
friend

The documentation for this class was generated from the following file: