We Keep Coding

Linq to objects is 20 times slower than plain C#. Is there a way to speed it up?

If I need just the maximum [or the 3 biggest items] of an array, and I do it with myArray.OrderBy(...).First() [or myArray.OrderBy(...).Take(3)], it is 20 times slower than calling myArray.Max(). Is there a way to write a faster linq query? This is my sample:
using System;
using System.Linq;
namespace ConsoleApp1
{
class Program
{
static void Main(string[] args)
{
var array = new int[1000000];
for (int i = 0; i < array.Length; i++)
{
array[i] = i;
}
var maxResults = new int[10];
var linqResults = new int[10];
var start = DateTime.Now;
for (int i = 0; i < maxResults.Length; i++)
{
maxResults[i] = array.Max();
}
var maxEnd = DateTime.Now;
for (int i = 0; i < maxResults.Length; i++)
{
linqResults[i] = array.OrderByDescending(it => it).First();
}
var linqEnd = DateTime.Now;
// 00:00:00.0748281
// 00:00:01.5321276
Console.WriteLine(maxEnd - start);
Console.WriteLine(linqEnd - maxEnd);
Console.ReadKey();
}
}
}

You sort the initial array 10 times in a loop:
for (int i = 0; i < maxResults.Length; i++)
{
linqResults[i] = array.OrderByDescending(it => it).First();
}
Let's do it once:
// 10 top item of the array
var linqResults = array
.OrderByDescending(it => it)
.Take(10)
.ToArray();
please, note, that
for (int i = 0; i < maxResults.Length; i++)
{
maxResults[i] = array.Max();
}
just repeat the same Max value 10 times (it doesn't return 10 top items)

Max method time consumption is O(n) and Ordering in the best time is O(n log(n))
First error of your code is that you are ordering 10 times which is worst scenario. You can order one time and take 10 of them like what Dmitry answered.
And also, calling Max method for 10 times does not give you 10 biggest values, just the biggest value for 10 times.
However Max method does iterating list once and keep the Max value in a seperate variable. You can rewrite this method to iterate you array and keep you 10 biggest values in your maxResults and this is the fastest way that you can get result.

It seems that others have filled the efficiency gap that Microsoft has left in linq-to-objects:
https://morelinq.github.io/3.1/ref/api/html/M_MoreLinq_MoreEnumerable_PartialSort__1_3.htm

Initiate a float list with zeros in C#

I want to initiate a list of N objects with zeros( 0.0 ) . I thought of doing it like that:
var TempList = new List<float>(new float[(int)(N)]);
Is there any better(more efficeint) way to do that?

Your current solution creates an array with the sole purpose of initialising a list with zeros, and then throws that array away. This might appear to be not efficient. However, as we shall see, it is in fact very efficient!
Here's a method that doesn't create an intermediary array:
int n = 100;
var list = new List<float>(n);
for (int i = 0; i < n; ++i)
list.Add(0f);
Alternatively, you can use Enumerable.Repeat() to provide 0f "n" times, like so:
var list = new List<float>(n);
list.AddRange(Enumerable.Repeat(0f, n));
But both these methods turn out to be a slower!
Here's a little test app to do some timings.
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.Linq;
namespace Demo
{
public class Program
{
private static void Main()
{
var sw = new Stopwatch();
int n = 1024*1024*16;
int count = 10;
int dummy = 0;
for (int trial = 0; trial < 4; ++trial)
{
sw.Restart();
for (int i = 0; i < count; ++i)
dummy += method1(n).Count;
Console.WriteLine("Enumerable.Repeat() took " + sw.Elapsed);
sw.Restart();
for (int i = 0; i < count; ++i)
dummy += method2(n).Count;
Console.WriteLine("list.Add() took " + sw.Elapsed);
sw.Restart();
for (int i = 0; i < count; ++i)
dummy += method3(n).Count;
Console.WriteLine("(new float[n]) took " + sw.Elapsed);
Console.WriteLine("\n");
}
}
private static List<float> method1(int n)
{
var list = new List<float>(n);
list.AddRange(Enumerable.Repeat(0f, n));
return list;
}
private static List<float> method2(int n)
{
var list = new List<float>(n);
for (int i = 0; i < n; ++i)
list.Add(0f);
return list;
}
private static List<float> method3(int n)
{
return new List<float>(new float[n]);
}
}
}
Here's my results for a RELEASE build:
Enumerable.Repeat() took 00:00:02.9508207
list.Add() took 00:00:01.1986594
(new float[n]) took 00:00:00.5318123
So it turns out that creating an intermediary array is quite a lot faster. However, be aware that this testing code is flawed because it doesn't account for garbage collection overhead caused by allocating the intermediary array (which is very hard to time properly).
Finally, there is a REALLY EVIL, NASTY way you can optimise this using reflection. But this is brittle, will probably fail to work in the future, and should never, ever be used in production code.
I present it here only as a curiosity:
private static List<float> method4(int n)
{
var list = new List<float>(n);
list.GetType().GetField("_size", BindingFlags.NonPublic | BindingFlags.Instance).SetValue(list, n);
return list;
}
Doing this reduces the time to less than a tenth of a second, compared to the next fastest method which takes half a second. But don't do it.

What is wrong with
float[] A = new float[N];
or
List<float> A = new List<float>(N);
Note that trying to micromanage the compiler is not optimization. Start with the cleanest code that does what you want and let the compiler do its thing.
Edit 1
The solution with List<float> produces an empty list, with only internally N items initialized. So we can trick it with some reflection
static void Main(string[] args)
{
int N=100;
float[] array = new float[N];
List<float> list=new List<float>(N);
var size=typeof(List<float>).GetField("_size", BindingFlags.Instance|BindingFlags.NonPublic);
size.SetValue(list, N);
// Now list has 100 zero items
}

Why not:
var itemsWithZeros = new float[length];

Segmented Aggregation within an Array

I have a large array of primitive value-types. The array is in fact one dimentional, but logically represents a 2-dimensional field. As you read from left to right, the values need to become (the original value of the current cell) + (the result calculated in the cell to the left). Obviously with the exception of the first element of each row which is just the original value.
I already have an implementation which accomplishes this, but is entirely iterative over the entire array and is extremely slow for large (1M+ elements) arrays.
Given the following example array,
0 0 1 0 0
2 0 0 0 3
0 4 1 1 0
0 1 0 4 1
Becomes
0 0 1 1 1
2 2 2 2 5
0 4 5 6 6
0 1 1 5 6
And so forth to the right, up to problematic sizes (1024x1024)
The array needs to be updated (ideally), but another array can be used if necessary. Memory footprint isn't much of an issue here, but performance is critical as these arrays have millions of elements and must be processed hundreds of times per second.
The individual cell calculations do not appear to be parallelizable given their dependence on values starting from the left, so GPU acceleration seems impossible. I have investigated PLINQ but requisite for indices makes it very difficult to implement.
Is there another way to structure the data to make it faster to process?
If efficient GPU processing is feasible using an innovative teqnique, this would be vastly preferable, as this is currently texture data which is having to be pulled from and pushed back to the video card.

Proper coding and a bit of insight in how .NET knows stuff helps as well :-)
Some rules of thumb that apply in this case:
If you can hint the JIT that the indexing will never get out of bounds of the array, it will remove the extra branche.
You should vectorize it only in multiple threads if it's really slow (f.ex. >1 second). Otherwise task switching, cache flushes etc will probably just eat up the added speed and you'll end up worse.
If possible, make memory access predictable, even sequential. If you need another array, so be it - if not, prefer that.
Use as few IL instructions as possible if you want speed. Generally this seems to work.
Test multiple iterations. A single iteration might not be good enough.
using these rules, you can make a small test case as follows. Note that I've upped the stakes to 4Kx4K since 1K is just so fast you cannot measure it :-)
public static void Main(string[] args)
{
int width = 4096;
int height = 4096;
int[] ar = new int[width * height];
Random rnd = new Random(213);
for (int i = 0; i < ar.Length; ++i)
{
ar[i] = rnd.Next(0, 120);
}
// (5)...
for (int j = 0; j < 10; ++j)
{
Stopwatch sw = Stopwatch.StartNew();
int sum = 0;
for (int i = 0; i < ar.Length; ++i) // (3) sequential access
{
if ((i % width) == 0)
{
sum = 0;
}
// (1) --> the JIT will notice this won't go out of bounds because [0<=i<ar.Length]
// (5) --> '+=' is an expression generating a 'dup'; this creates less IL.
ar[i] = (sum += ar[i]);
}
Console.WriteLine("This took {0:0.0000}s", sw.Elapsed.TotalSeconds);
}
Console.ReadLine();
}
One of these iterations wil take roughly 0.0174 sec here, and since this is about 16x the worst case scenario you describe, I suppose your performance problem is solved.
If you really want to parallize it to make it faster, I suppose that is possible, even though you will loose some of the optimizations in the JIT (Specifically: (1)). However, if you have a multi-core system like most people, the benefits might outweight these:
for (int j = 0; j < 10; ++j)
{
Stopwatch sw = Stopwatch.StartNew();
Parallel.For(0, height, (a) =>
{
int sum = 0;
for (var i = width * a + 1; i < width * (a + 1); i++)
{
ar[i] = (sum += ar[i]);
}
});
Console.WriteLine("This took {0:0.0000}s", sw.Elapsed.TotalSeconds);
}
If you really, really need performance, you can compile it to C++ and use P/Invoke. Even if you don't use the GPU, I suppose the SSE/AVX instructions might already give you a significant performance boost that you won't get with .NET/C#. Also I'd like to point out that the Intel C++ compiler can automatically vectorize your code - even to Xeon PHI's. Without a lot of effort, this might give you a nice boost in performance.

Well, I don't know too much about GPU, but I see no reason why you can't parallelize it as the dependencies are only from left to right.
There are no dependencies between rows.
0 0 1 0 0 - process on core1 |
2 0 0 0 3 - process on core1 |
-------------------------------
0 4 1 1 0 - process on core2 |
0 1 0 4 1 - process on core2 |
Although the above statement is not completely true. There's still hidden dependencies between rows when it comes to memory cache.
It's possible that there's going to be cache trashing. You can read about "cache false sharing", in order to understand the problem, and see how to overcome that.

As #Chris Eelmaa told you it is possible to do a parallel execution by row. Using Parallel.For could be rewritten like this:
static int[,] values = new int[,]{
{0, 0, 1, 0, 0},
{2, 0, 0, 0, 3},
{0, 4, 1, 1, 0},
{0, 1, 0, 4 ,1}};
static void Main(string[] args)
{
int rows=values.GetLength(0);
int columns=values.GetLength(1);
Parallel.For(0, rows, (row) =>
{
for (var column = 1; column < columns; column++)
{
values[row, column] += values[row, column - 1];
}
});
for (var row = 0; row < rows; row++)
{
for (var column = 0; column < columns; column++)
{
Console.Write("{0} ", values[row, column]);
}
Console.WriteLine();
}
So, as stated in your question, you have a one dimensional array, the code would be a bit faster:
static void Main(string[] args)
{
var values = new int[1024 * 1024];
Random r = new Random();
for (int i = 0; i < 1024; i++)
{
for (int j = 0; j < 1024; j++)
{
values[i * 1024 + j] = r.Next(25);
}
}
int rows = 1024;
int columns = 1024;
Stopwatch sw = Stopwatch.StartNew();
for (int i = 0; i < 100; i++)
{
Parallel.For(0, rows, (row) =>
{
for (var column = 1; column < columns; column++)
{
values[(row * columns) + column] += values[(row * columns) + column - 1];
}
});
}
Console.WriteLine(sw.Elapsed);
}
But not as fast as a GPU. To use parallel GPU processing you will have to rewrite it in C++ AMP or take a look on how to port this parallel for to cudafy: http://w8isms.blogspot.com.es/2012/09/cudafy-me-part-3-of-4.html

You may as well store the array as a jagged array, the memory layout will be the same. So, instead of,
int[] texture;
you have,
int[][] texture;
Isolate the row operation as,
private static Task ProcessRow(int[] row)
{
var v = row[0];
for (var i = 1; i < row.Length; i++)
{
v = row[i] += v;
}
return Task.FromResult(true);
}
then you can write a function that does,
Task.WhenAll(texture.Select(ProcessRow)).Wait();
If you want to remain with a 1-dimensional array, a similar approach will work, just change ProcessRow.
private static Task ProcessRow(int[] texture, int start, int limit)
{
var v = texture[start];
for (var i = start + 1; i < limit; i++)
{
v = texture[i] += v;
}
return Task.FromResult(true);
}
then once,
var rowSize = 1024;
var rows =
Enumerable.Range(0, texture.Length / rowSize)
.Select(i => Tuple.Create(i * rowSize, (i * rowSize) + rowSize))
.ToArray();
then on each cycle.
Task.WhenAll(rows.Select(t => ProcessRow(texture, t.Item1, t.Item2)).Wait();
Either way, each row is processed in parallel.

How to initialize integer array in C# [duplicate]

This question already has answers here:
Closed 10 years ago.
Possible Duplicate:
c# Leaner way of initializing int array
Basically I would like to know if there is a more efficent code than the one shown below
private static int[] GetDefaultSeriesArray(int size, int value)
{
int[] result = new int[size];
for (int i = 0; i < size; i++)
{
result[i] = value;
}
return result;
}
where size can vary from 10 to 150000. For small arrays is not an issue, but there should be a better way to do the above.
I am using VS2010(.NET 4.0)

C#/CLR does not have built in way to initalize array with non-default values.
Your code is as efficient as it could get if you measure in operations per item.
You can get potentially faster initialization if you initialize chunks of huge array in parallel. This approach will need careful tuning due to non-trivial cost of mutlithread operations.
Much better results can be obtained by analizing your needs and potentially removing whole initialization alltogether. I.e. if array is normally contains constant value you can implement some sort of COW (copy on write) approach where your object initially have no backing array and simpy returns constant value, that on write to an element it would create (potentially partial) backing array for modified segment.
Slower but more compact code (that potentially easier to read) would be to use Enumerable.Repeat. Note that ToArray will cause significant amount of memory to be allocated for large arrays (which may also endup with allocations on LOH) - High memory consumption with Enumerable.Range?.
var result = Enumerable.Repeat(value, size).ToArray();

One way that you can improve speed is by utilizing Array.Copy. It's able to work at a lower level in which it's bulk assigning larger sections of memory.
By batching the assignments you can end up copying the array from one section to itself.
On top of that, the batches themselves can be quite effectively paralleized.
Here is my initial code up. On my machine (which only has two cores) with a sample array of size 10 million items, I was getting a 15% or so speedup. You'll need to play around with the batch size (try to stay in multiples of your page size to keep it efficient) to tune it to the size of items that you have. For smaller arrays it'll end up almost identical to your code as it won't get past filling up the first batch, but it also won't be (noticeably) worse in those cases either.
private const int batchSize = 1048576;
private static int[] GetDefaultSeriesArray2(int size, int value)
{
int[] result = new int[size];
//fill the first batch normally
int end = Math.Min(batchSize, size);
for (int i = 0; i < end; i++)
{
result[i] = value;
}
int numBatches = size / batchSize;
Parallel.For(1, numBatches, batch =>
{
Array.Copy(result, 0, result, batch * batchSize, batchSize);
});
//handle partial leftover batch
for (int i = numBatches * batchSize; i < size; i++)
{
result[i] = value;
}
return result;
}

Another way to improve performance is with a pretty basic technique: loop unrolling.
I have written some code to initialize an array with 20 million items, this is done repeatedly 100 times and an average is calculated. Without unrolling the loop, this takes about 44 MS. With loop unrolling of 10 the process is finished in 23 MS.
private void Looper()
{
int repeats = 100;
float avg = 0;
ArrayList times = new ArrayList();
for (int i = 0; i < repeats; i++)
times.Add(Time());
Console.WriteLine(GetAverage(times)); //44
times.Clear();
for (int i = 0; i < repeats; i++)
times.Add(TimeUnrolled());
Console.WriteLine(GetAverage(times)); //22
}
private float GetAverage(ArrayList times)
{
long total = 0;
foreach (var item in times)
{
total += (long)item;
}
return total / times.Count;
}
private long Time()
{
Stopwatch sw = new Stopwatch();
int size = 20000000;
int[] result = new int[size];
sw.Start();
for (int i = 0; i < size; i++)
{
result[i] = 5;
}
sw.Stop();
Console.WriteLine(sw.ElapsedMilliseconds);
return sw.ElapsedMilliseconds;
}
private long TimeUnrolled()
{
Stopwatch sw = new Stopwatch();
int size = 20000000;
int[] result = new int[size];
sw.Start();
for (int i = 0; i < size; i += 10)
{
result[i] = 5;
result[i + 1] = 5;
result[i + 2] = 5;
result[i + 3] = 5;
result[i + 4] = 5;
result[i + 5] = 5;
result[i + 6] = 5;
result[i + 7] = 5;
result[i + 8] = 5;
result[i + 9] = 5;
}
sw.Stop();
Console.WriteLine(sw.ElapsedMilliseconds);
return sw.ElapsedMilliseconds;
}

Enumerable.Repeat(value, size).ToArray();

Reading up Enumerable.Repeat is 20 times slower than the ops standard for loop and the only thing I found which might improve its speed is
private static int[] GetDefaultSeriesArray(int size, int value)
{
int[] result = new int[size];
for (int i = 0; i < size; ++i)
{
result[i] = value;
}
return result;
}
NOTE the i++ is changed to ++i. i++ copies i, increments i, and returns the original value. ++i just returns the incremented value

As someone already mentioned, you can leverage parallel processing like this:
int[] result = new int[size];
Parallel.ForEach(result, x => x = value);
return result;
Sorry I had no time to do performance testing on this (don't have VS installed on this machine) but if you can do it and share the results it would be great.
EDIT: As per comment, while I still think that in terms of performance they are equivalent, you can try the parallel for loop:
Parallel.For(0, size, i => result[i] = value);

C# FindAll VS Where Speed

Anyone know any speed differences between Where and FindAll on List. I know Where is part of IEnumerable and FindAll is part of List, I'm just curious what's faster.

The FindAll method of the List<T> class actually constructs a new list object, and adds results to it. The Where extension method for IEnumerable<T> will simply iterate over an existing list and yield an enumeration of the matching results without creating or adding anything (other than the enumerator itself.)
Given a small set, the two would likely perform comparably. However, given a larger set, Where should outperform FindAll, as the new List created to contain the results will have to dynamically grow to contain additional results. Memory usage of FindAll will also start to grow exponentially as the number of matching results increases, where as Where should have constant minimal memory usage (in and of itself...excluding whatever you do with the results.)

FindAll is obviously slower than Where, because it needs to create a new list.
Anyway, I think you really should consider Jon Hanna comment - you'll probably need to do some operations on your results and list would be more useful than IEnumerable in many cases.
I wrote small test, just paste it in Console App project. It measures time/ticks of: function execution, operations on results collection(to get perf. of 'real' usage, and to be sure that compiler won't optimize unused data etc. - I'm new to C# and don't know how it works yet,sorry).
Notice: every measured function except WhereIENumerable() creates new List of elements. I might be doing something wrong, but clearly iterating IEnumerable takes much more time than iterating list.
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Diagnostics;
namespace Tests
{
public class Dummy
{
public int Val;
public Dummy(int val)
{
Val = val;
}
}
public class WhereOrFindAll
{
const int ElCount = 20000000;
const int FilterVal =1000;
const int MaxVal = 2000;
const bool CheckSum = true; // Checks sum of elements in list of resutls
static List<Dummy> list = new List<Dummy>();
public delegate void FuncToTest();
public static long TestTicks(FuncToTest function, string msg)
{
Stopwatch watch = new Stopwatch();
watch.Start();
function();
watch.Stop();
Console.Write("\r\n"+msg + "\t ticks: " + (watch.ElapsedTicks));
return watch.ElapsedTicks;
}
static void Check(List<Dummy> list)
{
if (!CheckSum) return;
Stopwatch watch = new Stopwatch();
watch.Start();
long res=0;
int count = list.Count;
for (int i = 0; i < count; i++) res += list[i].Val;
for (int i = 0; i < count; i++) res -= (long)(list[i].Val * 0.3);
watch.Stop();
Console.Write("\r\n\nCheck sum: " + res.ToString() + "\t iteration ticks: " + watch.ElapsedTicks);
}
static void Check(IEnumerable<Dummy> ieNumerable)
{
if (!CheckSum) return;
Stopwatch watch = new Stopwatch();
watch.Start();
IEnumerator<Dummy> ieNumerator = ieNumerable.GetEnumerator();
long res = 0;
while (ieNumerator.MoveNext()) res += ieNumerator.Current.Val;
ieNumerator=ieNumerable.GetEnumerator();
while (ieNumerator.MoveNext()) res -= (long)(ieNumerator.Current.Val * 0.3);
watch.Stop();
Console.Write("\r\n\nCheck sum: " + res.ToString() + "\t iteration ticks :" + watch.ElapsedTicks);
}
static void Generate()
{
if (list.Count > 0)
return;
var rand = new Random();
for (int i = 0; i < ElCount; i++)
list.Add(new Dummy(rand.Next(MaxVal)));
}
static void For()
{
List<Dummy> resList = new List<Dummy>();
int count = list.Count;
for (int i = 0; i < count; i++)
{
if (list[i].Val < FilterVal)
resList.Add(list[i]);
}
Check(resList);
}
static void Foreach()
{
List<Dummy> resList = new List<Dummy>();
int count = list.Count;
foreach (Dummy dummy in list)
{
if (dummy.Val < FilterVal)
resList.Add(dummy);
}
Check(resList);
}
static void WhereToList()
{
List<Dummy> resList = list.Where(x => x.Val < FilterVal).ToList<Dummy>();
Check(resList);
}
static void WhereIEnumerable()
{
Stopwatch watch = new Stopwatch();
IEnumerable<Dummy> iEnumerable = list.Where(x => x.Val < FilterVal);
Check(iEnumerable);
}
static void FindAll()
{
List<Dummy> resList = list.FindAll(x => x.Val < FilterVal);
Check(resList);
}
public static void Run()
{
Generate();
long[] ticks = { 0, 0, 0, 0, 0 };
for (int i = 0; i < 10; i++)
{
ticks[0] += TestTicks(For, "For \t\t");
ticks[1] += TestTicks(Foreach, "Foreach \t");
ticks[2] += TestTicks(WhereToList, "Where to list \t");
ticks[3] += TestTicks(WhereIEnumerable, "Where Ienum \t");
ticks[4] += TestTicks(FindAll, "FindAll \t");
Console.Write("\r\n---------------");
}
for (int i = 0; i < 5; i++)
Console.Write("\r\n"+ticks[i].ToString());
}
}
class Program
{
static void Main(string[] args)
{
WhereOrFindAll.Run();
Console.Read();
}
}
}
Results(ticks) - CheckSum enabled(some operations on results), mode: release without debugging(CTRL+F5):
- 16,222,276 (for ->list)
- 17,151,121 (foreach -> list)
- 4,741,494 (where ->list)
- 27,122,285 (where ->ienum)
- 18,821,571 (findall ->list)
CheckSum disabled (not using returned list at all):
- 10,885,004 (for ->list)
- 11,221,888 (foreach ->list)
- 18,688,433 (where ->list)
- 1,075 (where ->ienum)
- 13,720,243 (findall ->list)
Your results can be slightly different, to get real results you need more iterations.

UPDATE(from comment): Looking through that code I agree, .Where should have, at worst, equal performance but almost always better.
Original answer:
.FindAll() should be faster, it takes advantage of already knowing the List's size and looping through the internal array with a simple for loop. .Where() has to fire up an enumerator (a sealed framework class called WhereIterator in this case) and do the same job in a less specific way.
Keep in mind though, that .Where() is enumerable, not actively creating a List in memory and filling it. It's more like a stream, so the memory use on something very large can have a significant difference. Also, you could start using the results in a parallel fashion much faster using there .Where() approach in 4.0.

Where is much, much faster than FindAll. No matter how big the list is, Where takes exactly the same amount of time.
Of course Where just creates a query. It doesn't actually do anything, unlike FindAll which does create a list.

The answer from jrista makes senses. However, the new list adds the same objects, thus just growing with reference to existing objects, which should not be that slow.
As long as 3.5 / Linq extension are possible, Where stays better anyway.
FindAll makes much more sense when limited with 2.0