Concurrency
Download as PPTX, PDF0 likes365 views
This document discusses concurrency and parallelism in .NET. It covers three pillars of concurrency: scalability using Parallel.For, responsiveness using tasks and async/await, and consistency using locks, interlocked operations, and mutexes/events/semaphores. It also discusses challenges with shared state and race conditions when naively using Parallel.For on shared data. Several solutions are presented such as not sharing data, using smaller work chunks to reduce overhead, and running completely independent tasks in parallel. The document emphasizes the importance of measuring performance before and after optimizations.
![Which is fastest?
var ints = new int[InnerLoop];
var random = new Random();
for (var inner = 0; inner < InnerLoop; ++inner)
{
ints[inner] = random.Next();
}
// ------------------------------------------------
var ints = new int[InnerLoop];
var random = new Random();
Parallel.For(
0,
InnerLoop,
i => ints[i] = random.Next()
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-9-320.jpg)
![SHARED STATE Race condition
var ints = new int[InnerLoop];
var random = new Random();
for (var inner = 0; inner < InnerLoop; ++inner)
{
ints[inner] = random.Next();
}
// ------------------------------------------------
var ints = new int[InnerLoop];
var random = new Random();
Parallel.For(
0,
InnerLoop,
i => ints[i] = random.Next()
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-10-320.jpg)
![SHARED STATE Poor performance
var ints = new int[InnerLoop];
var random = new Random();
for (var inner = 0; inner < InnerLoop; ++inner)
{
ints[inner] = random.Next();
}
// ------------------------------------------------
var ints = new int[InnerLoop];
var random = new Random();
Parallel.For(
0,
InnerLoop,
i => ints[i] = random.Next()
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-11-320.jpg)
![2 Core CPU – L1 Cache
L1
CPU
L1
CPU
new Random ()
new int[InnerLoop]](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-23-320.jpg)
![4 Core CPU – L1 Cache
L1
CPU
L1
CPU
L1
CPU
L1
CPU
new Random ()
new int[InnerLoop]](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-24-320.jpg)
![Solution 1 – Locks
var ints = new int[InnerLoop];
var random = new Random();
Parallel.For(
0,
InnerLoop,
i => {lock (ints) {ints[i] = random.Next();}}
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-26-320.jpg)
![Solution 2 – No sharing
var ints = new int[InnerLoop];
Parallel.For(
0,
InnerLoop,
() => new Random(),
(i, pls, random) =>
{ints[i] = random.Next(); return random;},
random => {}
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-27-320.jpg)
![Parallel.For adds overhead
Level0
Level1
Level2
ints[0]
ints[1]
Level2
ints[2]
ints[3]
Level1
Level2
ints[4]
ints[5]
Level2
ints[6]
ints[7]](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-28-320.jpg)
![Solution 3 – Less overhead
var ints = new int[InnerLoop];
Parallel.For(
0,
InnerLoop / Modulus,
() => new Random(),
(i, pls, random) =>
{
var begin = i * Modulus ;
var end = begin + Modulus ;
for (var iter = begin; iter < end; ++iter)
{
ints[iter] = random.Next();
}
return random;
},
random => {}
);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-29-320.jpg)
![var ints = new int[InnerLoop];
var random = new Random();
for (var inner = 0; inner < InnerLoop; ++inner)
{
ints[inner] = random.Next();
}](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-30-320.jpg)
![Solution 4 – Independent runs
var tasks = Enumerable.Range (0, 8).Select (
i => Task.Factory.StartNew (
() =>
{
var ints = new int[InnerLoop];
var random = new Random ();
while (counter.CountDown ())
{
for (var inner = 0; inner < InnerLoop; ++inner)
{
ints[inner] = random.Next();
}
}
},
TaskCreationOptions.LongRunning))
.ToArray ();
Task.WaitAll (tasks);](https://image.slidesharecdn.com/concurrency-130829142119-phpapp02/85/Concurrency-31-320.jpg)
Concurrency
- 4. Responsive
- 6. Three pillars of Concurrency Scalability (CPU) Parallel.For Responsiveness Task async/await Consistency lock Interlocked.* Mutex/Event/Semaphore Monitor
- 7. Scalability
- 9. Which is fastest? var ints = new int[InnerLoop]; var random = new Random(); for (var inner = 0; inner < InnerLoop; ++inner) { ints[inner] = random.Next(); } // ------------------------------------------------ var ints = new int[InnerLoop]; var random = new Random(); Parallel.For( 0, InnerLoop, i => ints[i] = random.Next() );
- 10. SHARED STATE Race condition var ints = new int[InnerLoop]; var random = new Random(); for (var inner = 0; inner < InnerLoop; ++inner) { ints[inner] = random.Next(); } // ------------------------------------------------ var ints = new int[InnerLoop]; var random = new Random(); Parallel.For( 0, InnerLoop, i => ints[i] = random.Next() );
- 11. SHARED STATE Poor performance var ints = new int[InnerLoop]; var random = new Random(); for (var inner = 0; inner < InnerLoop; ++inner) { ints[inner] = random.Next(); } // ------------------------------------------------ var ints = new int[InnerLoop]; var random = new Random(); Parallel.For( 0, InnerLoop, i => ints[i] = random.Next() );
- 13. Then and now Metric VAX-11/750 (’80) Today Improvement MHz 6 3300 550x Memory MB 2 16384 8192x Memory MB/s 13 R ~10000 W ~2500 770x 190x
- 14. Then and now Metric VAX-11/750 (’80) Today Improvement MHz 6 3300 550x Memory MB 2 16384 8192x Memory MB/s 13 R ~10000 W ~2500 770x 190x Memory nsec 225 70 3x
- 15. Then and now Metric VAX-11/750 (’80) Today Improvement MHz 6 3300 550x Memory MB 2 16384 8192x Memory MB/s 13 R ~10000 W ~2500 770x 190x Memory nsec 225 70 3x Memory cycles 1.4 210 -150x
- 16. 299,792,458 m/s
- 18. Speed of light is too slow
- 19. 0.09 m/c
- 21. 99% - latency mitigation 1% - computation
- 23. 2 Core CPU – L1 Cache L1 CPU L1 CPU new Random () new int[InnerLoop]
- 24. 4 Core CPU – L1 Cache L1 CPU L1 CPU L1 CPU L1 CPU new Random () new int[InnerLoop]
- 26. Solution 1 – Locks var ints = new int[InnerLoop]; var random = new Random(); Parallel.For( 0, InnerLoop, i => {lock (ints) {ints[i] = random.Next();}} );
- 27. Solution 2 – No sharing var ints = new int[InnerLoop]; Parallel.For( 0, InnerLoop, () => new Random(), (i, pls, random) => {ints[i] = random.Next(); return random;}, random => {} );
- 29. Solution 3 – Less overhead var ints = new int[InnerLoop]; Parallel.For( 0, InnerLoop / Modulus, () => new Random(), (i, pls, random) => { var begin = i * Modulus ; var end = begin + Modulus ; for (var iter = begin; iter < end; ++iter) { ints[iter] = random.Next(); } return random; }, random => {} );
- 30. var ints = new int[InnerLoop]; var random = new Random(); for (var inner = 0; inner < InnerLoop; ++inner) { ints[inner] = random.Next(); }
- 31. Solution 4 – Independent runs var tasks = Enumerable.Range (0, 8).Select ( i => Task.Factory.StartNew ( () => { var ints = new int[InnerLoop]; var random = new Random (); while (counter.CountDown ()) { for (var inner = 0; inner < InnerLoop; ++inner) { ints[inner] = random.Next(); } } }, TaskCreationOptions.LongRunning)) .ToArray (); Task.WaitAll (tasks);
- 32. Parallel.For Only for CPU bound problems
- 33. Sharing is bad Kills performance Race conditions Dead-locks
- 34. Servers have natural concurrency Avoid Parallel.For
- 35. Act like an engineer Measure before and after
- 36. One more thing…