My MPI experience showed that the speedup as does not increase linearly with the number of nodes we use (because of the costs of communication). My experience is similar to this:.
Today a speaker said: "Magically (smiles), in some occasions we can get more speedup than the ideal one!".
He meant that ideally, when we use 4 nodes, we would get a speedup of 4. But in some occasions we can get a speedup greater than 4, with 4 nodes! The topic was related to MPI.
Is this true? If so, can anyone provide a simple example on that? Or maybe he was thinking about adding multithreading to the application (he went out of time and then had to leave ASAP, thus we could not discuss)?
Parallel efficiency (speed-up / number of parallel execution units) over unity is not at all uncommon.
The main reason for that is the total cache size available to the parallel program. With more CPUs (or cores), one has access to more cache memory. At some point, a large portion of the data fits inside the cache and this speeds up the computation considerably. Another way to look at it is that the more CPUs/cores you use, the smaller the portion of the data each one gets, until that portion could actually fit inside the cache of the individual CPU. This is sooner or later cancelled by the communication overhead though.
Also, your data shows the speed-up compared to the execution on a single node. Using OpenMP could remove some of the overhead when using MPI for intranode data exchange and therefore result in better speed-up compared to the pure MPI code.
The problem comes from the incorrectly used term ideal speed-up. Ideally, one would account for cache effects. I would rather use linear instead.
Not too sure this is on-topic here, but here goes nothing...
This super-linearity in speed-up can typically occur when you parallelise your code while distributing the data in memory with MPI. In some cases, by distributing the data across several nodes / processes, you end-up having sufficiently small chunks of data to deal with for each individual process that it fits in the cache of the processor. This cache effect might have a huge impact on the code's performance, leading to great speed-ups and compensating for the increased need of MPI communications... This can be observed in many situations, but this isn't something you can really count for for compensating a poor scalability.
Another case where you can observe this sort of super-linear scalability is when you have an algorithm where you distribute the task of finding a specific element in a large collection: by distributing your work, you can end up in one of the processes/threads finding almost immediately the results, just because it happened to be given range of indexes starting very close to the answer. But this case is even less reliable than the aforementioned cache effect.
Hope that gives you a flavour of what super-linearity is.
Cache has been mentioned, but it's not the only possible reason. For instance you could imagine a parallel program which does not have sufficient memory to store all its data structures at low node counts, but foes at high. Thus at low node counts the programmer may have been forced to write intermediate values to disk and then read them back in again, or alternatively re-calculate the data when required. However at high node counts these games are no longer required and the program can store all its data in memory. Thus super-linear speed-up is a possibility because at higher node counts the code is just doing less work by using the extra memory to avoid I/O or calculations.
Really this is the same as the cache effects noted in the other answers, using extra resources as they become available. And this is really the trick - more nodes doesn't just mean more cores, it also means more of all your resources, so as speed up really measures your core use if you can also use those other extra resources to good effect you can achieve super-linear speed up.
In implementing most algorithms (sort, search, graph traversal, etc.), there is frequently a trade-off that can be made in reducing memory accesses at the cost of additional ordinary operations.
Knuth has a useful method for comparing the complexity of various algorithm implementations by abstracting it from particular processors and only distinguishing between ordinary operations (oops) and memory operations (mems).
In compiled programs, one typically lets the compiler organise the low level operations, and hopes that the operating system will handle the question of whether data is held in cache memory (faster) or in virtual memory (slower). Furthermore, the exact number / cost of instructions is encapsulated by the compiler.
With Forth, there is no longer such encapsulation, and one is much closer to the machine, albeit perhaps to a stack machine running on top of a register processor.
Ignoring the effect of an operating system (so no memory stalls, etc.), and assuming for the moment a simple processor,
(1) Can anyone advise on how the ordinary stack operations in Forth (e.g. dup, rot, over, swap, etc.) compare with the cost of Forth's memory access fetch (#) or store (!) ?
(2) Is there a rule of thumb I can use to decide how many ordinary operations to trade-off against saving a memory access?
What I'm looking for is something like 'memory access costs as much as 50 ordinary ops, or 500 ordinary ops, or 5 ordinary ops' Ballpark is absolutely fine.
I'm trying to get a sense of the relative expense of fetch and store vs. rot, swap, dup, drop, over, correct to an order of magnitude.
This article How much time does it take to fetch one word from memory? talks about main memory stall times, with some rule of thumb type numbers, but basically you can do lots of instructions while stalling for main memory. As others have said, the numbers vary a lot between systems.
Main memory stalls is a big area of interest, especially as CPUs have more cores, but typically not much faster memory bandwidth. There is some research going on around compressing data in main memory too, so that the CPU can take advantage of 'spare' cycles and tightly packed cache lines http://oai.cwi.nl/oai/asset/15564/15564B.pdf
For those who are really interested in the details, most CPU manufacturers publish in depth guides on memory optimisations etc. mostly aimed at high end and compiler writers, but readable by all 2gl and 3gl programmers.
Ps. Go Forth.
A comparison between memory fetches and register operations is okay for assembler programs, as it is for the output of c-compilers, which is in fact an assembler program.
In Forth this question hardly makes sense. In the first place Forth is an interpreter and in using Forth one foregoes the ultimate in speed. Of course one could add an optimiser on top of Forth but then the question makes even less sense, because the output of a c-optimiser and a Forth optimiser converge to -- you guessed it -- an optimal solution.
Let's look at an elementary operation in Forth like AND.
This is implemented as
> CODE AND
> POP AX
> POP BX
> AND AX, BX
> PUSH AX
> NEXT
So we see already three memory operations for something that looks like an elementary calculation operation. It appears the Knuth metric is not applicable. Also Forth seems to be loosing big time.That is however not true. Those memory operations are all onto the L1 cache of a typical processor. That is about as efficient as local variables in small c functions,
We can compare stack operations with memory operations using VARIABLE's and the stack. The answer is simple. A VARIABLE risks a memory stall. A stack operation will almost certainly be a L1 cache hit. This is the single most important point of consideration. However the question explicitly asks not to consider it!
So there.
I've been trying to get a understanding of concurrency, and I've been trying to work out what's better, one big IORef lock or many TVars. I've came to the following guidelines, comments will be appreciated, regarding whether these are roughly right or whether I've missed the point.
Lets assume our concurrent data structure is a map m, accessed like m[i]. Lets also say we have two functions, f_easy and f_hard. The f_easy is quick, f_hard takes a long time. We'll assume the arguments to f_easy/f_hard are elements of m.
(1) If your transactions look roughly like this m[f_easy(...)] = f_hard(...), use an IORef with atomicModifyIORef. Laziness will ensure that m is only locked for a short time as it's updated with a thunk. Calculating the index effectively locks the structure (as something is going to get updated, but we don't know what yet), but once it's known what that element is, the thunk over the entire structure moves to a thunk only over that particular element, and then only that particular element is "locked".
(2) If your transactions look roughly like this m[f_hard(...)] = f_easy(...), and the don't conflict too much, use lots of TVars. Using an IORef in this case will effectively make the app single threaded, as you can't calculate two indexes at the same time (as there will be an unresolved thunk over the entire structure). TVars let you work out two indexes at the same time, however, the negative is that if two concurrent transactions both access the same element, and one of them is a write, one transaction must be scrapped, which wastes time (which could have been used elsewhere). If this happens a lot, you may be better with locks that come (via blackholing) from IORef, but if it doesn't happen very much, you'll get better parallelism with TVars.
Basically in case (2), with IORef you may get 100% efficiency (no wasted work) but only use 1.1 threads, but with TVar if you have a low number of conflicts you might get 80% efficiency but use 10 threads, so you still end up 7 times faster even with the wasted work.
Your guidelines are somewhat similar to the findings of [1] (Section 6) where the performance of the Haskell STM is analyzed:
"In particular, for programs that do not perform much work inside transactions, the commit overhead appears to be very high. To further observe this overhead, an analysis needs to be conducted on the performance of commit-time course-grain and fine-grain STM locking mechanisms."
I use atomicModifyIORef or an MVar when all the synchronization I need is something that simple locking will ensure. When looking at concurrent accesses to a data structure, it also depends on how this data structure is implemented. For example, if you store your data inside a IORef Data.Map and frequently perform read/write access then I think atmoicModifyIORef will degrade to a single thread performance, as you have conjectured, but the same will be true for a TVar Data.Map. My point is that it's important to use a data structure that is suitable for concurrent programming (balanced trees aren't).
That said, in my opinion the winning argument for using STM is composability: you can combine multiple operations into a single transactions without headaches. In general, this isn't possible using IORef or MVar without introducing new locks.
[1] The limits of software transactional memory (STM): dissecting Haskell STM applications on a many-core environment.
http://dx.doi.org/10.1145/1366230.1366241
Answer to #Clinton's comment:
If a single IORef contains all your data, you can simply use atomicModifyIORef for composition. But if you need to process lots of parallel read/write requests to that data, the performance loss might become significant, since every pair of parallel read/write requests to that data might cause a conflict.
The approach that I would try is to use a data structure where the entries themselves are stored inside a TVar (vs putting the whole data structure into a single TVar). That should reduce the possibility of livelocks, as transactions won't conflict that often.
Of course, you still want to keep your transactions as small as possible and use composability only if it's absolutely necessary to guarantee consistency. So far I haven't encountered a scenario where combining more than a few insert/lookup operations into a single transaction was necessary.
Beyond performance, I see a more fundamental reason to using TVar--the type system ensures you dont do any "unsafe" operations like readIORef or writeIORef. That your data is shared is a property of the type, not of the implementation. EDIT: unsafePerformIO is always unsafe. readIORef is only unsafe if you are also using atomicModifyIORef. At the very least wrap your IORef in a newtype and only expose a wrapped atomicModifyIORef
Beyond that, don't use IORef, use MVar or TVar
The first usage pattern you describe probably does not have nice performance characteristics. You likely end up being (almost) entirely single threaded--because of laziness no actual work happens each time you update the shared state, but whenever you need to use this shared state, the entire accumulated pile of thunks needs to be forced, and has a linear data dependency structure.
Having 80% efficiency but substantially higher parallelism allows you to exploit growing number of cores. You can expect minimal performance improvements over the coming years on single threaded code.
Many word CAS is likely coming to a processor near you in the form of "Hardware Transactional Memory" allowing STMs to become far more efficient.
Your code will be more modular--every piece of code has to be changed if you add more shared state when your design has all shared state behind a single reference. TVars and to a lesser extent MVars support natural modularity.
This could sound like a subjective question, but what I am looking for are specific instances, which you could have encountered related to this.
How to make code, cache effective/cache friendly (more cache hits, as few cache misses as possible)? From both perspectives, data cache & program cache (instruction cache),
i.e. what things in one's code, related to data structures and code constructs, should one take care of to make it cache effective.
Are there any particular data structures one must use/avoid, or is there a particular way of accessing the members of that structure etc... to make code cache effective.
Are there any program constructs (if, for, switch, break, goto,...), code-flow (for inside an if, if inside a for, etc ...) one should follow/avoid in this matter?
I am looking forward to hearing individual experiences related to making cache efficient code in general. It can be any programming language (C, C++, Assembly, ...), any hardware target (ARM, Intel, PowerPC, ...), any OS (Windows, Linux,S ymbian, ...), etc..
The variety will help to better to understand it deeply.
The cache is there to reduce the number of times the CPU would stall waiting for a memory request to be fulfilled (avoiding the memory latency), and as a second effect, possibly to reduce the overall amount of data that needs to be transfered (preserving memory bandwidth).
Techniques for avoiding suffering from memory fetch latency is typically the first thing to consider, and sometimes helps a long way. The limited memory bandwidth is also a limiting factor, particularly for multicores and multithreaded applications where many threads wants to use the memory bus. A different set of techniques help addressing the latter issue.
Improving spatial locality means that you ensure that each cache line is used in full once it has been mapped to a cache. When we have looked at various standard benchmarks, we have seen that a surprising large fraction of those fail to use 100% of the fetched cache lines before the cache lines are evicted.
Improving cache line utilization helps in three respects:
It tends to fit more useful data in the cache, essentially increasing the effective cache size.
It tends to fit more useful data in the same cache line, increasing the likelyhood that requested data can be found in the cache.
It reduces the memory bandwidth requirements, as there will be fewer fetches.
Common techniques are:
Use smaller data types
Organize your data to avoid alignment holes (sorting your struct members by decreasing size is one way)
Beware of the standard dynamic memory allocator, which may introduce holes and spread your data around in memory as it warms up.
Make sure all adjacent data is actually used in the hot loops. Otherwise, consider breaking up data structures into hot and cold components, so that the hot loops use hot data.
avoid algorithms and datastructures that exhibit irregular access patterns, and favor linear datastructures.
We should also note that there are other ways to hide memory latency than using caches.
Modern CPU:s often have one or more hardware prefetchers. They train on the misses in a cache and try to spot regularities. For instance, after a few misses to subsequent cache lines, the hw prefetcher will start fetching cache lines into the cache, anticipating the application's needs. If you have a regular access pattern, the hardware prefetcher is usually doing a very good job. And if your program doesn't display regular access patterns, you may improve things by adding prefetch instructions yourself.
Regrouping instructions in such a way that those that always miss in the cache occur close to each other, the CPU can sometimes overlap these fetches so that the application only sustain one latency hit (Memory level parallelism).
To reduce the overall memory bus pressure, you have to start addressing what is called temporal locality. This means that you have to reuse data while it still hasn't been evicted from the cache.
Merging loops that touch the same data (loop fusion), and employing rewriting techniques known as tiling or blocking all strive to avoid those extra memory fetches.
While there are some rules of thumb for this rewrite exercise, you typically have to carefully consider loop carried data dependencies, to ensure that you don't affect the semantics of the program.
These things are what really pays off in the multicore world, where you typically wont see much of throughput improvements after adding the second thread.
I can't believe there aren't more answers to this. Anyway, one classic example is to iterate a multidimensional array "inside out":
pseudocode
for (i = 0 to size)
for (j = 0 to size)
do something with ary[j][i]
The reason this is cache inefficient is because modern CPUs will load the cache line with "near" memory addresses from main memory when you access a single memory address. We are iterating through the "j" (outer) rows in the array in the inner loop, so for each trip through the inner loop, the cache line will cause to be flushed and loaded with a line of addresses that are near to the [j][i] entry. If this is changed to the equivalent:
for (i = 0 to size)
for (j = 0 to size)
do something with ary[i][j]
It will run much faster.
The basic rules are actually fairly simple. Where it gets tricky is in how they apply to your code.
The cache works on two principles: Temporal locality and spatial locality.
The former is the idea that if you recently used a certain chunk of data, you'll probably need it again soon. The latter means that if you recently used the data at address X, you'll probably soon need address X+1.
The cache tries to accomodate this by remembering the most recently used chunks of data. It operates with cache lines, typically sized 128 byte or so, so even if you only need a single byte, the entire cache line that contains it gets pulled into the cache. So if you need the following byte afterwards, it'll already be in the cache.
And this means that you'll always want your own code to exploit these two forms of locality as much as possible. Don't jump all over memory. Do as much work as you can on one small area, and then move on to the next, and do as much work there as you can.
A simple example is the 2D array traversal that 1800's answer showed. If you traverse it a row at a time, you're reading the memory sequentially. If you do it column-wise, you'll read one entry, then jump to a completely different location (the start of the next row), read one entry, and jump again. And when you finally get back to the first row, it will no longer be in the cache.
The same applies to code. Jumps or branches mean less efficient cache usage (because you're not reading the instructions sequentially, but jumping to a different address). Of course, small if-statements probably won't change anything (you're only skipping a few bytes, so you'll still end up inside the cached region), but function calls typically imply that you're jumping to a completely different address that may not be cached. Unless it was called recently.
Instruction cache usage is usually far less of an issue though. What you usually need to worry about is the data cache.
In a struct or class, all members are laid out contiguously, which is good. In an array, all entries are laid out contiguously as well. In linked lists, each node is allocated at a completely different location, which is bad. Pointers in general tend to point to unrelated addresses, which will probably result in a cache miss if you dereference it.
And if you want to exploit multiple cores, it can get really interesting, as usually, only one CPU may have any given address in its L1 cache at a time. So if both cores constantly access the same address, it will result in constant cache misses, as they're fighting over the address.
I recommend reading the 9-part article What every programmer should know about memory by Ulrich Drepper if you're interested in how memory and software interact. It's also available as a 104-page PDF.
Sections especially relevant to this question might be Part 2 (CPU caches) and Part 5 (What programmers can do - cache optimization).
Apart from data access patterns, a major factor in cache-friendly code is data size. Less data means more of it fits into the cache.
This is mainly a factor with memory-aligned data structures. "Conventional" wisdom says data structures must be aligned at word boundaries because the CPU can only access entire words, and if a word contains more than one value, you have to do extra work (read-modify-write instead of a simple write). But caches can completely invalidate this argument.
Similarly, a Java boolean array uses an entire byte for each value in order to allow operating on individual values directly. You can reduce the data size by a factor of 8 if you use actual bits, but then access to individual values becomes much more complex, requiring bit shift and mask operations (the BitSet class does this for you). However, due to cache effects, this can still be considerably faster than using a boolean[] when the array is large. IIRC I once achieved a speedup by a factor of 2 or 3 this way.
The most effective data structure for a cache is an array. Caches work best, if your data structure is laid out sequentially as CPUs read entire cache lines (usually 32 bytes or more) at once from main memory.
Any algorithm which accesses memory in random order trashes the caches because it always needs new cache lines to accomodate the randomly accessed memory. On the other hand an algorithm, which runs sequentially through an array is best because:
It gives the CPU a chance to read-ahead, e.g. speculatively put more memory into the cache, which will be accessed later. This read-ahead gives a huge performance boost.
Running a tight loop over a large array also allows the CPU to cache the code executing in the loop and in most cases allows you to execute an algorithm entirely from cache memory without having to block for external memory access.
One example I saw used in a game engine was to move data out of objects and into their own arrays. A game object that was subject to physics might have a lot of other data attached to it as well. But during the physics update loop all the engine cared about was data about position, speed, mass, bounding box, etc. So all of that was placed into its own arrays and optimized as much as possible for SSE.
So during the physics loop the physics data was processed in array order using vector math. The game objects used their object ID as the index into the various arrays. It was not a pointer because pointers could become invalidated if the arrays had to be relocated.
In many ways this violated object-oriented design patterns but it made the code a lot faster by placing data close together that needed to be operated on in the same loops.
This example is probably out of date because I expect most modern games use a prebuilt physics engine like Havok.
A remark to the "classic example" by user 1800 INFORMATION (too long for a comment)
I wanted to check the time differences for two iteration orders ( "outter" and "inner"), so I made a simple experiment with a large 2D array:
measure::start();
for ( int y = 0; y < N; ++y )
for ( int x = 0; x < N; ++x )
sum += A[ x + y*N ];
measure::stop();
and the second case with the for loops swapped.
The slower version ("x first") was 0.88sec and the faster one, was 0.06sec. That's the power of caching :)
I used gcc -O2 and still the loops were not optimized out. The comment by Ricardo that "most of the modern compilers can figure this out by itselves" does not hold
Only one post touched on it, but a big issue comes up when sharing data between processes. You want to avoid having multiple processes attempting to modify the same cache line simultaneously. Something to look out for here is "false" sharing, where two adjacent data structures share a cache line and modifications to one invalidates the cache line for the other. This can cause cache lines to unnecessarily move back and forth between processor caches sharing the data on a multiprocessor system. A way to avoid it is to align and pad data structures to put them on different lines.
I can answer (2) by saying that in the C++ world, linked lists can easily kill the CPU cache. Arrays are a better solution where possible. No experience on whether the same applies to other languages, but it's easy to imagine the same issues would arise.
Cache is arranged in "cache lines" and (real) memory is read from and written to in chunks of this size.
Data structures that are contained within a single cache-line are therefore more efficient.
Similarly, algorithms which access contiguous memory blocks will be more efficient than algorithms which jump through memory in a random order.
Unfortunately the cache line size varies dramatically between processors, so there's no way to guarantee that a data structure that's optimal on one processor will be efficient on any other.
To ask how to make a code, cache effective-cache friendly and most of the other questions , is usually to ask how to Optimize a program, that's because the cache has such a huge impact on performances that any optimized program is one that is cache effective-cache friendly.
I suggest reading about Optimization, there are some good answers on this site.
In terms of books, I recommend on Computer Systems: A Programmer's Perspective which has some fine text about the proper usage of the cache.
(b.t.w - as bad as a cache-miss can be, there is worse - if a program is paging from the hard-drive...)
There has been a lot of answers on general advices like data structure selection, access pattern, etc. Here I would like to add another code design pattern called software pipeline that makes use of active cache management.
The idea is borrow from other pipelining techniques, e.g. CPU instruction pipelining.
This type of pattern best applies to procedures that
could be broken down to reasonable multiple sub-steps, S[1], S[2], S[3], ... whose execution time is roughly comparable with RAM access time (~60-70ns).
takes a batch of input and do aforementioned multiple steps on them to get result.
Let's take a simple case where there is only one sub-procedure.
Normally the code would like:
def proc(input):
return sub-step(input))
To have better performance, you might want to pass multiple inputs to the function in a batch so you amortize function call overhead and also increases code cache locality.
def batch_proc(inputs):
results = []
for i in inputs:
// avoids code cache miss, but still suffer data(inputs) miss
results.append(sub-step(i))
return res
However, as said earlier, if the execution of the step is roughly the same as RAM access time you can further improve the code to something like this:
def batch_pipelined_proc(inputs):
for i in range(0, len(inputs)-1):
prefetch(inputs[i+1])
# work on current item while [i+1] is flying back from RAM
results.append(sub-step(inputs[i-1]))
results.append(sub-step(inputs[-1]))
The execution flow would look like:
prefetch(1) ask CPU to prefetch input[1] into cache, where prefetch instruction takes P cycles itself and return, and in the background input[1] would arrive in cache after R cycles.
works_on(0) cold miss on 0 and works on it, which takes M
prefetch(2) issue another fetch
works_on(1) if P + R <= M, then inputs[1] should be in the cache already before this step, thus avoid a data cache miss
works_on(2) ...
There could be more steps involved, then you can design a multi-stage pipeline as long as the timing of the steps and memory access latency matches, you would suffer little code/data cache miss. However, this process needs to be tuned with many experiments to find out right grouping of steps and prefetch time. Due to its required effort, it sees more adoption in high performance data/packet stream processing. A good production code example could be found in DPDK QoS Enqueue pipeline design:
http://dpdk.org/doc/guides/prog_guide/qos_framework.html Chapter 21.2.4.3. Enqueue Pipeline.
More information could be found:
https://software.intel.com/en-us/articles/memory-management-for-optimal-performance-on-intel-xeon-phi-coprocessor-alignment-and
http://infolab.stanford.edu/~ullman/dragon/w06/lectures/cs243-lec13-wei.pdf
Besides aligning your structure and fields, if your structure if heap allocated you may want to use allocators that support aligned allocations; like _aligned_malloc(sizeof(DATA), SYSTEM_CACHE_LINE_SIZE); otherwise you may have random false sharing; remember that in Windows, the default heap has a 16 bytes alignment.
Write your program to take a minimal size. That is why it is not always a good idea to use -O3 optimisations for GCC. It takes up a larger size. Often, -Os is just as good as -O2. It all depends on the processor used though. YMMV.
Work with small chunks of data at a time. That is why a less efficient sorting algorithms can run faster than quicksort if the data set is large. Find ways to break up your larger data sets into smaller ones. Others have suggested this.
In order to help you better exploit instruction temporal/spatial locality, you may want to study how your code gets converted in to assembly. For example:
for(i = 0; i < MAX; ++i)
for(i = MAX; i > 0; --i)
The two loops produce different codes even though they are merely parsing through an array. In any case, your question is very architecture specific. So, your only way to tightly control cache use is by understanding how the hardware works and optimising your code for it.