Related
I am trying to find a situation where changing multithreading to coroutines will speed up processing of the affected code section. As far as I discovered that coroutines use less CPU and Heap space comparing to threads, I still can't find the case where coroutines are faster than threads. Although I know that coroutines creation and context-switching are much cheaper than corresponding operations with threads, I've got imperceptible results in speed difference (without measuring thread creation both cases will be absolutely the same).
So, is it even possible to find a case where coroutines will faster an execution more than threads?
One thing to note is that coroutines are vastly superior when you have lots and lots of them. You can create and execute thousands of coroutines without a second thought, if you attempted to do that via threads all the overhead associated with threads might quickly kill the host. So, this enables you to think about massive parallelization without having to manage worker threads and runnables. They also make it easy to implement asynchronous computation patterns which would be very unwieldy to implement with basic threads, like channels and actors.
Out of scope regarding your question, but still noteworthy is the genericity of the concept, as the use cases for coroutines are not just limited to asynchronous computation. The core of coroutines are suspendable functions, which for example also enables generators like you have in python, which you would not immediately connect to asynchronous programming.
I've been reading (and working) quite a bit with massively multi-threaded applications, and with IO, and I've found that the term asynchronous has become some sort of catch-all for multiple vague ideas. I'm wondering if I understand it correctly. The way I see it is that there are two main branches of "asynchronicity".
Asynchronous I/O. Such as network read/write. What this really boils down to is efficient parallel processing between multiple CPUs, such as your main CPU and your NIC CPU. The idea is to have multiple processors running in parallel, exchanging data, without blocking waiting for the other to finish and return the results of it's job.
Minimizing context-switching penalties by minimizing use of threads. This seems to be what the .NET framework is focusing on with it's async/await features. Instead of spawning/closing/blocking threads, break parallel jobs into tasks, and use a software task scheduler to keep a pool of threads as busy as possible without resorting to spawning new threads.
These seem like two entirely separate concepts with no similarities that could tie them together, but are both referred to by the same "asynchronous computing" vocabulary.
Am I understanding all of this correctly?
Asynchronous basically means not blocking, i.e. not having to wait for an operation to complete.
Threads are just one way of accomplishing that. There are many ways of doing this, from hardware level, SO level, software level.
Someone with more experience than me can give examples of asyncronicity not related to threads.
What this really boils down to is efficient parallel processing between multiple CPUs, such as your main CPU and your NIC CPU. The idea is to have multiple processors running in parallel...
Asynchronous programming is not all about multi-core CPU's and parallelism: consider a single core CPU, with just one thread creating email messages and sends them. In a synchronous fashion, it would spend a few micro seconds to create the message, and a lot more time to send it through network, and only then create the next message. But in asynchronous program, the thread could create a new message while the previous one is being sent through the network. One implementation for that kind of program can be using .NET async/await feature, where you can have just one thread. But even a blocking IO program could be considered asynchronous: If the main thread creates the messages and queues them in a buffer, which another thread pulls them from and sends them in a blocking IO way. From the main thread's point of view - it's completely async.
.NET async/await just uses the OS api's which are already async - reading /writing a file, send /receive data through network, they are all async anyway - the OS doesn't block on them (the drivers themselves are async).
Asynchronous is a general term, which does not have widely accepted meaning. Different domains have different meanings to it.
For instance, async IO means that instead of blocking on IO call, something else happens. Something else can be really different things, but it usually involves some sort of notification of call completion. Details might differ. For instance, a notification might be built into the call itself - like in MS Completeion Ports (if memory serves). Or, it can be something verify do before you make a call so that the call can not block - this is what poll() and friends do.
Async might also well mean simply parallel execution. For instance, one might say that 'database is updated asynchronously' meaning that there is a dedicated thread which handles database connectivity, and that thread does not slow down the main processing thread.
I don't quite understand the difference between threads and lightweight threads. From an API perspective both types of threads are identical so where exactly does the difference come in. Is it at the implementation level where a lightweight thread is managed by a higher level runtime than the OS thread scheduler or is it something else? Also, is there set of heuristics that people use to decide which type of thread to use in specific scenarios?
In what context, lightweight threads could represent threads which are implemented by a library, for example threads can be simulated in a library by switching between lightweight threads at an event handling layer, these lightweight threads are queued up and processed by a singe OS thread, the advantage of this is that since context switching is handled in the library switching can occur when the processing of data is complete and so the data does not need to be loaded back into the CPU's cache next time this lightweight thread becomes active.
Lightweight threads could also refer to co-operative threads (or fibers), these are threads where you have to explicitly yield to give other lightweight threads a chance, this has the same advantage in that the context switching can occur at a place you know you have finished processing some data and so you know it will not be need again.
Alternativly Lightweight threads could mean normal OS threads and the non-lightweight threads could mean processes, process have at least one thread within them and also have there own memory and other resources, they are more expensive than threads because you can not share data between thread easily and it can be a more expensive operation for the OS to create processes.
I searched the web on some technical details about blocking I/O and non blocking I/O and I found several people stating that non-blocking I/O would be faster than blocking I/O. For example in this document.
If I use blocking I/O, then of course the thread that is currently blocked can't do anything else... Because it's blocked. But as soon as a thread starts being blocked, the OS can switch to another thread and not switch back until there is something to do for the blocked thread. So as long as there is another thread on the system that needs CPU and is not blocked, there should not be any more CPU idle time compared to an event based non-blocking approach, is there?
Besides reducing the time the CPU is idle I see one more option to increase the number of tasks a computer can perform in a given time frame: Reduce the overhead introduced by switching threads. But how can this be done? And is the overhead large enough to show measurable effects? Here is an idea on how I can picture it working:
To load the contents of a file, an application delegates this task to an event-based i/o framework, passing a callback function along with a filename
The event framework delegates to the operating system, which programs a DMA controller of the hard disk to write the file directly to memory
The event framework allows further code to run.
Upon completion of the disk-to-memory copy, the DMA controller causes an interrupt.
The operating system's interrupt handler notifies the event-based i/o framework about the file being completely loaded into memory. How does it do that? Using a signal??
The code that is currently run within the event i/o framework finishes.
The event-based i/o framework checks its queue and sees the operating system's message from step 5 and executes the callback it got in step 1.
Is that how it works? If it does not, how does it work? That means that the event system can work without ever having the need to explicitly touch the stack (such as a real scheduler that would need to backup the stack and copy the stack of another thread into memory while switching threads)? How much time does this actually save? Is there more to it?
The biggest advantage of nonblocking or asynchronous I/O is that your thread can continue its work in parallel. Of course you can achieve this also using an additional thread. As you stated for best overall (system) performance I guess it would be better to use asynchronous I/O and not multiple threads (so reducing thread switching).
Let's look at possible implementations of a network server program that shall handle 1000 clients connected in parallel:
One thread per connection (can be blocking I/O, but can also be non-blocking I/O).
Each thread requires memory resources (also kernel memory!), that is a disadvantage. And every additional thread means more work for the scheduler.
One thread for all connections.
This takes load from the system because we have fewer threads. But it also prevents you from using the full performance of your machine, because you might end up driving one processor to 100% and letting all other processors idle around.
A few threads where each thread handles some of the connections.
This takes load from the system because there are fewer threads. And it can use all available processors. On Windows this approach is supported by Thread Pool API.
Of course having more threads is not per se a problem. As you might have recognized I chose quite a high number of connections/threads. I doubt that you'll see any difference between the three possible implementations if we are talking about only a dozen threads (this is also what Raymond Chen suggests on the MSDN blog post Does Windows have a limit of 2000 threads per process?).
On Windows using unbuffered file I/O means that writes must be of a size which is a multiple of the page size. I have not tested it, but it sounds like this could also affect write performance positively for buffered synchronous and asynchronous writes.
The steps 1 to 7 you describe give a good idea of how it works. On Windows the operating system will inform you about completion of an asynchronous I/O (WriteFile with OVERLAPPED structure) using an event or a callback. Callback functions will only be called for example when your code calls WaitForMultipleObjectsEx with bAlertable set to true.
Some more reading on the web:
Multiple Threads in the User Interface on MSDN, also shortly handling the cost of creating threads
Section Threads and Thread Pools says "Although threads are relatively easy to create and use, the operating system allocates a significant amount of time and other resources to manage them."
CreateThread documentation on MSDN says "However, your application will have better performance if you create one thread per processor and build queues of requests for which the application maintains the context information.".
Old article Why Too Many Threads Hurts Performance, and What to do About It
I/O includes multiple kind of operations like reading and writing data from hard drives, accessing network resources, calling web services or retrieving data from databases. Depending on the platform and on the kind of operation, asynchronous I/O will usually take advantage of any hardware or low level system support for performing the operation. This means that it will be performed with as little impact as possible on the CPU.
At application level, asynchronous I/O prevents threads from having to wait for I/O operations to complete. As soon as an asynchronous I/O operation is started, it releases the thread on which it was launched and a callback is registered. When the operation completes, the callback is queued for execution on the first available thread.
If the I/O operation is executed synchronously, it keeps its running thread doing nothing until the operation completes. The runtime doesn't know when the I/O operation completes, so it will periodically provide some CPU time to the waiting thread, CPU time that could have otherwise be used by other threads that have actual CPU bound operations to perform.
So, as #user1629468 mentioned, asynchronous I/O does not provide better performance but rather better scalability. This is obvious when running in contexts that have a limited number of threads available, like it is the case with web applications. Web application usually use a thread pool from which they assign threads to each request. If requests are blocked on long running I/O operations there is the risk of depleting the web pool and making the web application freeze or slow to respond.
One thing I have noticed is that asynchronous I/O isn't the best option when dealing with very fast I/O operations. In that case the benefit of not keeping a thread busy while waiting for the I/O operation to complete is not very important and the fact that the operation is started on one thread and it is completed on another adds an overhead to the overall execution.
You can read a more detailed research I have recently made on the topic of asynchronous I/O vs. multithreading here.
To presume a speed improvement due to any form of multi-computing you must presume either that multiple CPU-based tasks are being executed concurrently upon multiple computing resources (generally processor cores) or else that not all of the tasks rely upon the concurrent usage of the same resource -- that is, some tasks may depend on one system subcomponent (disk storage, say) while some tasks depend on another (receiving communication from a peripheral device) and still others may require usage of processor cores.
The first scenario is often referred to as "parallel" programming. The second scenario is often referred to as "concurrent" or "asynchronous" programming, although "concurrent" is sometimes also used to refer to the case of merely allowing an operating system to interleave execution of multiple tasks, regardless of whether such execution must take place serially or if multiple resources can be used to achieve parallel execution. In this latter case, "concurrent" generally refers to the way that execution is written in the program, rather than from the perspective of the actual simultaneity of task execution.
It's very easy to speak about all of this with tacit assumptions. For example, some are quick to make a claim such as "Asynchronous I/O will be faster than multi-threaded I/O." This claim is dubious for several reasons. First, it could be the case that some given asynchronous I/O framework is implemented precisely with multi-threading, in which case they are one in the same and it doesn't make sense to say one concept "is faster than" the other.
Second, even in the case when there is a single-threaded implementation of an asynchronous framework (such as a single-threaded event loop) you must still make an assumption about what that loop is doing. For example, one silly thing you can do with a single-threaded event loop is request for it to asynchronously complete two different purely CPU-bound tasks. If you did this on a machine with only an idealized single processor core (ignoring modern hardware optimizations) then performing this task "asynchronously" wouldn't really perform any differently than performing it with two independently managed threads, or with just one lone process -- the difference might come down to thread context switching or operating system schedule optimizations, but if both tasks are going to the CPU it would be similar in either case.
It is useful to imagine a lot of the unusual or stupid corner cases you might run into.
"Asynchronous" does not have to be concurrent, for example just as above: you "asynchronously" execute two CPU-bound tasks on a machine with exactly one processor core.
Multi-threaded execution doesn't have to be concurrent: you spawn two threads on a machine with a single processor core, or ask two threads to acquire any other kind of scarce resource (imagine, say, a network database that can only establish one connection at a time). The threads' execution might be interleaved however the operating system scheduler sees fit, but their total runtime cannot be reduced (and will be increased from the thread context switching) on a single core (or more generally, if you spawn more threads than there are cores to run them, or have more threads asking for a resource than what the resource can sustain). This same thing goes for multi-processing as well.
So neither asynchronous I/O nor multi-threading have to offer any performance gain in terms of run time. They can even slow things down.
If you define a specific use case, however, like a specific program that both makes a network call to retrieve data from a network-connected resource like a remote database and also does some local CPU-bound computation, then you can start to reason about the performance differences between the two methods given a particular assumption about hardware.
The questions to ask: How many computational steps do I need to perform and how many independent systems of resources are there to perform them? Are there subsets of the computational steps that require usage of independent system subcomponents and can benefit from doing so concurrently? How many processor cores do I have and what is the overhead for using multiple processors or threads to complete tasks on separate cores?
If your tasks largely rely on independent subsystems, then an asynchronous solution might be good. If the number of threads needed to handle it would be large, such that context switching became non-trivial for the operating system, then a single-threaded asynchronous solution might be better.
Whenever the tasks are bound by the same resource (e.g. multiple needs to concurrently access the same network or local resource), then multi-threading will probably introduce unsatisfactory overhead, and while single-threaded asynchrony may introduce less overhead, in such a resource-limited situation it too cannot produce a speed-up. In such a case, the only option (if you want a speed-up) is to make multiple copies of that resource available (e.g. multiple processor cores if the scarce resource is CPU; a better database that supports more concurrent connections if the scarce resource is a connection-limited database, etc.).
Another way to put it is: allowing the operating system to interleave the usage of a single resource for two tasks cannot be faster than merely letting one task use the resource while the other waits, then letting the second task finish serially. Further, the scheduler cost of interleaving means in any real situation it actually creates a slowdown. It doesn't matter if the interleaved usage occurs of the CPU, a network resource, a memory resource, a peripheral device, or any other system resource.
The main reason to use AIO is for scalability. When viewed in the context of a few threads, the benefits are not obvious. But when the system scales to 1000s of threads, AIO will offer much better performance. The caveat is that AIO library should not introduce further bottlenecks.
One possible implementation of non-blocking I/O is exactly what you said, with a pool of background threads that do blocking I/O and notify the thread of the originator of the I/O via some callback mechanism. In fact, this is how the AIO module in glibc works. Here are some vague details about the implementation.
While this is a good solution that is quite portable (as long as you have threads), the OS is typically able to service non-blocking I/O more efficiently. This Wikipedia article lists possible implementations besides the thread pool.
I am currently in the process of implementing async io on an embedded platform using protothreads. Non blocking io makes the difference between running at 16000fps and 160fps. The biggest benefit of non blocking io is that you can structure your code to do other things while hardware does its thing. Even initialization of devices can be done in parallel.
Martin
In Node, multiple threads are being launched, but it's a layer down in the C++ run-time.
"So Yes NodeJS is single threaded, but this is a half truth, actually it is event-driven and single-threaded with background workers. The main event loop is single-threaded but most of the I/O works run on separate threads, because the I/O APIs in Node.js are asynchronous/non-blocking by design, in order to accommodate the event loop. "
https://codeburst.io/how-node-js-single-thread-mechanism-work-understanding-event-loop-in-nodejs-230f7440b0ea
"Node.js is non-blocking which means that all functions ( callbacks ) are delegated to the event loop and they are ( or can be ) executed by different threads. That is handled by Node.js run-time."
https://itnext.io/multi-threading-and-multi-process-in-node-js-ffa5bb5cde98
The "Node is faster because it's non-blocking..." explanation is a bit of marketing and this is a great question. It's efficient and scaleable, but not exactly single threaded.
The improvement as far as I know is that Asynchronous I/O uses ( I'm talking about MS System, just to clarify ) the so called I/O completion ports. By using the Asynchronous call the framework leverage such architecture automatically, and this is supposed to be much more efficient that standard threading mechanism. As a personal experience I can say that you would sensibly feel your application more reactive if you prefer AsyncCalls instead of blocking threads.
Let me give you a counterexample that asynchronous I/O does not work.
I am writing a proxy similar to below-using boost::asio.
https://github.com/ArashPartow/proxy/blob/master/tcpproxy_server.cpp
However, the scenario of my case is, incoming (from clients side) messages are fast while outgoing (to server side) is slow for one session, to keep up with the incoming speed or to maximize the total proxy throughput, we have to use multiple sessions under one connection.
Thus this async I/O framework does not work anymore. We do need a thread pool to send to the server by assigning each thread a session.
I'm reading up on concurrency. I've got a bit over my head with terms that have confusingly similar definitions. Namely:
Processes
Threads
"Green threads"
Protothreads
Fibers
Coroutines
"Goroutines" in the Go language
My impression is that the distinctions rest on (1) whether truly parallel or multiplexed; (2) whether managed at the CPU, at the OS, or in the program; and (3..5) a few other things I can't identify.
Is there a succinct and unambiguous guide to the differences between these approaches to parallelism?
OK, I'm going to do my best. There are caveats everywhere, but I'm going to do my best to give my understanding of these terms and references to something that approximates the definition I've given.
Process: OS-managed (possibly) truly concurrent, at least in the presence of suitable hardware support. Exist within their own address space.
Thread: OS-managed, within the same address space as the parent and all its other threads. Possibly truly concurrent, and multi-tasking is pre-emptive.
Green Thread: These are user-space projections of the same concept as threads, but are not OS-managed. Probably not truly concurrent, except in the sense that there may be multiple worker threads or processes giving them CPU time concurrently, so probably best to consider this as interleaved or multiplexed.
Protothreads: I couldn't really tease a definition out of these. I think they are interleaved and program-managed, but don't take my word for it. My sense was that they are essentially an application-specific implementation of the same kind of "green threads" model, with appropriate modification for the application domain.
Fibers: OS-managed. Exactly threads, except co-operatively multitasking, and hence not truly concurrent.
Coroutines: Exactly fibers, except not OS-managed.
Goroutines: They claim to be unlike anything else, but they seem to be exactly green threads, as in, process-managed in a single address space and multiplexed onto system threads. Perhaps somebody with more knowledge of Go can cut through the marketing material.
It's also worth noting that there are other understandings in concurrency theory of the term "process", in the process calculus sense. This definition is orthogonal to those above, but I just thought it worth mentioning so that no confusion arises should you see process used in that sense somewhere.
Also, be aware of the difference between parallel and concurrent. It's possible you were using the former in your question where I think you meant the latter.
I mostly agree with Gian's answer, but I have different interpretations of a few concurrency primitives. Note that these terms are often used inconsistently by different authors. These are my favorite definitions (hopefully not too far from the modern consensus).
Process:
OS-managed
Each has its own virtual address space
Can be interrupted (preempted) by the system to allow another process to run
Can run in parallel with other processes on different processors
The memory overhead of processes is high (includes virtual memory tables, open file handles, etc)
The time overhead for creating and context switching between processes is relatively high
Threads:
OS-managed
Each is "contained" within some particular process
All threads in the same process share the same virtual address space
Can be interrupted by the system to allow another thread to run
Can run in parallel with other threads on different processors
The memory and time overheads associated with threads are smaller than processes, but still non-trivial
(For example, typically context switching involves entering the kernel and invoking the system scheduler.)
Cooperative Threads:
May or may not be OS-managed
Each is "contained" within some particular process
In some implementations, each is "contained" within some particular OS thread
Cannot be interrupted by the system to allow a cooperative peer to run
(The containing process/thread can still be interrupted, of course)
Must invoke a special yield primitive to allow peer cooperative threads to run
Generally cannot be run in parallel with cooperative peers
(Though some people think it's possible: http://ocm.dreamhosters.com/.)
There are lots of variations on the cooperative thread theme that go by different names:
Fibers
Green threads
Protothreads
User-level threads (user-level threads can be interruptable/preemptive, but that's a relatively unusual combination)
Some implementations of cooperative threads use techniques like split/segmented stacks or even individually heap-allocating every call frame to reduce the memory overhead associated with pre-allocating a large chunk of memory for the stack
Depending on the implementation, calling a blocking syscall (like reading from the network or sleeping) will either cause a whole group of cooperative threads to block or implicitly cause the calling thread to yield
Coroutines:
Some people use "coroutine" and "cooperative thread" more or less synonymously
I do not prefer this usage
Some coroutine implementations are actually "shallow" cooperative threads; yield can only be invoked by the "coroutine entry procedure"
The shallow (or semi-coroutine) version is easier to implement than threads, because each coroutine does not need a complete stack (just one frame for the entry procedure)
Often coroutine frameworks have yield primitives that require the invoker to explicitly state which coroutine control should transfer to
Generators:
Restricted (shallow) coroutines
yield can only return control back to whichever code invoked the generator
Goroutines:
An odd hybrid of cooperative and OS threads
Cannot be interrupted (like cooperative threads)
Can run in parallel on a language runtime-managed pool of OS threads
Event handlers:
Procedures/methods that are invoked by an event dispatcher in response to some action happening
Very popular for user interface programming
Require little to no language/system support; can be implemented in a library
At most one event handler can be running at a time; the dispatcher must wait for a handler to finish (return) before starting the next
Makes synchronization relatively simple; different handler executions never overlap in time
Implementing complex tasks with event handlers tends to lead to "inverted control flow"/"stack ripping"
Tasks:
Units of work that are doled out by a manager to a pool of workers
The workers can be threads, processes or machines
Of course the kind of worker a task library uses has a significant impact on how one implements the tasks
In this list of inconsistently and confusingly used terminology, "task" takes the crown. Particularly in the embedded systems community, "task" is sometimes used to mean "process", "thread" or "event handler" (usually called an "interrupt service routine"). It is also sometimes used generically/informally to refer to any kind of unit of computation.
One pet peeve that I can't stop myself from airing: I dislike the use of the phrase "true concurrency" for "processor parallelism". It's quite common, but I think it leads to much confusion.
For most applications, I think task-based frameworks are best for parallelization. Most of the popular ones (Intel's TBB, Apple's GCD, Microsoft's TPL & PPL) use threads as workers. I wish there were some good alternatives that used processes, but I'm not aware of any.
If you're interested in concurrency (as opposed to processor parallelism), event handlers are the safest way to go. Cooperative threads are an interesting alternative, but a bit of a wild west. Please do not use threads for concurrency if you care about the reliability and robustness of your software.
Protothreads are just a switch case implementation that acts like a state machine but makes implementation of the software a whole lot simpler. It is based around idea of saving a and int value before a case label and returning and then getting back to the point after the case by reading back that variable and using switch to figure out where to continue. So protothread are a sequential implementation of a state machine.
Protothreads are great when implementing sequential state machines. Protothreads are not really threads at all, but rather a syntax abstraction that makes it much easier to write a switch/case state machine that has to switch states sequentially (from one to the next etc..).
I have used protothreads to implement asynchronous io: http://martinschroder.se/asynchronous-io-using-protothreads/