I am looking for a cross-platform C++ master/worker library or work queue library. The general idea is that my application would create some sort of Task or Work objects, pass them to the work master or work queue, which would in turn execute the work in separate threads or processes. To provide a bit of context, the application is a CD ripper, and the the tasks that I want to parallelize are things like "rip track", "encode WAV to Mp3", etc.
My basic requirements are:
Must support a configurable number of concurrent tasks.
Must support dependencies between tasks, such that tasks are not executed until all tasks that they depend on have completed.
Must allow for cancellation of tasks (or at least not prevent me from coding cancellation into my own tasks).
Must allow for reporting of status and progress information back to the main application thread.
Must work on Windows, Mac OS X, and Linux
Must be open source.
It would be especially nice if this library also:
Integrated with Qt's signal/slot mechanism.
Supported the use of threads or processes for executing tasks.
By way of analogy, I'm looking for something similar to Java's ExecutorService or some other similar thread pooling library, but in cross-platform C++. Does anyone know of such a beast?
Thanks!
I haven't used it in long enough that I'm not positive whether it exactly meets your needs, but check out the Adaptive Communications Environment (ACE). This library allows you to construct "active objects" which have work queues and execute their main body in their own threads, as well as thread pools that can be shared amoung objects. Then you can pass queue work objects on to active objects for them to process. Objects can be chained in various ways. The library is fairly heavy and has a lot to it to learn, but there have been a couple of books written about it and theres a fair amount of tutorial information available online as well. It should be able to do everything you want plus more, my only concern is whether it possesses the interfaces you are looking for 'out of the box' or if you'd need to build on top of it to get exactly what you are looking for.
I think this calls for intel's Threading Building Blocks, which pretty much does what you want.
Check out Intels' Thread Building Blocks library.
Sounds like you require some kind of "Time Sharing System".
There are some good open source ones out there, but I don't know
if they have built-in QT slot support.
This is probably a huge overkill for what you need but still worth mentioning -
BOINC is a distributed framework for such tasks. There's a main server that gives out tasks to perform and a cloud of workers that do its bidding. It is the framework behind projects like SETI#Home and many others.
See this post for creating threads using the boost library in C++:
Simple example of threading in C++
(it is a c++ thread even though the title says c)
basically, create your own "master" object that takes a "runnable" object and starts it running in a new thread.
Then you can create new classes that implement "runnable" and throw them over to your master runner any old time you want.
Related
My general idea is that a single-threaded application ( the Lua interpreter ) will always deteriorate the performance of a multi-threaded application that depends on it ( a generic C++ application ).
To circumvent this problem I'm thinking about an asynchronous approach on the interpreter while keeping the C++ application multi-threaded, this basically means that based on my approach a Lua interpreter should somehow push the entire script/file in a scheduler with an asynchronous approach ( without waiting for the result ) and it's up to the well designed C++ multi-threaded messaging system to keep everything sequential.
The usual relationship is C/C++ function <-> Lua ( with a sequential approach ) ; I would like to have something like C++ messaging system <-> entire Lua script .
I'm also open to any kind of approach that can solve this and really help the mix between Lua and a C++ application designed for multi-threading.
Is this approach made possible by some piece of software ?
EDIT
I need something "user-proof" and I need to implement this behaviour right in C++/Lua API design.
One option is to implement communication to lua as a co-routine. Messages are sent to C++ via coroutine.yield(messagedata), and then it sends back results via lua_resume. (See also: lua_newthread). You could even wrap your functions to provide a nicer event UI.
function doThing(thing, other, data)
return coroutine.yield("doThing", thing, other, data)
end
You can still only have one thread running the lua interpreter at any given time (you will have to do locking) but you can have multiple such co-routines running concurrently.
Concurrency in Lua is a topic that has many many solutions. Here is a resource:
http://lua-users.org/wiki/MultiTasking
You actually can make it easy for yourself since you do not actually have to run Lua itself multithreaded, which would impose a number of additional issues.
The obvious solution is running Lua in a separate thread but providing only a thin API for Lua in which every single API call immediately either forks a new thread/process or uses some sort of message passing for asynchronous data transfer, or even uses short-duration semaphores to read/write some values. This solution requires some sort of idle loop or event listeners unless you want to do busy waiting...
Another option that I think is still quite easy to implement with a new API, is actually the approach of node.js:
Run Lua in a separate thread
Make your whole API of functions that only take callbacks. These callbacks are queued and can be scheduled by your C++ application.
You can even, but do not have to, provide callback wrappers for the standard Lua API.
Example:
local version;
Application.requestVersionNumber(function(val) version = val; end)
Of course this example is riduculously trivial, but you get the idea.
One thing you should know though is that with the callback approach the scripts quickly get highly tiered if you are not careful. While that's not bad for performance, they can get hard to read.
I want to write a simple multiplayer game as part of my C++ learning project.
So I thought, since I am at it, I would like to do it properly, as opposed to just getting-it-done.
If I understood correctly: Apache uses a Thread-per-connection architecture, while nginx uses an event-loop and then dedicates a worker [x] for the incoming connection. I guess nginx is wiser, since it supports a higher concurrency level. Right?
I have also come across this clever analogy, but I am not sure if it could be applied to my situation. The analogy also seems to be very idealist. I have rarely seen my computer run at 100% CPU (even with a umptillion Chrome tabs open, Photoshop and what-not running simultaneously)
Also, I have come across a SO post (somehow it vanished from my history) where a user asked how many threads they should use, and one of the answers was that it's perfectly acceptable to have around 700, even up to 10,000 threads. This question was related to JVM, though.
So, let's estimate a fictional user-base of around 5,000 users. Which approach should would be the "most concurrent" one?
A reactor pattern running everything in a single thread.
A reactor pattern with a thread-pool (approximately, how big do you suggest the thread pool should be?
Creating a thread per connection and then destroying the thread the connection closes.
I admit option 2 sounds like the best solution to me, but I am very green in all of this, so I might be a bit naive and missing some obvious flaw. Also, it sounds like it could be fairly difficult to implement.
PS: I am considering using POCO C++ Libraries. Suggesting any alternative libraries (like boost) is fine with me. However, many say POCO's library is very clean and easy to understand. So, I would preferably use that one, so I can learn about the hows of what I'm using.
Reactive Applications certainly scale better, when they are written correctly. This means
Never blocking in a reactive thread:
Any blocking will seriously degrade the performance of you server, you typically use a small number of reactive threads, so blocking can also quickly cause deadlock.
No mutexs since these can block, so no shared mutable state. If you require shared state you will have to wrap it with an actor or similar so only one thread has access to the state.
All work in the reactive threads should be cpu bound
All IO has to be asynchronous or be performed in a different thread pool and the results feed back into the reactor.
This means using either futures or callbacks to process replies, this style of code can quickly become unmaintainable if you are not used to it and disciplined.
All work in the reactive threads should be small
To maintain responsiveness of the server all tasks in the reactor must be small (bounded by time)
On an 8 core machine you cannot cannot allow 8 long tasks arrive at the same time because no other work will start until they are complete
If a tasks could take a long time it must be broken up (cooperative multitasking)
Tasks in reactive applications are scheduled by the application not the operating system, that is why they can be faster and use less memory. When you write a Reactive application you are saying that you know the problem domain so well that you can organise and schedule this type of work better than the operating system can schedule threads doing the same work in a blocking fashion.
I am a big fan of reactive architectures but they come with costs. I am not sure I would write my first c++ application as reactive, I normally try to learn one thing at a time.
If you decide to use a reactive architecture use a good framework that will help you design and structure your code or you will end up with spaghetti. Things to look for are:
What is the unit of work?
How easy is it to add new work? can it only come in from an external event (eg network request)
How easy is it to break work up into smaller chunks?
How easy is it to process the results of this work?
How easy is it to move blocking code to another thread pool and still process the results?
I cannot recommend a C++ library for this, I now do my server development in Scala and Akka which provide all of this with an excellent composable futures library to keep the code clean.
Best of luck learning C++ and with which ever choice you make.
Option 2 will most efficiently occupy your hardware. Here is the classic article, ten years old but still good.
http://www.kegel.com/c10k.html
The best library combination these days for structuring an application with concurrency and asynchronous waiting is Boost Thread plus Boost ASIO. You could also try a C++11 std thread library, and std mutex (but Boost ASIO is better than mutexes in a lot of cases, just always callback to the same thread and you don't need protected regions). Stay away from std future, cause it's broken:
http://bartoszmilewski.com/2009/03/03/broken-promises-c0x-futures/
The optimal number of threads in the thread pool is one thread per CPU core. 8 cores -> 8 threads. Plus maybe a few extra, if you think it's possible that your threadpool threads might call blocking operations sometimes.
FWIW, Poco supports option 2 (ParallelReactor) since version 1.5.1
I think that option 2 is the best one. As for tuning of the pool size, I think the pool should be adaptive. It should be able to spawn more threads (with some high hard limit) and remove excessive threads in times of low activity.
as the analogy you linked to (and it's comments) suggest. this is somewhat application dependent. now what you are building here is a game server. let's analyze that.
game servers (generally) do a lot of I/O and relatively few calculations, so they are far from 100% CPU applications.
on the other hand they also usually change values in some database (a "game world" model). all players create reads and writes to this database. which is exactly the intersection problem in the analogy.
so while you may gain some from handling the I/O in separate threads, you will also lose from having separate threads accessing the same database and waiting for its locks.
so either option 1 or 2 are acceptable in your situation. for scalability reasons I would not recommend option 3.
I'm working on a game server, written in C++, and I'm trying to decide how many threads to use and what tasks to thread. The basic server skeleton consists of keyboard I/O and output to a console, accepting incoming connects, sending outgoing connects, and doing the game "stuff".
What I'd like to know is which things should be given a separate thread. Should each connect have its own thread? I know this is variable, it depends on the project or so, but I would like it to support a pretty decent number of players (somewhere in the hundreds if possible).
The standard answer should always be: Try it the simplest way first, and only look for ways to improve performance if the simple way isn't good enough. However, re-architecting a large C++ program can be a painful experience, so some guesses about performance in advance may be appropriate.
Theoretically, hundreds of threads are probably OK on modern machines. The NPTL implementation for Linux was tested with tens of thousands of threads, as I recall. If that's the easiest way for you to implement, it may be the right answer.
However, high-performance web servers and similar typically use event-driven models instead. Consider a library like libevent. I'm sure there are C++ libraries for the same purpose.
I personally believe that languages without first-class continuations, or at least coroutines, are poor choices for this kind of work, but the C language family is how we get work done today, so off we go. :-)
A good solution could be to use a Thread pool.
Idea is to let the main thread dispatch equitably all connexions in a fixed number of threads.
With a good design, you can easily set the number of thread on runtime.
You can find more informations here.
Create more threads than you have CPU cores is not productive, and adding too threads decrease performances due to time taken for switching between threads.
By example, for compiling a large project (it's not exactly the same thing, but it's valid for both case), it's often recommended to use no more thread than number of CPU cores + 1.
A very common technique is to have the game server run on one thread to monitor several connections (i.e. sockets) by using a select on each socket. When data is available, grab the data and enqueue it in a producer/consumer type model for the game engine to pick up.
This is by no means the be-all-end-all implementation, but it should be enough to get you started. Sounds like a cool project. Good luck!
If you setup the connections and utilize them in a manner that cause the thread to block waiting on IO then you should be able to service all of the connections and the keyboard on one thread. You may not want to put the console output on that same thread, as I've seen cases (on windows at least), where the speed of writing to the console is actually a bottleneck (i.e. if the console window is minimized the process runs considerably faster).
If the work of your game engine parallelizes well then you probably want to set use as many threads as there are CPUs less one (for the OS and the other two threads). If you expect the client to run on the same machine the server will want to detect that and scale back the number of threads it uses.
I am building a CD ripper application in C++ and Qt. I would like to parallelize the application such that multiple tracks can be encoded concurrently. Therefore, I have structured the application in such a way that encoding a track is a "Task", and I'm working on a mechanism to run some number of these Tasks concurrently. I could, of course, accomplish this using threads and write my own Task queue or work manager, but I thought Intel's Threading Building Blocks (TBB) might be a better tool for the job. I have a couple of questions, however.
Is encoding a WAV file into a FLAC, Ogg Vorbis, or Mp3 file something that would work well as a tbb::task? The tutorial document states that "if threads block frequently, there is a performance loss when using the task scheduler". I don't think my encoding tasks would block for mutexes frequently, but the will need to access disk relatively frequently, since they must read the WAV data from disk in order to encode. Is this level of disk activity problematic in the sense described by the tutorial?
Does TBB work well with Qt? When using Qt threads, you can use Qt's signals/slots mechanism transparently across threads. Would the same be true if I were using tbb::tasks instead of Qt threads? Would there be any other "gotchas"?
Thanks for any insights you can provide.
Why not using Qt Concurrent ?
TBB is suppose to work well, even transparently, with other threading mechanisms so theoretically there should be nothing preventing you from using QT's thread classes in the same program. If there is something that works more naturally with QT threads, like the GUI, use them and keep the TBB stuff segregated as best you can or want.
I don't see that you are making the best use of TBB as you currently outlined your design. You parallelize at the grossest level, the file. As you suspect, since the CD is a pretty slow device, you may spend more time seeking back and forth for data from multiple files than you actually save.
The real bang for the buck with TBB should involve exploiting whatever data and/or task parallelism there is in the transformation process. Can you, for instance, pull any block of bytes out of the stream and apply whatever transform to it independently of any part of the stream before or after? Are there multiple steps to the transform that can be parallelized?
I need to manage CPU-heavy multitaskable jobs in an interactive application. Just as background, my specific application is an engineering design interface. As a user tweaks different parameters and options to a model, multiple simulations are run in the background and results displayed as they complete, likely even as the user is still editing values. Since the multiple simulations take variable time (some are milliseconds, some take 5 seconds, some take 10 minutes), it's basically a matter of getting feedback displayed as fast as possible, but often aborting jobs that started previously but are now no longer needed because of the user's changes have already invalidated them. Different user changes may invalidate different computations so at any time I may have 10 different simulations running. Somesimulations have multiple parts which have dependencies (simulations A and B can be seperately computed, but I need their results to seed simulation C so I need to wait for both A and B to finish first before starting C.)
I feel pretty confident that the code-level method to handle this kind of application is some kind of multithreaded job queue. This would include features of submitting jobs for execution, setting task priorities, waiting for jobs to finish, specifying dependencies (do this job, but only after job X and job Y have finished), canceling subsets of jobs that fit some criteria, querying what jobs remain, setting worker thread counts and priorities, and so on. And multiplatform support is very useful too.
These are not new ideas or desires in software, but I'm at the early design phase of my application where I need to make a choice about what library to use for managing such tasks. I've written my own crude thread managers in the past in C (I think it's a rite of passage) but I want to use modern tools to base my work on, not my own previous hacks.
The first thought is to run to OpenMP but I'm not sure it's what I want. OpenMP is great for parallelizing at a fine level, automatically unrolling loops and such. While multiplatform, it also invades your code with #pragmas. But mostly it's not designed for managing large tasks.. especially cancelling pending jobs or specifying dependencies. Possible, yes, but it's not elegant.
I noticed that Google Chrome uses such a job manager for even the most trivial tasks. The design goal seems to be to keep the user interaction thread as light and nimble as possible, so anything that can get spawned off asynchronously, should be. From looking at the Chrome source this doesn't seem to be a generic library, but it still is interesting to see how the design uses asynchronous launches to keep interaction fast. This is getting to be similar to what I'm doing.
There are a still other options:
Surge.Act: a Boost-like library for defining jobs. It builds on OpenMP, but does allow chaining of dependencies which is nice. It doesn't seem to feel like it's got a manager that can be queried, jobs cancelled, etc. It's a stale project so it's scary to depend on it.
Job Queue is quite close to what I'm thinking of, but it's a 5 year old article, not a supported library.
Boost.threads does have nice platform independent synchronization but that's not a job manager. POCO has very clean designs for task launching, but again not a full manager for chaining tasks. (Maybe I'm underestimating POCO though).
So while there are options available, I'm not satisfied and I feel the urge to roll my own library again. But I'd rather use something that's already in existence. Even after searching (here on SO and on the net) I haven't found anything that feels right, though I imagine this must be a kind of tool that is often needed, so surely there's some community library or at least common design.
On SO there's been some posts about job queues, but nothing that seems to fit.
My post here is to ask you all what existing tools I've missed, and/or how you've rolled your own such multithreaded job queue.
We had to build our own job queue system to meet requirements similar to yours ( UI thread must always respond within 33ms, jobs can run from 15-15000ms ), because there really was nothing out there that quite met our needs, let alone was performant.
Unfortunately our code is about as proprietary as proprietary gets, but I can give you some of the most salient features:
We start up one thread per core at the beginning of the program. Each pulls work from a global job queue. Jobs consist of a function object and a glob of associated data (really an elaboration on a func_ptr and void *). Thread 0, the fast client loop, isn't allowed to work on jobs, but the rest grab as they can.
The job queue itself ought to be a lockless data structure, such as a lock-free singly linked list (Visual Studio comes with one). Avoid using a mutex; contention for the queue is surprisingly high, and grabbing mutexes is costly.
Pack up all the necessary data for the job into the job object itself -- avoid having pointer from the job back into the main heap, where you'll have to deal with contention between jobs and locks and all that other slow, annoying stuff. For example, all the simulation parameters should go into the job's local data blob. The results structure obviously needs to be something that outlives the job: you can deal with this either by a) hanging onto the job objects even after they've finished running (so you can use their contents from the main thread), or b) allocating a results structure specially for each job and stuffing a pointer into the job's data object. Even though the results themselves won't live in the job, this effectively gives the job exclusive access to its output memory so you needn't muss with locks.
Actually I'm simplifying a bit above, since we need to choreograph exactly which jobs run on which cores, so each core gets its own job queue, but that's probably unnecessary for you.
I rolled my own, based on Boost.threads. I was quite surprised by how much bang I got from writing so little code. If you don't find something pre-made, don't be afraid to roll your own. Between Boost.threads and your experience since writing your own, it might be easier than you remember.
For premade options, don't forget that Chromium is licensed very friendly, so you may be able to roll your own generic library around its code.
Microsoft is working on a set of technologies for the next Version of Visual Studio 2010 called the Concurrency Runtime, the Parallel Pattern Library and the Asynchronous Agents Library which will probably help. The Concurrency Runtime will offer policy based scheduling, i.e. allowing you to manage and compose multiple scheduler instances (similar to thread pools but with affinitization and load balancing between instances), the Parallel Pattern Library will offer task based programming and parallel loops with an STL like programming model. The Agents library offers an actor based programming model and has support for building concurrent data flow pipelines, i.e. managing those dependencies described above. Unfortunately this isn't released yet, so you can read about it on our team blog or watch some of the videos on channel9 there is also a very large CTP that is available for download as well.
If you're looking for a solution today, Intel's Thread Building Blocks and boost's threading library are both good libraries and available now. JustSoftwareSolutions has released an implementation of std::thread which matches the C++0x draft and of course OpenMP is widely available if you're looking at fine-grained loop based parallelism.
The real challenge as other folks have alluded to is to correctly identify and decompose work into tasks suitable for concurrent execution (i.e. no unprotected shared state), understand the dependencies between them and minimize the contention that can occur on bottlenecks (whether the bottleneck is protecting shared state or ensuring the dispatch loop of a work queue is low contention or lock-free)... and to do this without scheduling implementation details leaking into the rest of your code.
-Rick
Would something like threadpool be useful to you? It's based on boost::threads and basically implements a simple thread task queue that passes worker functions off to the pooled threads.
I've been looking for near the same requirements. I'm working on a game with 4x-ish mechanics and scheduling different parts of what gets done almost exploded my brain. I have a complex set of work that needs to get accomplished at different time resolutions, and to a different degree of actual simulation depending on what system/region the player has actively loaded. This means as the player moves from system to system, I need to load a system to the current high resolution simulation, offload the last system to a lower resolution simulation, and do the same for active/inactive regions of systems. The different simulations are big lists of population, political, military, and economic actions based on profiles of each entity. I'm going to try to describe my issue and my approach so far and I hope it's useful at describe an alternative for you or someone else. The rough outline of the structure I'm building will use the following:
cpp-taskflow (A Modern C++ Parallel Task Programming Library) I'm going to make a library of modules that will be used as job construction parts. Each entry will have an API for initializing and destruction as well as pointers for communication. I'm hoping to write it in a way that they will be nest-able using the cpp-taskflow API to set-up all the dependencies at job creation time, but provide a means of live adjustment and having a kill-switch available. Most of what I'm making will be decision trees of state machines, or state machines of behavior trees so the job data structure will be settings and states of time-resolution tagged data pointing to actual stats and object values.
FlatBuffers I'm looking to use this library to build a "job list entry" as well as an "object wrapper" system. Each entry in the job queues will be a flatbuffer object describing the work needed done(settings for the module), as well as containing the data(or shared pointers to the data) for the work that needs done. The object storage flatbuffers will contain the data that represents entity tables. For me, most of the actual data will me arrays that need deciding/working on. I'm also looking to use flatbuffers as a communication/control channel between threads. I'm torn on making a master "router" thread all the others communicate through, or each one containing their own, and having some mechanism of discovery.
SQLite Since only the active regions/systems need higher resolution work done, some of the background job lists the game will create(for thousands of systems and their entities) will be pretty large and long lived. 100's of thousands - millions of jobs(big in my mind), each requiring an unknown amount of time to complete. In my case, I don't care when they get done, as long as they all do(long campains). I plan on each thread getting a table of an in-memory sqlite db as a job queue. Each entry will contain a blob of flatbuffer work, a pointer to a buffer to notify upon completion, a pointer to a control buffer for updates, and other fields decorating the job item(location, data ranges, priority) that will get filled as the job entry makes new jobs, and as the items are consumed into the database. This give me a way I can create relational ties between jobs and simply construct queries if I need to re-work/update jobs, remove them and their dependencies, or update/re-order priorities or dependencies. All this being used in an sqlite db also means that at any time I can dump the whole thing to disk and reload it later, or switch to attaching to and processing it from disk. Additionally, this gives me access to a lot of search and ordering algorithmic work I'd normally need a bunch of different types of containers for. Being able to use SQL queries gives me a lot of options to process the jobs.
The communication queue(as a db) is what I'm torn as to whether I should make access via the corresponding thread(each thread contains it's own messaging db, and the module API has locks/mutex abstracted for access), or have all updates, adds/removes, and communication via some master router thread into one large db. I have no idea which will give me the least headaches as far as mutexing and locks. I got a few days into making a monster spaghetti beast of shared pointers to sbuffer pools and lookup tables, so each thread had it's own buffer in, and separate out buffers. That's when I decided to just offload the giant list keeping to sqlite. Then I thought, why not just feed the flatbuffer objects of everything else into tables.
Having almost everything in a db means from each module, I can write sql statements that represent the view of the data I need to work on as well as pivot on the fly as to how the data is worked on. Having the jobs themselves in a db means I can do the same for them as well. SQLite has multi-threading access, so using it as a Multithreaded job queue manager shouldn't be too much of a stretch.
In summary, Cpp-Taskflow will allow you to setup complicated nested loops with dependency chaining and job-pool multithreading. Out of the box it comes with most of the structure you need. FlatBuffers will allow you to create job declarations and object wrappers easy to feed into stream-buffers as one unit of work and pass them between job threads, and SQLite will allow you to tag and queue the stream-buffer jobs into blob entries in a way that should allow adding, searching, ordering, updating, and removal with minimal work on your end. It also makes saving and reloading a breeze. Snapshots and roll-backs should also be doable, you just have to keep your mind wrapped around the order and resolution of events for the db.
Edit: Take this with a grain of salt though, I found your question because I'm trying to accomplish what Crashworks described. I'm thinking of using affinity to open long living threads and have the master thread run the majority of the Cpp-Taskflow hierarchy work, feeding jobs to the others. I've yet to use the sqlite meothod of job-queue/control communication, that's just my plan so far.
I hope someone finds this helpful.
You might want to look at Flow-Based Programming - it is based on data chunks streaming between asynchronous components. There are Java and C# versions of the driver, plus a number of precoded components. It is intrinsically multithreaded - in fact the only single-threaded code is within the components, although you can add timing constraints to the standard scheduling rules. Although it may be at too fine-grained a level for what you need, there may be stuff here you can use.
Take a look at boost::future (but see also this discussion and proposal) which looks like a really nice foundation for parallelism (in particular it seems to offer excellent support for C-depends-on-A-and-B type situations).
I looked at OpenMP a bit but (like you) wasn't convinced it would work well for anything but Fortran/C numeric code. Intel's Threading Building Blocks looked more interesting to me.
If it comes to it, it's not too hard to roll your own on top of boost::thread.
[Explanation: a thread farm (most people would call it a pool) draws work from a thread-safe queue of functors (tasks or jobs). See the tests and benchmark for examples of use. I have some extra complication to (optionally) support tasks with priorities, and the case where executing tasks can spawn more tasks into the work queue (this makes knowing when all the work is actually completed a bit more problematic; the references to "pending" are the ones which can deal with the case). Might give you some ideas anyway.]
You may like to look at Intel Thread Building Blocks. I beleave it does what you want and with version 2 it's Open Source.
There's plenty of distributed resource managers out there. The software that meets nearly all of your requirements is Sun Grid Engine. SGE is used on some of the worlds largest supercomputers and is in active development.
There's also similar solutions in Torque, Platform LSF, and Condor.
It sounds like you may want to roll your own but there's plenty of functionality in all of the above.
I don't know if you're looking for a C++ library (which I think you are), but Doug Lea's Fork/Join framework for Java 7 is pretty nifty, and does exactly what you want. You'd probably be able to implement it in C++ or find a pre-implemented library.
More info here:
http://artisans-serverintellect-com.si-eioswww6.com/default.asp?W1
A little late to the punch perhaps, but take a look also at ThreadWeaver:
http://en.wikipedia.org/wiki/ThreadWeaver