I have recently begun working on a project to establish how best to leverage the processing power available in modern graphics cards for general programming. It seems that the field general purpose GPU programming (GPGPU) has a large bias towards scientific applications with a lot of heavy math as this fits well with the GPU computational model. This is all good and well, but most people don't spend all their time running simulation software and the like so we figured it might be possible to create a common foundation for easily building GPU-enabled software for the masses.
This leads to the question I would like to pose; What are the most common types of work performed by programs? It is not a requirement that the work translates extremely well to GPU programming as we are willing to accept modest performance improvements (Better little than nothing, right?).
There are a couple of subjects we have in mind already:
Data management - Manipulation of large amounts of data from databases
and otherwise.
Spreadsheet type programs (Is somewhat related to the above).
GUI programming (Though it might be impossible to get access to the
relevant code).
Common algorithms like sorting and searching.
Common collections (And integrating them with data manipulation
algorithms)
Which other coding tasks are very common? I suspect a lot of the code being written is of the category of inventory management and otherwise tracking of real 'objects'.
As I have no industry experience I figured there might be a number of basic types of code which is done more often than I realize but which just doesn't materialize as external products.
Both high level programming tasks as well as specific low level operations will be appreciated.
General programming translates terribly to GPUs. GPUs are dedicated to performing fairly simple tasks on streams of data at a massive rate, with massive parallelism. They do not deal well with the rich data and control structures of general programming, and there's no point trying to shoehorn that into them.
General programming translates terribly to GPUs. GPUs are dedicated to performing fairly simple tasks on streams of data at a massive rate, with massive parallelism. They do not deal well with the rich data and control structures of general programming, and there's no point trying to shoehorn that into them.
This isn't too far away from my impression of the situation but at this point we are not concerning ourselves too much with that. We are starting out by getting a broad picture of which options we have to focus on. After that is done we will analyse them a bit deeper and find out which, if any, are plausible options. If we end up determining that it is impossible to do anything within the field, and we are only increasing everybody's electricity bill then that is a valid result as well.
Things that modern computers do a lot of, where a little benefit could go a long way? Let's see...
Data management: relational database management could benefit from faster relational joins (especially joins involving a large number of relations). Involves massive homogeneous data sets.
Tokenising, lexing, parsing text.
Compilation, code generation.
Optimisation (of queries, graphs, etc).
Encryption, decryption, key generation.
Page layout, typesetting.
Full text indexing.
Garbage collection.
I do a lot of simplifying of configuration. That is I wrap the generation/management of configuration values inside a UI. The primary benefit is I can control work flow and presentation to make it simpler for non-techie users to configure apps/sites/services.
The other thing to consider when using a GPU is the bus speed, Most Graphics cards are designed to have a higher bandwidth when transferring data from the CPU out to the GPU as that's what they do most of the time. The bandwidth from the GPU back up to the CPU, which is needed to return results etc, isn't as fast. So they work best in a pipelined mode.
You might want to take a look at the March/April issue of ACM's Queue magazine, which has several articles on GPUs and how best to use them (besides doing graphics, of course).
Related
I have code doing a lot of operations with objects which can be represented as arrays.
When does it make to sense to use GPGPU environments (like CUDA) in an application? Can I predict performance gains before writing real code?
The convenience depends on a number of factors. Elementwise independent operations on large arrays/matrices are a good candidate.
For your particular problem (machine learning/fuzzy logic), I would recommend reading some related documents, as
Large Scale Machine Learning using NVIDIA CUDA
and
Fuzzy Logic-Based Image Processing Using Graphics Processor Units
to have a feeling on the speedup achieved by other people.
As already mentioned, you should specify your problem. However, if large parts of your code involve operations on your objects that are independent in a sense that object n does not have to wait for the results of the operations objects 0 to n-1, GPUs may enhance performance.
You could go to CUDA Zone to get yourself a general idea about what CUDA can do and do better than CPU.
https://developer.nvidia.com/category/zone/cuda-zone
CUDA has already provided lots of performance libraries, tools and ecosystems to reduce the development difficulty. It could also help you understand what kind of operations CUDA are good at.
https://developer.nvidia.com/cuda-tools-ecosystem
Further more, CUDA provided benchmark report on some of the most common and representative operations. You could find if your code can benefit from that.
https://developer.nvidia.com/sites/default/files/akamai/cuda/files/CUDADownloads/CUDA_5.0_Math_Libraries_Performance.pdf
What are some more interesting graph data structures for working with networks? I am interested in structures which may offer some particular advantage in terms of traversing the network, finding random nodes, size in memory or for insertion/deletion/temporary hiding of nodes for example.
Note: I'm not so much interested in database like designs for addressing external memory problems.
One of my personal favorites is the link/cut tree, a data structure for partitioning a graph into a family of directed trees. This lets you solve network flow problems asymptotically faster than more traditional methods and can be used as a more powerful generalization of the union/find structure you may have heard of before.
I've heard of Skip Graphs ( http://www.google.com/search?ie=UTF-8&oe=UTF-8&sourceid=navclient&gfns=1&q=skip+graphs ), a probabilistic graph structure that is - as far as I know - already in use in some peer-to-peer applications.
These graphs are kind of self-organizing and their goal is to achieve a good connectivity and a small diameter. There is a distributed algorithm that tries to achieve such graphs: http://www14.informatik.tu-muenchen.de/personen/jacob/Publications/podc09.pdf
What is the purpose of abstraction in coding:
Programmer's efficiency or program's efficiency?
Our professor said that it is used merely for helping the programmer comprehend & modify programs faster to suit different scenarios. He also contended that it adds an extra burden on the program's performance. I am not exactly clear by what this means.
Could someone kindly elaborate?
I would say he's about half right.
The biggest purpose is indeed to help the programmer. The computer couldn't care less how abstracted your program is. However, there is a related, but different, benefit - code reuse. This isn't just for readability though, abstraction is what lets us plug various components into our programs that were written by others. If everything were just mixed together in one code file, and with absolutely no abstraction, you would never be able to write anything even moderately complex, because you'd be starting with the bare metal every single time. Just writing text on the screen could be a week long project.
About performance, that's a questionable claim. I'm sure it depends on the type and depth of the abstraction, but in most cases I don't think the system will notice a hit. Especially modern compiled languages, which actually "un-abstract" the code for you (things like loop unrolling and function inlining) sometimes to make it easier on the system.
Your professor is correct; abstraction in coding exists to make it easier to do the coding, and it increases the workload of the computer in running the program. The trick, though, is to make the (hopefully very tiny) increase in computer workload be dwarfed by the increase in programmer efficiency.
For example, on an extremely low-level; object-oriented code is an abstraction that helps the programmer, but adds some overhead to the program in the end in extra 'stuff' in memory, and extra function calls.
Since Abstraction is really the process of pulling out common pieces of functionality into re-usable components (be it abstract classes, parent classes, interfaces, etc.) I would say that it is most definitely a Programmer's efficiency.
Saying that Abstraction comes at the cost of performance is treading on unstable ground at best though. With most modern languages, abstraction (thus enhanced flexibility) can be had a little to no cost to the performance of the application.
What abstraction is is effectively outlined in the link Tesserex posted. To your professor's point about adding an additional burden on the program, this is actually fairly true. However, the burden in modern systems is negligible. Think of it in terms of what actually happens when you call a method: each additional method you call requires adding a number of additional data structures to the stack and then handling the return values also placed on the stack. So for instance, calling
c = add(a, b);
which looks something like
public int add(int a, int b){
return a + b;
}
requires pushing two integers onto the stack for the parameters and then pushing an additional one onto the stack for the return value. However, no memory interaction is required if both values are already in registers -- it's a simple, one-instruction call. Given that memory operations are much slower than register operations, you can see where the notion of a performance hit comes from.
Ultimately, every method call you make is going to increase the overhead of your program a little bit. However as #Tesserex points out, it's minute in most modern computer systems and as #Andrew Barber points out, that compromise is usually totally dwarfed by the increase in programmer efficiency.
Abstraction is a tool to make it easier for the programmer. The abstraction may or may not have an effect on the runtime performance of the system.
For an example of an abstraction that doesn't alter performance consider assembly. The pneumonics like mov and add are an abstraction that makes opcodes easier to remember, as compared to remembering byte-codes and other instruction encoding details. However, given the 1 to 1 mapping I'd suggest its clear that this abstraction has 0 effect on final performance.
There's not a clear-cut situation that abstraction makes life easier for the programmer at the expense of more work for the computer.
Although a higher level of abstraction typically adds at least a small amount of overhead to executing a discrete unit of code, it's also what allows the programmer to think about a problem in larger "units" so he can do a better job of understanding an entire problem, and avoid executing mane (or at least some) of those discrete units of code.
Therefore, a higher level of abstraction will often lead to faster-executing programs as long as you avoid adding too much overhead. The problem, of course, is that there's no easy or simple definition of how much overhead is too much. That stems largely from the fact that the amount of overhead that's acceptable depends heavily on the problem being solved, and the degree to which working at a higher level of abstraction allows the programmer to recognize operations that are truly unnecessary, and eliminate them.
I had started working on GPGPU some days ago and successfully implemented cholesky factorization with good performacne and I attended a conference on High Performance Computing where some people said that "GPGPU is a Hack".
I am still confused what does it mean and why they were saying it hack. One said that this is hack because you are converting your problem into a matrix and doing operations on it. But still I am confused that does people think it is a hack or if yes then why?
Can anyone help me, why they called it a hack while I found nothing wrong with it.
One possible reason for such opinion is that the GPU was not originally intended for general purpose computations. Also programming a GPU is less traditional and more hardcore and therefore more likely to be perceived as a hack.
The point that "you convert the problem into a matrix" is not reasonable at all. Whatever task you solve with writing code you choose reasonable data structures. In case of GPU matrices are likely the most reasonable datastructures and it's not a hack but just a natural choice to use them.
However I suppose that it's a matter of time for GPGPU becoming widespread. People just have to get used to the idea. After all who cares which unit of the computer runs the program?
On the GPU, having efficient memory access is paramount to achieving optimal performance. This often involves restructuring or even choosing entirely new algorithms and data structures. This is reason why GPU programming can be perceived as a hack.
Secondly, adapting an existing algorithm to run on the GPU is not in and of itself science. The relatively low scientific contribution of some GPU algorithm-related papers has led to a negative perception of GPU programming as strictly "engineering".
Obviously, only the person who said that can say for certain why he said it, but, here's my take:
A "Hack" is not a bad thing.
It forces people to learn new programming languages and concepts. For people who are just trying to model the weather or protein folding or drug reactions, this is an unwelcome annoyance. They didn't really want to learn FORTRAN (or whatever) in the first place, and now the have to learn another programming system.
The programming tools are NOT very mature yet.
The hardware isn't as reliable as CPUs (yet) so all of the calculations have to be done twice to make sure you've got the right answer. One reason for this is that GPUs don't come with error-correcting memory yet, so if you're trying to build a supercomputer with thousands of processors, the probability of a cosmic ray flipping a bit in you numbers approaches certainty.
As for the comment "you are converting your problem into a matrix and doing operations on it", I think that shows a lot of ignorance. Virtually ALL of high-performance computing fits that description!
One of the major problems in GPGPU for the past few years and probably for the next few is that programming them for arbitrary tasks was not very easy. Up until DX10 there was no integer support among GPUs and branching is still very poor. This is very much a situation where in order to get maximum benefit you have to write your code in a very awkward manner to extract all sorts of efficiency gains from the GPU. This is because you're running on hardware that is still dedicated to processing polygons and textures, rather than abstract parallel tasks.
Obviously, thats my take on it and YMMV
The GPGPU harks back to the days of the math co-processor. A hack is a shortcut to solving a long winded problem. GPGPU is a hack just like NAT on top of IPV4 is a hack. Computational problems just like networks are getting bigger as we try to do more, GPGPU is an useful interim solution, whether it stays outside the core CPU chip and has separate cranky API or gets sucked into the CPU via API or manufacture is up to the path finders.
I suppose he meant that using GPGPU forced you to restructure your implementation, so that it fitted the hardware, not the problem domain. Elegant implementation should fit the latter.
Note, that the word "hack" may have several different meanings:
http://www.urbandictionary.com/define.php?term=hack
I have to build a distributed application, using MPI.
One of the decision that I have to take is how to map instances of classes into process (and then into machines), in order to take maximum advantages from a distributed environment.
My question is: there is a model that let me choose the better mapping? I mean, some arrangements are surely wrong (for ex., putting in two different machines two objects that should process together a fairly large amount of data, in a sequential manner, without a stream of tokens to process), but there's a systematically way to determine such wrong arrangements, determined by flow of execution, message complexity, time taken by the computation done by the algorithmic components?
Well, there are data flow diagrams. Those can help identify parallelism's opportunities and pitfalls. The references on the wikipedia page might give you some more theoretical grounding.
When I worked at Lockheed Martin, I was exposed to CSIM, a tool they developed for modeling algorithm mapping to processing blocks.
Another thing you might try is the Join Calculus. I've found examples of programming with it to be surprisingly intuitive, and I think it's well grounded in theory. I'm not sure why it hasn't caught on more.
The other approach is the Pi Calculus, and I think that might be more popular, though it seems harder to understand.
A practical solution to this would be using a different model of distributed-memory parallel programming, that directly addresses your concerns. I work on the Charm++ programming system, whose model is that of individual objects sending messages from one to another. The runtime system facilitates automatic mapping of these objects to available processors, to account for issues of load balance and communication locality.