Related
I'm working on optimizing a function which uses floating point instructions.
For bench marking I need to know execution latency of the instructions to know theoretical possible performance.
I have found such manual for A57: http://infocenter.arm.com/help/topic/com.arm.doc.uan0015b/Cortex_A57_Software_Optimization_Guide_external.pdf
But, I don't find any such doc for A53.
Am I missing something?
Is there any such optimization guide available for A53?
Unfortunately, to the best of my knowledge, there's very little information about the Cortex A53 cycle timings. Crucially, from my measurements, the NEON latencies are quite different from what's happening on the Cortex A7.
I've written microbenchmarking software to figure out what's going on with regards to the instruction cycle timings. There are some tables in the paper that was related to this work (also on IACR eprint).
Some saint independently measured the instruction latencies.
http://hardwarebug.org/2014/05/15/cortex-a7-instruction-cycle-timings/
Note that the a53 evolved from the a7 and hence the timings are likely similar. It is a completely different design from the a57, which has a much longer pipeline and out of order execution.
Hypothetically speaking, if my scientific work was leading toward the development of functions/modules/subroutines (on a desktop), what would I need to know to incorporate it into a large-scale simulation to be run on a supercomputer (which might simulate molecules, fluids, reactions, and so on)?
My impression is that it has to do with taking advantage of certain libraries (e.g., BLAS, LAPLACK) where possible, revising algorithms (reducing iteration), profiling, parallelizing, considering memory-hard disk-processor use/access... I am aware of the adage, "want to optimize your code? don't do it", but if one were interested in learning about writing efficient code, what references might be available?
I think this question is language agnostic, but since many number-crunching packages for biomolecular simulation, climate modeling, etc. are written in some version of Fortran, this language would probably be my target of interest (and I have programmed rather extensively in Fortran 77).
Profiling is a must at any level of machinery. In common usage, I've found that scaling to larger and larger grids requires a better understanding of the grid software and the topology of the grid. In that sense, everything you learn about optimizing for one machine is still applicable, but understanding the grid software gets you additional mileage. Hadoop is one of the most popular and widespread grid systems, so learning about the scheduler options, interfaces (APIs and web interfaces), and other aspects of usage will help. Although you may not use Hadoop for a given supercomputer, it is one of the less painful methods for learning about distributed computing. For parallel computing, you may pursue MPI and other systems.
Additionally, learning to parallelize code on a single machine, across multiple cores or processors, is something you can begin learning on a desktop machine.
Recommendations:
Learn to optimize code on a single machine:
Learn profiling
Learn to use optimized libraries (after profiling: so that you see the speedup)
Be sure you know algorithms and data structures very well (*)
Learn to do embarrassingly parallel programming on multiple core machines.
Later: consider multithreaded programming. It's harder and may not pay off for your problem.
Learn about basic grid software for distributed processing
Learn about tools for parallel processing on a grid
Learn to program for alternative hardware, e.g. GPUs, various specialized computing systems.
This is language agnostic. I have had to learn the same sequence in multiple languages and multiple HPC systems. At each step, take a simpler route to learn some of the infrastructure and tools; e.g. learn multicore before multithreaded, distributed before parallel, so that you can see what fits for the hardware and problem, and what doesn't.
Some of the steps may be reordered depending on local computing practices, established codebases, and mentors. If you have a large GPU or MPI library in place, then, by all means, learn that rather than foist Hadoop onto your collaborators.
(*) The reason to know algorithms very well is that as soon as your code is running on a grid, others will see it. When it is hogging up the system, they will want to know what you're doing. If you are running a process that is polynomial and should be constant, you may find yourself mocked. Others with more domain expertise may help you find good approximations for NP-hard problems, but you should know that the concept exists.
Parallelization would be the key.
Since the problems you cited (e.g. CFD, multiphysics, mass transfer) are generally expressed as large-scale linear algebra problems, you need matrix routines that parallelize well. MPI is a standard for those types of problems.
Physics can influence as well. For example, it's possible to solve some elliptical problems efficiently using explicit dynamics and artificial mass and damping matricies.
3D multiphysics means coupled differential equations with varying time scales. You'll want a fine mesh to resolve details in both space and time, so the number of degrees of freedom will rise rapidly; time steps will be governed by the stability requirements of your problem.
If someone ever figures out how to run linear algebra as a map-reduce problem they'll have it knocked.
Hypothetically speaking, if my scientific work was leading toward the development of functions/modules/subroutines (on a desktop), what would I need to know to incorporate it into a large-scale simulation to be run on a supercomputer (which might simulate molecules, fluids, reactions, and so on)?
First, you would need to understand the problem. Not all problems can be solved in parallel (and I'm using the term parallel in as wide meaning as it can get). So, see how the problem is now solved. Can it be solved with some other method quicker. Can it be divided in independent parts ... and so on ...
Fortran is the language specialized for scientific computing, and during the recent years, along with the development of new language features, there has also been some very interesting development in terms of features that are aiming for this "market". The term "co-arrays" could be an interesting read.
But for now, I would suggest reading first into a book like Using OpenMP - OpenMP is a simpler model but the book (fortran examples inside) explains nicely the fundamentals. Message parsing interface (for friends, MPI :) is a larger model, and one of often used. Your next step from OpenMP should probably go in this direction. Books on the MPI programming are not rare.
You mentioned also libraries - yes, some of those you mentioned are widely used. Others are also available. A person who does not know exactly where the problem in performance lies should IMHO never try to undertake the task of trying to rewrite library routines.
Also there are books on parallel algorithms, you might want to check out.
I think this question is language agnostic, but since many number-crunching packages for biomolecular simulation, climate modeling, etc. are written in some version of Fortran, this language would probably be my target of interest (and I have programmed rather extensively in Fortran 77).
In short it comes down to understanding the problem, learning where the problem in performance is, re-solving the whole problem again with a different approach, iterating a few times, and by that time you'll already know what you're doing and where you're stuck.
We're in a position similar to yours.
I'm most in agreement with #Iterator's answer, but I think there's more to say.
First of all, I believe in "profiling" by the random-pausing method, because I'm not really interested in measuring things (It's easy enough to do that) but in pinpointing code that is causing time waste, so I can fix it. It's like the difference between a floodlight and a laser.
For one example, we use LAPACK and BLAS. Now, in taking my stack samples, I saw a lot of the samples were in the routine that compares characters. This was called from a general routine that multiplies and scales matrices, and that was called from our code. The matrix-manipulating routine, in order to be flexible, has character arguments that tell it things like, if a matrix is lower-triangular or whatever. In fact, if the matrices are not very large, the routine can spend more than 50% of its time just classifying the problem. Of course, the next time it is called from the same place, it does the same thing all over again. In a case like that, a special routine should be written. When it is optimized by the compiler, it will be as fast as it reasonably can be, and will save all that classifying time.
For another example, we use a variety of ODE solvers. These are optimized to the nth degree of course. They work by calling user-provided routines to calculate derivatives and possibly a jacobian matrix. If those user-provided routines don't actually do much, samples will indeed show the program counter in the ODE solver itself. However, if the user-provided routines do much more, samples will find the lower end of the stack in those routines mostly, because they take longer, while the ODE code takes roughly the same time. So, optimization should be concentrated in the user-provided routines, not the ODE code.
Once you've done several of the kind of optimization that is pinpointed by stack sampling, which can speed things up by 1-2 orders of magnitude, then by all means exploit parallelism, MPI, etc. if the problem allows it.
I'm a CS master student, and next semester I will have to start working on my thesis. I've had trouble coming up with a thesis idea, but I decided it will be related to Computer Graphics as I'm passionate about game development and wish to work as a professional game programmer one day.
Unfortunately I'm kinda new to the field of 3D Computer Graphics, I took an undergraduate course on the subject and hope to take an advanced course next semester, and I'm already reading a variety of books and articles to learn more. Still, my supervisor thinks its better if I come up with a general thesis idea now and then spend time learning about it in preparation for doing my thesis proposal. My supervisor has supplied me with some good ideas but I'd rather do something more interesting on my own, which hopefully has to do with games and gives me more opportunities to learn more about the field. I don't care if it's already been done, for me the thesis is more of an opportunity to learn about things in depth and to do substantial work on my own.
I don't know much about GPU programming and I'm still learning about shaders and languages like CUDA. One idea I had is to program an entire game (or as much as possible) on the GPU, including all the game logic, AI, and tests. This is inspired by reading papers on GPGPU and questions like this one I don't know how feasible that is with my knowledge, and my supervisor doesn't know a lot about recent GPUs. I'm sure with time I will be able to answer this question on my own, but it'd be handy if I could know the answer in advance so I could also consider other ideas.
So, if you've got this far, my question: Using only shaders or something like CUDA, can you make a full, simple 3D game that exploits the raw power and parallelism of GPUs? Or am I missing some limitation or difference between GPUs and CPUs that will always make a large portion of my code bound to CPU? I've read about physics engines running on the GPU, so why not everything else?
DISCLAIMER: I've done a PhD, but have never supervised a student of my own, so take all of what I'm about to say with a grain of salt!
I think trying to force as much of a game as possible onto a GPU is a great way to start off your project, but eventually the point of your work should be: "There's this thing that's an important part of many games, but in it's present state doesn't fit well on a GPU: here is how I modified it so it would fit well".
For instance, fortran mentioned that AI algorithms are a problem because they tend to rely on recursion. True, but, this is not necessarily a deal-breaker: the art of converting recursive algorithms into an iterative form is looked upon favorably by the academic community, and would form a nice center-piece for your thesis.
However, as a masters student, you haven't got much time so you would really need to identify the kernel of interest very quickly. I would not bother trying to get the whole game to actually fit onto the GPU as part of the outcome of your masters: I would treat it as an exercise just to see which part won't fit, and then focus on that part alone.
But be careful with your choice of supervisor. If your supervisor doesn't have any relevant experience, you should pick someone else who does.
I'm still waiting for a Gameboy Emulator that runs entirely on the GPU, which is just fed the game ROM itself and current user input and results in a texture displaying the game - maybe a second texture for sound output :)
The main problem is that you can't access persistent storage, user input or audio output from a GPU. These parts have to be on the CPU, by definition (even though cards with HDMI have audio output, but I think you can't control it from the GPU). Apart from that, you can already push large parts of the game code into the GPU, but I think it's not enough for a 3D game, since someone has to feed the 3D data into the GPU and tell it which shaders should apply to which part. You can't really randomly access data on the GPU or run arbitrary code, someone has to do the setup.
Some time ago, you would just setup a texture with the source data, a render target for the result data, and a pixel shader that would do the transformation. Then you rendered a quad with the shader to the render target, which would perform the calculations, and then read the texture back (or use it for further rendering). Today, things have been made simpler by the fourth and fifth generation of shaders (Shader Model 4.0 and whatever is in DirectX 11), so you can have larger shaders and access memory more easily. But still they have to be setup from the outside, and I don't know how things are today regarding keeping data between frames. In worst case, the CPU has to read back from the GPU and push again to retain game data, which is always a slow thing to do. But if you can really get to a point where a single generic setup/rendering cycle would be sufficient for your game to run, you could say that the game runs on the GPU. The code would be quite different from normal game code, though. Most of the performance of GPUs comes from the fact that they execute the same program in hundreds or even thousands of parallel shading units, and you can't just write a shader that can draw an image to a certain position. A pixel shader always runs, by definition, on one pixel, and the other shaders can do things on arbitrary coordinates, but they don't deal with pixels. It won't be easy, I guess.
I'd suggest just trying out the points I said. The most important is retaining state between frames, in my opinion, because if you can't retain all data, all is impossible.
First, Im not a computer engineer so my assumptions cannot even be a grain of salt, maybe nano scale.
Artificial intelligence? No problem.There are countless neural network examples running in parallel in google. Example: http://www.heatonresearch.com/encog
Pathfinding? You just try some parallel pathfinding algorithms that are already on internet. Just one of them: https://graphics.tudelft.nl/Publications-new/2012/BB12a/BB12a.pdf
Drawing? Use interoperability of dx or gl with cuda or cl so drawing doesnt cross pci-e lane. Can even do raytracing at corners so no z-fighting anymore, even going pure raytraced screen is doable with mainstream gpu using a low depth limit.
Physics? The easiest part, just iterate a simple Euler or Verlet integration and frequently stability checks if order of error is big.
Map/terrain generation? You just need a Mersenne-twister and a triangulator.
Save game? Sure, you can compress the data parallelly before writing to a buffer. Then a scheduler writes that data piece by piece to HDD through DMA so no lag.
Recursion? Write your own stack algorithm using main vram, not local memory so other kernels can run in wavefronts and GPU occupation is better.
Too much integer needed? You can cast to a float then do 50-100 calcs using all cores then cast the result back to integer.
Too much branching? Compute both cases if they are simple, so every core is in line and finish in sync. If not, then you can just put a branch predictor of yourself so the next time, it predicts better than the hardware(could it be?) with your own genuine algorithm.
Too much memory needed? You can add another GPU to system and open DMA channel or a CF/SLI for faster communication.
Hardest part in my opinion is the object oriented design since it is very weird and hardware dependent to build pseudo objects in gpu. Objects should be represented in host(cpu) memory but they must be separated over many arrays in gpu to be efficient. Example objects in host memory: orc1xy_orc2xy_orc3xy. Example objects in gpu memory: orc1_x__orc2_x__ ... orc1_y__orc2_y__ ...
The answer has already been chosen 6 years ago but for those interested to the actual question, Shadertoy, a live-coding WebGL platform, recently added the "multipass" feature allowing preservation of state.
Here's a live demo of the Bricks game running on Gpu.
I don't care if it's already been
done, for me the thesis is more of an
opportunity to learn about things in
depth and to do substantial work on my
own.
Then your idea of what a thesis is is completely wrong. A thesis must be an original research. --> edit: I was thinking about a PhD thesis, not a master thesis ^_^
About your question, the GPU's instruction sets and capabilities are very specific to vector floating point operations. The game logic usually does little floating point, and much logic (branches and decision trees).
If you take a look to the CUDA wikipedia page you will see:
It uses a recursion-free,
function-pointer-free subset of the C
language
So forget about implementing any AI algorithms there, that are essentially recursive (like A* for pathfinding). Maybe you could simulate the recursion with stacks, but if it's not allowed explicitly it should be for a reason. Not having function pointers also limits somewhat the ability to use dispatch tables for handling the different actions depending on state of the game (you could use again chained if-else constructions, but something smells bad there).
Those limitations in the language reflect that the underlying HW is mostly thought to do streaming processing tasks. Of course there are workarounds (stacks, chained if-else), and you could theoretically implement almost any algorithm there, but they will probably make the performance suck a lot.
The other point is about handling the IO, as already mentioned up there, this is a task for the main CPU (because it is the one that executes the OS).
It is viable to do a masters thesis on a subject and with tools that you are, when you begin, unfamiliar. However, its a big chance to take!
Of course a masters thesis should be fun. But ultimately, its imperative that you pass with distinction and that might mean tackling a difficult subject that you have already mastered.
Equally important is your supervisor. Its imperative that you tackle some problem they show an interest in - that they are themselves familiar with - so that they can become interested in helping you get a great grade.
You've had lots of hobby time for scratching itches, you'll have lots more hobby time in the future too no doubt. But master thesis time is not the time for hobbies unfortunately.
Whilst GPUs today have got some immense computational power, they are, regardless of things like CUDA and OpenCL limited to a restricted set of uses, whereas the CPU is more suited towards computing general things, with extensions like SSE to speed up specific common tasks. If I'm not mistaken, some GPUs have the inability to do a division of two floating point integers in hardware. Certainly things have improved greatly compared to 5 years ago.
It'd be impossible to develop a game to run entirely in a GPU - it would need the CPU at some stage to execute something, however making a GPU perform more than just the graphics (and physics even) of a game would certainly be interesting, with the catch that game developers for PC have the biggest issue of having to contend with a variety of machine specification, and thus have to restrict themselves to incorporating backwards compatibility, complicating things. The architecture of a system will be a crucial issue - for example the Playstation 3 has the ability to do multi gigabytes a second of throughput between the CPU and RAM, GPU and Video RAM, however the CPU accessing GPU memory peaks out just past 12MiB/s.
The approach you may be looking for is called "GPGPU" for "General Purpose GPU". Good starting points may be:
http://en.wikipedia.org/wiki/GPGPU
http://gpgpu.org/
Rumors about spectacular successes in this approach have been around for a few years now, but I suspect that this will become everyday practice in a few years (unless CPU architectures change a lot, and make it obsolete).
The key here is parallelism: if you have a problem where you need a large number of parallel processing units. Thus, maybe neural networks or genetic algorithms may be a good range of problems to attack with the power of a GPU. Maybe also looking for vulnerabilities in cryptographic hashes (cracking the DES on a GPU would make a nice thesis, I imagine :)). But problems requiring high-speed serial processing don't seem so much suited for the GPU. So emulating a GameBoy may be out of scope. (But emulating a cluster of low-power machines might be considered.)
I would think a project dealing with a game architecture that targets multiple core CPUs and GPUs would be interesting. I think this is still an area where a lot of work is being done. In order to take advantage of current and future computer hardware, new game architectures are going to be needed. I went to GDC 2008 and there were ome talks related to this. Gamebryo had an interesting approach where they create threads for processing computations. You can designate the number of cores you want to use so that if you don't starve out other libraries that might be multi-core. I imagine the computations could be targeted to GPUs as well.
Other approaches included targeting different systems for different cores so that computations could be done in parallel. For instance, the first split a talk suggested was to put the renderer on its own core and the rest of the game on another. There are other more complex techniques but it all basically boils down to how do you get the data around to the different cores.
We've got a fairly large application running on VxWorks 5.5.1 that's been developed and modified for around 10 years now. We have some simple home-grown tools to show that we are not using too much memory or too much processor, but we don't have a good feel for how much headroom we actually have. It's starting to make it difficult to do estimates for future enhancements.
Does anybody have any suggestions on how to profile such a system? We've never had much luck getting the Wind River tools to work.
For bonus points: the other complication is that our system has very different behaviors at different times; during start-up it does a lot of stuff, then it sits relatively idle except for brief bursts of activity. If there is a profiler with some programmatic way to have to record state information, I think that'd be very useful too.
FWIW, this is compiled with GCC and written entirely in C.
I've done a lot of performance tuning of various kinds of software, including embedded applications. I won't discuss memory profiling - I think that is a different issue.
I can only guess where the "well-known" idea originated that to find performance problems you need to measure performance of various parts. That is a top-down approach, similar to the way governments try to control budget waste, by subdividing. IMHO, it doesn't work very well.
Measurement is OK for seeing if what you did made a difference, but it is poor at telling you what to fix.
What is good at telling you what to fix is a bottom-up approach, in which you examine a representative sample of microscopic units of what is being spent, and finding out the full explanation of why each one is being spent. This works for a simple statistical reason. If there is a reason why some percent (for example 40%) of samples can be saved, on average 40% of samples will show it, and it doesn't require a huge number of samples. It does require that you examine each sample carefully, and not just sort of aggregate them into bigger bunches.
As a historical example, this is what Harry Truman did at the outbreak of the U.S. involvement in WW II. There was terrific waste in the defense industry. He just got in his car, drove out to the factories, and interviewed the people standing around. Then he went back to the U.S. Senate, explained what the problems were exactly, and got them fixed.
Maybe this is more of an answer than you wanted. Specifically, this is the method I use, and this is a blow-by-blow example of it.
ADDED: I guess the idea of finding-by-measuring is simply natural. Around '82 I was working on an embedded system, and I needed to do some performance tuning. The hardware engineer offered to put a timer on the board that I could read (providing from his plenty). IOW he assumed that finding performance problems required timing. I thanked him and declined, because by that time I knew and trusted the random-halt technique (done with an in-circuit-emulator).
If you have the Auxiliary Clock available, you could use the SPY utility (configurable via the config.h file) which does give you a very rough approximation of which tasks are using the CPU.
The nice thing about it is that it does not require being attached to the Tornado environment and you can use it from the Kernel shell.
Otherwise, btpierre's suggestion of using taskHookAdd has been used successfully in the past.
I've worked on systems that have had luck using locally-built monitoring utilities based on taskSwitchHookAdd and related functions (delete hook, etc).
"Simply" use this to track the number of ticks a given task runs. I realize that this is fairly gross scale information for profiling, but it can be useful depending on your needs.
To see how much cpu% each task is using, calculate the percentage of ticks assigned to each task.
To see how much headroom you have, add a lowest priority "idle" task that just does "while(1){}", and see how much cpu% it is assigned to it. Roughly speaking, that's your headroom.
A two parter:
1) Say you're designing a new type of application and you're in the process of coming up with new algorithms to express the concepts and content -- does it make sense to attempt to actively not consider optimisation techniques at that stage, even if in the back of your mind you fear it might end up as O(N!) over millions of elements?
2) If so, say to avoid limiting cool functionality which you might be able to optimise once the proof-of-concept is running -- how do you stop yourself from this programmers habit of a lifetime? I've been trying mental exercises, paper notes, but I grew up essentially counting clock cycles in assembler and I continually find myself vetoing potential solutions for being too wasteful before fully considering the functional value.
Edit: This is about designing something which hasn't been done before (the unknown), when you're not even sure if it can be done in theory, never mind with unlimited computing power at hand. So answers along the line of "of course you have to optimise before you have a prototype because it's an established computing principle," aren't particularly useful.
I say all the following not because I think you don't already know it, but to provide moral support while you suppress your inner critic :-)
The key is to retain sanity.
If you find yourself writing a Theta(N!) algorithm which is expected to scale, then you're crazy. You'll have to throw it away, so you might as well start now finding a better algorithm that you might actually use.
If you find yourself worrying about whether a bit of Pentium code, that executes precisely once per user keypress, will take 10 cycles or 10K cycles, then you're crazy. The CPU is 95% idle. Give it ten thousand measly cycles. Raise an enhancement ticket if you must, but step slowly away from the assembler.
Once thing to decide is whether the project is "write a research prototype and then evolve it into a real product", or "write a research prototype". With obviously an expectation that if the research succeeds, there will be another related project down the line.
In the latter case (which from comments sounds like what you have), you can afford to write something that only works for N<=7 and even then causes brownouts from here to Cincinnati. That's still something you weren't sure you could do. Once you have a feel for the problem, then you'll have a better idea what the performance issues are.
What you're doing, is striking a balance between wasting time now (on considerations that your research proves irrelevant) with wasting time later (because you didn't consider something now that turns out to be important). The more risky your research is, the more you should be happy just to do something, and worry about what you've done later.
My big answer is Test Driven Development. By writing all your tests up front then you force yourself to only write enough code to implement the behavior you are looking for. If timing and clock cycles becomes a requirement then you can write tests to cover that scenario and then refactor your code to meet those requirements.
Like security and usability, performance is something that has to be considered from the beginning of the project. As such, you should definitely be designing with good performance in mind.
The old Knuth line is "We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil." O(N!) to O(poly(N)) is not a "small efficiency"!
The best way to handle type 1 is to start with the simplest thing that could possibly work (O(N!) cannot possibly work unless you're not scaling past a couple dozen elements!) and encapsulate it from the rest of the application so you could rewrite it to a better approach assuming that there is going to be a performance issue.
Optimization isn't exactly a danger; its good to think about speed to some extent when writing code, because it stops you from implementing slow and messy solutions when something simpler and faster would do. It also gives you a check in your mind on whether something is going to be practical or not.
The worst thing that can happen is you design a large program explicitly ignoring optimization, only to go back and find that your entire design is completely useless because it cannot be optimized without completely rewriting it. This never happens if you consider everything when writing it--and part of that "everything" is potential performance issues.
"Premature optimization is the root of all evil" is the root of all evil. I've seen projects crippled by overuse of this concept. At my company we have a software program that broadcasts transport streams from disk on the network. It was originally created for testing purposes (so we would just need a few streams at once), but it was always in the program's spec requirements that it work for larger numbers of streams so it could later be used for video on demand.
Because it was written completely ignoring speed, it was a mess; it had tons of memcpys despite the fact that they should never be necessary, its TS processing code was absurdly slow (it actually parsed every single TS packet multiple times), and so forth. It handled a mere 40 streams at a time instead of the thousands it was supposed to, and when it actually came time to use it for VOD, we had to go back and spend a huge amount of time cleaning it up and rewriting large parts of it.
"First, make it run. Then make it run fast."
or
"To finish first, first you have to finish."
Slow existing app is usually better than ultra-fast non-existing app.
First of all peopleclaim that finishign is only thing that matters (or almost).
But if you finish a product that has O(N!) complexity on its main algorithm, as a rule of thumb you did not finished it! You have an incomplete and unacceptable product for 99% of the cases.
A reasonable performance is part of a working product. A perfect performance might not be. If you finish a text editor that needs 6 GB of memory to write a short note, then you have not finished a product at all, you have only a waste of time at your hands.. You must remember always that is not only delivering code that makes a product complete, is making it achieve capability of supplying the costumer/users needs. If you fail at that it matters nothing that you have finished the code writing in the schedule.
So all optimizations that avoid a resulting useless product are due to be considered and applied as soon as they do not compromise the rest of design and implementation proccess.
"actively not consider optimisation" sounds really weird to me. Usually 80/20 rule works quite good. If you spend 80% of your time to optimize program for less than 20% of use cases, it might be better to not waste time unless those 20% of use-cases really matter.
As for perfectionism, there is nothing wrong with it unless it starts to slow you down and makes you miss time-frames. Art of computer programming is an act of balancing between beauty and functionality of your applications. To help yourself consider learning time-management. When you learn how to split and measure your work, it would be easy to decide whether to optimize it right now, or create working version.
I think it is quite reasonable to forget about O(N!) worst case for an algorithm. First you need to determine that a given process is possible at all. Keep in mind that Moore's law is still in effect, so even bad algorithms will take less time in 10 or 20 years!
First optimize for Design -- e.g. get it to work first :-) Then optimize for performance. This is the kind of tradeoff python programmers do inherently. By programming in a language that is typically slower at run-time, but is higher level (e.g. compared to C/C++) and thus faster to develop, python programmers are able to accomplish quite a bit. Then they focus on optimization.
One caveat, if the time it takes to finish is so long that you can't determine if your algorithm is right, then it is a very good time to worry about optimization earlier up stream. I've encountered this scenario only a few times -- but good to be aware of it.
Following on from onebyone's answer there's a big difference between optimising the code and optimising the algorithm.
Yes, at this stage optimising the code is going to be of questionable benefit. You don't know where the real bottlenecks are, you don't know if there is going to be a speed problem in the first place.
But being mindful of scaling issues even at this stage of the development of your algorithm/data structures etc. is not only reasonable but I suspect essential. After all there's not going to be a lot of point continuing if your back-of-the-envelope analysis says that you won't be able to run your shiny new application once to completion before the heat death of the universe happens. ;-)
I like this question, so I'm giving an answer, even though others have already answered it.
When I was in grad school, in the MIT AI Lab, we faced this situation all the time, where we were trying to write programs to gain understanding into language, vision, learning, reasoning, etc.
My impression was that those who made progress were more interested in writing programs that would do something interesting than do something fast. In fact, time spent worrying about performance was basically subtracted from time spent conceiving interesting behavior.
Now I work on more prosaic stuff, but the same principle applies. If I get something working I can always make it work faster.
I would caution however that the way software engineering is now taught strongly encourages making mountains out of molehills. Rather than just getting it done, folks are taught to create a class hierarchy, with as many layers of abstraction as they can make, with services, interface specifications, plugins, and everything under the sun. They are not taught to use these things as sparingly as possible.
The result is monstrously overcomplicated software that is much harder to optimize because it is much more complicated to change.
I think the only way to avoid this is to get a lot of experience doing performance tuning and in that way come to recognize the design approaches that lead to this overcomplication. (Such as: an over-emphasis on classes and data structure.)
Here is an example of tuning an application that has been written in the way that is generally taught.
I will give a little story about something that happened to me, but not really an answer.
I am developing a project for a client where one part of it is processing very large scans (images) on the server. When i wrote it i was looking for functionality, but i thought of several ways to optimize the code so it was faster and used less memory.
Now an issue has arisen. During Demos to potential clients for this software and beta testing, on the demo unit (self contained laptop) it fails due to too much memory being used. It also fails on the dev server with really large files.
So was it an optimization, or was it a known future bug. Do i fix it or oprtimize it now? well, that is to be determined as their are other priorities as well.
It just makes me wish I did spend the time to reoptimize the code earlier on.
Think about the operational scenarios. ( use cases)
Say that we're making a pizza-shop finder gizmo.
The user turns on the machine. It has to show him the nearest Pizza shop in meaningful time. It Turns out our users want to know fast: in under 15 seconds.
So now, any idea you have, you think: is this going to ever, realistically run in some time less than 15 seconds, less all other time spend doing important stuff..
Or you're a trading system: accurate sums. Less than a millisecond per trade if you can, please. (They'd probably accept 10ms), so , agian: you look at every idea from the relevant scenarios point of view.
Say it's a phone app: has to start in under (how many seconds)
Demonstrations to customers fomr laptops are ALWAYS a scenario. We've got to sell the product.
Maintenance, where some person upgrades the thing are ALWAYS a scenario.
So now, as an example: all the hard, AI heavy, lisp-customized approaches are not suitable.
Or for different strokes, the XML server configuration file is not user friendly enough.
See how that helps.
If I'm concerned about the codes ability to handle data growth, before I get too far along I try to set up sample data sets in large chunk increments to test it with like:
1000 records
10000 records
100000 records
1000000 records
and see where it breaks or becomes un-usable. Then you can decide based on real data if you need to optimize or re-design the core algorithms.