Search Results

Search found 671 results on 27 pages for 'optimizing'.

Page 9/27 | < Previous Page | 5 6 7 8 9 10 11 12 13 14 15 16  | Next Page >

  • [Qt/C++] Need help in optimizing a drawing code ...

    - by Ahmad
    Hello all ... I needed some help in trying to optimize this code portion ... Basically here's the thing .. I'm making this 'calligraphy pen' which gives the calligraphy effect by simply drawing a lot of adjacent slanted lines ... The problem is this: When I update the draw region using update() after every single draw of a slanted line, the output is correct, in the sense that updates are done in a timely manner, so that everything 'drawn' using the pen is immediately 'seen' the drawing.. however, because a lot (100s of them) of updates are done, the program slows down a little when run on the N900 ... When I try to do a little optimization by running update after drawing all the slanted lines (so that all lines are updated onto the drawing board through a single update() ), the output is ... odd .... That is, immediately after drawing the lines, they lines seem broken (they have vacant patches where the drawing should have happened as well) ... however, if I trigger a redrawing of the form window (say, by changing the size of the form), the broken patches are immediately fixed !! When I run this program on my N900, it gets the initial broken output and stays like that, since I don't know how to enforce a redraw in this case ... Here is the first 'optimized' code and output (partially correct/incorrect) void Canvas::drawLineTo(const QPoint &endPoint) { QPainter painter(&image); painter.setPen(QPen(Qt::black,1,Qt::SolidLine,Qt::RoundCap,Qt::RoundJoin)); int fx=0,fy=0,k=0; qPoints.clear(); connectingPointsCalculator2(qPoints,lastPoint.x(),lastPoint.y(),endPoint.x(),endPoint.y()); int i=0; int x,y; for(i=0;i<qPoints.size();i++) { x=qPoints.at(i).x(); y=qPoints.at(i).y(); painter.setPen(Qt::black); painter.drawLine(x-5,y-5,x+5,y+5); **// Drawing slanted lines** } **//Updating only once after many draws:** update (QRect(QPoint(lastPoint.x()-5,lastPoint.y()-5), QPoint(endPoint.x()+5,endPoint.y()+5)).normalized()); modified = true; lastPoint = endPoint; } Image right after scribbling on screen: http://img823.imageshack.us/img823/8755/59943912.png After re-adjusting the window size, all the broken links above are fixed like they should be .. Here is the second un-optimized code (its output is correct right after drawing, just like in the second picture above): void Canvas::drawLineTo(const QPoint &endPoint) { QPainter painter(&image); painter.setPen(QPen(Qt::black,1,Qt::SolidLine,Qt::RoundCap,Qt::RoundJoin)); int fx=0,fy=0,k=0; qPoints.clear(); connectingPointsCalculator2(qPoints,lastPoint.x(),lastPoint.y(),endPoint.x(),endPoint.y()); int i=0; int x,y; for(i=0;i<qPoints.size();i++) { x=qPoints.at(i).x(); y=qPoints.at(i).y(); painter.setPen(Qt::black); painter.drawLine(x-5,y-5,x+5,y+5); **// Drawing slanted lines** **//Updating repeatedly during the for loop:** update(QRect(QPoint(x-5,y-5), QPoint(x+5,y+5)).normalized());//.adjusted(-rad,-rad,rad,rad)); } modified = true; int rad = (myPenWidth / 2) + 2; lastPoint = endPoint; } Can anyone see what the issue might be ?

    Read the article

  • Getting Optimal Performance from Oracle E-Business Suite

    - by Steven Chan (Oracle Development)
    Performance tuning and optimization in E-Business Suite environments can involve many different components and diagnostic tools.  Samer Barakat, Senior Architect in our Applications Performance group, held an OpenWorld 2013 session that covered: Performance triage, analysis and diagnostic tools Optimizing the E-Business Suite application tier, including Concurrent Manager Optimizing the E-Business Suite database tier Optimizing the E-Business Suite on Real Application Clusters (RAC) E-Business Suite on engineered systems, including Exadata and Exalogic Optimizing E-Business Suite data management, including archiving and purging  The Applications Performance group works with the world's largest E-Business Suite customers to isolate and resolve performance bottlenecks. This team has helped tune the E-Business Suite environments of world's largest companies to handle staggering amounts of transactional volume in multi-terabyte databases.  This group also publishes our official Oracle Apps benchmarks, white papers, and performance metrics. This is an essential set of tips and techniques that all EBS sysadmins and DBAs can use to improve the performance of their environments: Getting Optimal Performance from Oracle E-Business Suite (PDF, 1.7 MB) OpenWorld 2013 presentations are only available for approximately six months -- until ~March 2013.  Download this one while it's still available. Related Articles E-Business Suite Technology Sessions at OpenWorld 2013 OAUG/Collaborate Recap: Best Practices for E-Business Suite Performance Tuning

    Read the article

  • How to smartly optimize ads on website

    - by YardenST
    I've a content website that presents ads. Now, my team want to optimize it for a better experience for the users. (we really believe our ads are good for our users.) We are sure that every website deals with this issue and there must be some known ways and methods to deal with it, that smart people thought of before. so what i'm looking is a tested, working method to optimize ads. for example: if i was asking about optimizing my website in Google, I would expect you to answer me: learn SEO if i was asking about optimizing the use of my website: usability testing. navigation: information architecture what is the field that deals with optimizing ads?

    Read the article

  • How to Capitalize on Traffic From Google Images

    A little used technique in SEO is optimizing your site's photos for Google images. Most people either don't know of this or are simply too lazy to do it. You can dramatically increase your traffic by optimizing the images on your site for high rank on Google images. Why not pick up this extra traffic?

    Read the article

  • Why Are Inbound Links Important to My Online Identity?

    I have to admit--I'm hooked on Website Grader by HubSpot. The information I get on optimizing my website is pretty cool. I had never configured a 301 redirect until I submitted my website for a grade. For a free website, the advice you receive on optimizing your website is pretty fantastic!

    Read the article

  • SEO - The Most Important Aspect of Internet Marketing Strategy

    Your online visibility and trafficking can be improved by SEO which means Search Engine Optimization. It's a process of improving the quality or traffic to your website or web page. This active practice of changing and optimizing internal as well as external aspects in order to increase the traffic and hits on your website or web page is called optimizing your search result.

    Read the article

  • Creating the Best SEO Content

    SEO, or search engine optimization, encompasses optimizing the volume and quality of a website's content and can even include the optimizing of the website's overall layout. SEO efforts are implemented to increase the search engine page ranking of a website and that website's web pages.

    Read the article

  • Creating Engaging Online Experiences is Easy and Intuitive for Marketers with Oracle WebCenter Sites 11g

    - by Christie Flanagan
    Last month, we announced the availability of Oracle WebCenter Sites 11g, the latest release of our web experience management solution. This new release is really geared toward enabling marketers and business users to drive customer acquisition and brand loyalty by simplifying the whole process of creating, managing and optimizing engaging online experiences.  To show you just how this works, we’ve created the video below which takes you through the tasks a typical marketer might execute using Oracle WebCenter Sites to manage their online presence -- everything from page editing to page creation, right on through to optimizing the mobile experience and moderating user-generated comments and reviews is covered here. I hope this video has give you a flavor for just how easy and intuitive it is for marketers and other business users to manage engaging and interactive online experiences using Oracle WebCenter Sites.  To see more about the new release, please check out the recording of our launch webcast. On Demand Webcast - Introducing Oracle WebCenter Sites: Transforming the Online Experience Enabling marketers and business users is a key requirement for creating and managing contextually relevant, social, and interactive online experiences. Oracle WebCenter Sites transforms the online experience into one that is simple and intuitive to manage as a content contributor, encourages interaction between site visitors and their social networks, and provides marketers with automated targeting options for optimizing online engagement. View this webcast now to learn more.

    Read the article

  • Good book/resource recommendation for HTML5 mobile game development?

    - by Greg Bala
    The problem: I am taking an existing, 5 year old, html based MMORTS game and "HTML5-ing" it, "AJAX-ing" it and most importantly, optimizing for mobile devices like iPhone, android etc. For these devices, the application will be packaged as a downloadable app that is a wrapper for a browser which actually shows the game.. The Question Looking for a good book, or books, or in-depth articles that would help me learn: what tools I have in iOS, andriod applications for optimizing an html based game. things like caching of images, etc what kind of connectivity, or interactivity I can expect between the html/javascript pages and the wrapper - can I play sounds in the wrapper by triggering them from javascript? etc tip and tricks to optimize html/html5 & Javascript application to run well on mobile devices ETC :) Any recommendations would be greatly appreciated!!

    Read the article

  • Best practices when loading images for improving page loading speed

    - by Naoise Golden
    I am working on optimizing a page's loading speed. Here are some analytics: Notice how the images, although only accounting for 65% of the total size (1.1MB), are by far the slowest loading assets: 96% of time. I'd like to know which are the recommended practices on optimizing loading speed, only taking images into account. Some of the techniques we are already applying: image compression images hosted on cookieless domain and CDN spriting everything that can be sprited http headers: keep alive and Expires to one year. Disclaimer: I have gone through the available documentation, I think by focusing on image loading optimization I am not creating a duplicate or a subjective question.

    Read the article

  • Full-text indexing? You must read this

    - by Kyle Hatlestad
    For those of you who may have missed it, Peter Flies, Principal Technical Support Engineer for WebCenter Content, gave an excellent webcast on database searching and indexing in WebCenter Content.  It's available for replay along with a download of the slidedeck.  Look for the one titled 'WebCenter Content: Database Searching and Indexing'. One of the items he led with...and concluded with...was a recommendation on optimizing your search collection if you are using full-text searching with the Oracle database.  This can greatly improve your search performance.  And this would apply to both Oracle Text Search and DATABASE.FULLTEXT search methods.  Peter describes how a collection can become fragmented over time as content is added, updated, and deleted.  Just like you should defragment your hard drive from time to time to get your files placed on the disk in the most optimal way, you should do the same for the search collection. And optimizing the collection is just a simple procedure call that can be scheduled to be run automatically.   [Read more] 

    Read the article

  • Full-text indexing? You must read this

    - by Kyle Hatlestad
    For those of you who may have missed it, Peter Flies, Principal Technical Support Engineer for WebCenter Content, gave an excellent webcast on database searching and indexing in WebCenter Content.  It's available for replay along with a download of the slidedeck.  Look for the one titled 'WebCenter Content: Database Searching and Indexing'. One of the items he led with...and concluded with...was a recommendation on optimizing your search collection if you are using full-text searching with the Oracle database.  This can greatly improve your search performance.  And this would apply to both Oracle Text Search and DATABASE.FULLTEXT search methods.  Peter describes how a collection can become fragmented over time as content is added, updated, and deleted.  Just like you should defragment your hard drive from time to time to get your files placed on the disk in the most optimal way, you should do the same for the search collection. And optimizing the collection is just a simple procedure call that can be scheduled to be run automatically.   beginctx_ddl.optimize_index('FT_IDCTEXT1','FULL', parallel_degree =>'1');end; When I checked my own test instance, I found my collection had a row fragmentation of about 80% After running the optimization procedure, it went down to 0% The knowledgebase article On Index Fragmentation and Optimization When Using OracleTextSearch or DATABASE.FULLTEXT [ID 1087777.1] goes into detail on how to check your current index fragmentation, how to run the procedure, and then how to schedule the procedure to run automatically.  While the article mentions scheduling the job weekly, Peter says he now is recommending this be run daily, especially on more active systems. And just as a reminder, be sure to involve your DBA with your WebCenter Content implementation as you go to production and over time.  We recently had a customer complain of slow performance of the application when it was discovered the database was starving for memory.  So it's always helpful to keep a watchful eye on your database.

    Read the article

  • EclipseCon 2011

    - by Marcus Hirt
    I sadly could not make it to EclipseCon last year. It was sad for so many reasons, not the least being that Sweden during that part of the year is cold and dark. ;) This year, however, I will be contributing two talks: ---> HotRockit – What to Expect from Oracle’s Converged JVM Oracle is converging the HotSpot and JRockit JVMs to produce a "best of breed JVM". Internally the project is sometimes referred to as the HotRockit project. There is already a large influx of ideas and solutions provided by the JRockit JVM into the Open JDK. Examples of improvements include: Better monitoring and profiling Improved performance Better ergonomics This talk will discuss what to expect from the converged JVM over the next two years, and how this will benefit the Eclipse community. Production-time Problem Solving in Eclipse This session will look at some common problems and pitfalls in Java applications. The focus will be on non-invasive profiling and diagnostics of running production systems. Problems tackled will be: Excessive GC Finding hotspots and optimizing them Optimizing the choice of data structures Synchronization problems Finding out where exceptions are thrown Finding memory leaks All problems will be demonstrated and solved running both the bad-behaving applications and the tools to analyze them from within the Eclipse Java IDE. <--- I hope to meet you there!

    Read the article

  • AT&T Application Resource Analyzer in NetBeans IDE

    - by Geertjan
    Here at Øredev in Malmö I met Doug Sillars who does developer outreach for the AT&T Application Resource Optimizer. In this YouTube clip you see Doug explaining how it works and what it can do for optimizing performance of mobile applications. There's a free and open source Android app on GitHub that you can install on Android to collect data and then there's a Java Swing application for analyzing the results. And here's what that application looks like as a plugin in NetBeans IDE, click to enlarge the image, which shows the Android sources of the Data Collector, as well as the Data Analyzer ready to be used to collect data: Since the ARO Data Analyzer is written in Java and has JPanels defining its UI layer, integrating the user interface wasn't hard. Now working on the Actions, so there'll be a new ARO menu with start/stop data collecting menu items, etc, reusing as much of the original code as possible. That part is actually already working. I started up an Android emulator, then started the data collection process from the IDE. Now need to include the Actions for importing the data into the analyzer, together with a few other related features. A pretty cool feature in ARO is video capture, so that a movie can be made by ARO of all the steps taken on the device during the collection process, which will also be nice to have integrated into the NetBeans plugin. Ultimately, this will be handy for anyone creating Android applications in NetBeans IDE since they'll be able to use AT&T's ARO tool for optimizing the performance of the applications they're developing. It will also be useful for those using the built-in Cordova tools in NetBeans IDE to create iOS applications because ARO is also applicable to analyzing iOS application performance.

    Read the article

  • How John Got 15x Improvement Without Really Trying

    - by rchrd
    The following article was published on a Sun Microsystems website a number of years ago by John Feo. It is still useful and worth preserving. So I'm republishing it here.  How I Got 15x Improvement Without Really Trying John Feo, Sun Microsystems Taking ten "personal" program codes used in scientific and engineering research, the author was able to get from 2 to 15 times performance improvement easily by applying some simple general optimization techniques. Introduction Scientific research based on computer simulation depends on the simulation for advancement. The research can advance only as fast as the computational codes can execute. The codes' efficiency determines both the rate and quality of results. In the same amount of time, a faster program can generate more results and can carry out a more detailed simulation of physical phenomena than a slower program. Highly optimized programs help science advance quickly and insure that monies supporting scientific research are used as effectively as possible. Scientific computer codes divide into three broad categories: ISV, community, and personal. ISV codes are large, mature production codes developed and sold commercially. The codes improve slowly over time both in methods and capabilities, and they are well tuned for most vendor platforms. Since the codes are mature and complex, there are few opportunities to improve their performance solely through code optimization. Improvements of 10% to 15% are typical. Examples of ISV codes are DYNA3D, Gaussian, and Nastran. Community codes are non-commercial production codes used by a particular research field. Generally, they are developed and distributed by a single academic or research institution with assistance from the community. Most users just run the codes, but some develop new methods and extensions that feed back into the general release. The codes are available on most vendor platforms. Since these codes are younger than ISV codes, there are more opportunities to optimize the source code. Improvements of 50% are not unusual. Examples of community codes are AMBER, CHARM, BLAST, and FASTA. Personal codes are those written by single users or small research groups for their own use. These codes are not distributed, but may be passed from professor-to-student or student-to-student over several years. They form the primordial ocean of applications from which community and ISV codes emerge. Government research grants pay for the development of most personal codes. This paper reports on the nature and performance of this class of codes. Over the last year, I have looked at over two dozen personal codes from more than a dozen research institutions. The codes cover a variety of scientific fields, including astronomy, atmospheric sciences, bioinformatics, biology, chemistry, geology, and physics. The sources range from a few hundred lines to more than ten thousand lines, and are written in Fortran, Fortran 90, C, and C++. For the most part, the codes are modular, documented, and written in a clear, straightforward manner. They do not use complex language features, advanced data structures, programming tricks, or libraries. I had little trouble understanding what the codes did or how data structures were used. Most came with a makefile. Surprisingly, only one of the applications is parallel. All developers have access to parallel machines, so availability is not an issue. Several tried to parallelize their applications, but stopped after encountering difficulties. Lack of education and a perception that parallelism is difficult prevented most from trying. I parallelized several of the codes using OpenMP, and did not judge any of the codes as difficult to parallelize. Even more surprising than the lack of parallelism is the inefficiency of the codes. I was able to get large improvements in performance in a matter of a few days applying simple optimization techniques. Table 1 lists ten representative codes [names and affiliation are omitted to preserve anonymity]. Improvements on one processor range from 2x to 15.5x with a simple average of 4.75x. I did not use sophisticated performance tools or drill deep into the program's execution character as one would do when tuning ISV or community codes. Using only a profiler and source line timers, I identified inefficient sections of code and improved their performance by inspection. The changes were at a high level. I am sure there is another factor of 2 or 3 in each code, and more if the codes are parallelized. The study’s results show that personal scientific codes are running many times slower than they should and that the problem is pervasive. Computational scientists are not sloppy programmers; however, few are trained in the art of computer programming or code optimization. I found that most have a working knowledge of some programming language and standard software engineering practices; but they do not know, or think about, how to make their programs run faster. They simply do not know the standard techniques used to make codes run faster. In fact, they do not even perceive that such techniques exist. The case studies described in this paper show that applying simple, well known techniques can significantly increase the performance of personal codes. It is important that the scientific community and the Government agencies that support scientific research find ways to better educate academic scientific programmers. The inefficiency of their codes is so bad that it is retarding both the quality and progress of scientific research. # cacheperformance redundantoperations loopstructures performanceimprovement 1 x x 15.5 2 x 2.8 3 x x 2.5 4 x 2.1 5 x x 2.0 6 x 5.0 7 x 5.8 8 x 6.3 9 2.2 10 x x 3.3 Table 1 — Area of improvement and performance gains of 10 codes The remainder of the paper is organized as follows: sections 2, 3, and 4 discuss the three most common sources of inefficiencies in the codes studied. These are cache performance, redundant operations, and loop structures. Each section includes several examples. The last section summaries the work and suggests a possible solution to the issues raised. Optimizing cache performance Commodity microprocessor systems use caches to increase memory bandwidth and reduce memory latencies. Typical latencies from processor to L1, L2, local, and remote memory are 3, 10, 50, and 200 cycles, respectively. Moreover, bandwidth falls off dramatically as memory distances increase. Programs that do not use cache effectively run many times slower than programs that do. When optimizing for cache, the biggest performance gains are achieved by accessing data in cache order and reusing data to amortize the overhead of cache misses. Secondary considerations are prefetching, associativity, and replacement; however, the understanding and analysis required to optimize for the latter are probably beyond the capabilities of the non-expert. Much can be gained simply by accessing data in the correct order and maximizing data reuse. 6 out of the 10 codes studied here benefited from such high level optimizations. Array Accesses The most important cache optimization is the most basic: accessing Fortran array elements in column order and C array elements in row order. Four of the ten codes—1, 2, 4, and 10—got it wrong. Compilers will restructure nested loops to optimize cache performance, but may not do so if the loop structure is too complex, or the loop body includes conditionals, complex addressing, or function calls. In code 1, the compiler failed to invert a key loop because of complex addressing do I = 0, 1010, delta_x IM = I - delta_x IP = I + delta_x do J = 5, 995, delta_x JM = J - delta_x JP = J + delta_x T1 = CA1(IP, J) + CA1(I, JP) T2 = CA1(IM, J) + CA1(I, JM) S1 = T1 + T2 - 4 * CA1(I, J) CA(I, J) = CA1(I, J) + D * S1 end do end do In code 2, the culprit is conditionals do I = 1, N do J = 1, N If (IFLAG(I,J) .EQ. 0) then T1 = Value(I, J-1) T2 = Value(I-1, J) T3 = Value(I, J) T4 = Value(I+1, J) T5 = Value(I, J+1) Value(I,J) = 0.25 * (T1 + T2 + T5 + T4) Delta = ABS(T3 - Value(I,J)) If (Delta .GT. MaxDelta) MaxDelta = Delta endif enddo enddo I fixed both programs by inverting the loops by hand. Code 10 has three-dimensional arrays and triply nested loops. The structure of the most computationally intensive loops is too complex to invert automatically or by hand. The only practical solution is to transpose the arrays so that the dimension accessed by the innermost loop is in cache order. The arrays can be transposed at construction or prior to entering a computationally intensive section of code. The former requires all array references to be modified, while the latter is cost effective only if the cost of the transpose is amortized over many accesses. I used the second approach to optimize code 10. Code 5 has four-dimensional arrays and loops are nested four deep. For all of the reasons cited above the compiler is not able to restructure three key loops. Assume C arrays and let the four dimensions of the arrays be i, j, k, and l. In the original code, the index structure of the three loops is L1: for i L2: for i L3: for i for l for l for j for k for j for k for j for k for l So only L3 accesses array elements in cache order. L1 is a very complex loop—much too complex to invert. I brought the loop into cache alignment by transposing the second and fourth dimensions of the arrays. Since the code uses a macro to compute all array indexes, I effected the transpose at construction and changed the macro appropriately. The dimensions of the new arrays are now: i, l, k, and j. L3 is a simple loop and easily inverted. L2 has a loop-carried scalar dependence in k. By promoting the scalar name that carries the dependence to an array, I was able to invert the third and fourth subloops aligning the loop with cache. Code 5 is by far the most difficult of the four codes to optimize for array accesses; but the knowledge required to fix the problems is no more than that required for the other codes. I would judge this code at the limits of, but not beyond, the capabilities of appropriately trained computational scientists. Array Strides When a cache miss occurs, a line (64 bytes) rather than just one word is loaded into the cache. If data is accessed stride 1, than the cost of the miss is amortized over 8 words. Any stride other than one reduces the cost savings. Two of the ten codes studied suffered from non-unit strides. The codes represent two important classes of "strided" codes. Code 1 employs a multi-grid algorithm to reduce time to convergence. The grids are every tenth, fifth, second, and unit element. Since time to convergence is inversely proportional to the distance between elements, coarse grids converge quickly providing good starting values for finer grids. The better starting values further reduce the time to convergence. The downside is that grids of every nth element, n > 1, introduce non-unit strides into the computation. In the original code, much of the savings of the multi-grid algorithm were lost due to this problem. I eliminated the problem by compressing (copying) coarse grids into continuous memory, and rewriting the computation as a function of the compressed grid. On convergence, I copied the final values of the compressed grid back to the original grid. The savings gained from unit stride access of the compressed grid more than paid for the cost of copying. Using compressed grids, the loop from code 1 included in the previous section becomes do j = 1, GZ do i = 1, GZ T1 = CA(i+0, j-1) + CA(i-1, j+0) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) S1 = T1 + T4 - 4 * CA1(i+0, j+0) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 enddo enddo where CA and CA1 are compressed arrays of size GZ. Code 7 traverses a list of objects selecting objects for later processing. The labels of the selected objects are stored in an array. The selection step has unit stride, but the processing steps have irregular stride. A fix is to save the parameters of the selected objects in temporary arrays as they are selected, and pass the temporary arrays to the processing functions. The fix is practical if the same parameters are used in selection as in processing, or if processing comprises a series of distinct steps which use overlapping subsets of the parameters. Both conditions are true for code 7, so I achieved significant improvement by copying parameters to temporary arrays during selection. Data reuse In the previous sections, we optimized for spatial locality. It is also important to optimize for temporal locality. Once read, a datum should be used as much as possible before it is forced from cache. Loop fusion and loop unrolling are two techniques that increase temporal locality. Unfortunately, both techniques increase register pressure—as loop bodies become larger, the number of registers required to hold temporary values grows. Once register spilling occurs, any gains evaporate quickly. For multiprocessors with small register sets or small caches, the sweet spot can be very small. In the ten codes presented here, I found no opportunities for loop fusion and only two opportunities for loop unrolling (codes 1 and 3). In code 1, unrolling the outer and inner loop one iteration increases the number of result values computed by the loop body from 1 to 4, do J = 1, GZ-2, 2 do I = 1, GZ-2, 2 T1 = CA1(i+0, j-1) + CA1(i-1, j+0) T2 = CA1(i+1, j-1) + CA1(i+0, j+0) T3 = CA1(i+0, j+0) + CA1(i-1, j+1) T4 = CA1(i+1, j+0) + CA1(i+0, j+1) T5 = CA1(i+2, j+0) + CA1(i+1, j+1) T6 = CA1(i+1, j+1) + CA1(i+0, j+2) T7 = CA1(i+2, j+1) + CA1(i+1, j+2) S1 = T1 + T4 - 4 * CA1(i+0, j+0) S2 = T2 + T5 - 4 * CA1(i+1, j+0) S3 = T3 + T6 - 4 * CA1(i+0, j+1) S4 = T4 + T7 - 4 * CA1(i+1, j+1) CA(i+0, j+0) = CA1(i+0, j+0) + DD * S1 CA(i+1, j+0) = CA1(i+1, j+0) + DD * S2 CA(i+0, j+1) = CA1(i+0, j+1) + DD * S3 CA(i+1, j+1) = CA1(i+1, j+1) + DD * S4 enddo enddo The loop body executes 12 reads, whereas as the rolled loop shown in the previous section executes 20 reads to compute the same four values. In code 3, two loops are unrolled 8 times and one loop is unrolled 4 times. Here is the before for (k = 0; k < NK[u]; k++) { sum = 0.0; for (y = 0; y < NY; y++) { sum += W[y][u][k] * delta[y]; } backprop[i++]=sum; } and after code for (k = 0; k < KK - 8; k+=8) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (y = 0; y < NY; y++) { sum0 += W[y][0][k+0] * delta[y]; sum1 += W[y][0][k+1] * delta[y]; sum2 += W[y][0][k+2] * delta[y]; sum3 += W[y][0][k+3] * delta[y]; sum4 += W[y][0][k+4] * delta[y]; sum5 += W[y][0][k+5] * delta[y]; sum6 += W[y][0][k+6] * delta[y]; sum7 += W[y][0][k+7] * delta[y]; } backprop[k+0] = sum0; backprop[k+1] = sum1; backprop[k+2] = sum2; backprop[k+3] = sum3; backprop[k+4] = sum4; backprop[k+5] = sum5; backprop[k+6] = sum6; backprop[k+7] = sum7; } for one of the loops unrolled 8 times. Optimizing for temporal locality is the most difficult optimization considered in this paper. The concepts are not difficult, but the sweet spot is small. Identifying where the program can benefit from loop unrolling or loop fusion is not trivial. Moreover, it takes some effort to get it right. Still, educating scientific programmers about temporal locality and teaching them how to optimize for it will pay dividends. Reducing instruction count Execution time is a function of instruction count. Reduce the count and you usually reduce the time. The best solution is to use a more efficient algorithm; that is, an algorithm whose order of complexity is smaller, that converges quicker, or is more accurate. Optimizing source code without changing the algorithm yields smaller, but still significant, gains. This paper considers only the latter because the intent is to study how much better codes can run if written by programmers schooled in basic code optimization techniques. The ten codes studied benefited from three types of "instruction reducing" optimizations. The two most prevalent were hoisting invariant memory and data operations out of inner loops. The third was eliminating unnecessary data copying. The nature of these inefficiencies is language dependent. Memory operations The semantics of C make it difficult for the compiler to determine all the invariant memory operations in a loop. The problem is particularly acute for loops in functions since the compiler may not know the values of the function's parameters at every call site when compiling the function. Most compilers support pragmas to help resolve ambiguities; however, these pragmas are not comprehensive and there is no standard syntax. To guarantee that invariant memory operations are not executed repetitively, the user has little choice but to hoist the operations by hand. The problem is not as severe in Fortran programs because in the absence of equivalence statements, it is a violation of the language's semantics for two names to share memory. Codes 3 and 5 are C programs. In both cases, the compiler did not hoist all invariant memory operations from inner loops. Consider the following loop from code 3 for (y = 0; y < NY; y++) { i = 0; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += delta[y] * I1[i++]; } } } Since dW[y][u] can point to the same memory space as delta for one or more values of y and u, assignment to dW[y][u][k] may change the value of delta[y]. In reality, dW and delta do not overlap in memory, so I rewrote the loop as for (y = 0; y < NY; y++) { i = 0; Dy = delta[y]; for (u = 0; u < NU; u++) { for (k = 0; k < NK[u]; k++) { dW[y][u][k] += Dy * I1[i++]; } } } Failure to hoist invariant memory operations may be due to complex address calculations. If the compiler can not determine that the address calculation is invariant, then it can hoist neither the calculation nor the associated memory operations. As noted above, code 5 uses a macro to address four-dimensional arrays #define MAT4D(a,q,i,j,k) (double *)((a)->data + (q)*(a)->strides[0] + (i)*(a)->strides[3] + (j)*(a)->strides[2] + (k)*(a)->strides[1]) The macro is too complex for the compiler to understand and so, it does not identify any subexpressions as loop invariant. The simplest way to eliminate the address calculation from the innermost loop (over i) is to define a0 = MAT4D(a,q,0,j,k) before the loop and then replace all instances of *MAT4D(a,q,i,j,k) in the loop with a0[i] A similar problem appears in code 6, a Fortran program. The key loop in this program is do n1 = 1, nh nx1 = (n1 - 1) / nz + 1 nz1 = n1 - nz * (nx1 - 1) do n2 = 1, nh nx2 = (n2 - 1) / nz + 1 nz2 = n2 - nz * (nx2 - 1) ndx = nx2 - nx1 ndy = nz2 - nz1 gxx = grn(1,ndx,ndy) gyy = grn(2,ndx,ndy) gxy = grn(3,ndx,ndy) balance(n1,1) = balance(n1,1) + (force(n2,1) * gxx + force(n2,2) * gxy) * h1 balance(n1,2) = balance(n1,2) + (force(n2,1) * gxy + force(n2,2) * gyy)*h1 end do end do The programmer has written this loop well—there are no loop invariant operations with respect to n1 and n2. However, the loop resides within an iterative loop over time and the index calculations are independent with respect to time. Trading space for time, I precomputed the index values prior to the entering the time loop and stored the values in two arrays. I then replaced the index calculations with reads of the arrays. Data operations Ways to reduce data operations can appear in many forms. Implementing a more efficient algorithm produces the biggest gains. The closest I came to an algorithm change was in code 4. This code computes the inner product of K-vectors A(i) and B(j), 0 = i < N, 0 = j < M, for most values of i and j. Since the program computes most of the NM possible inner products, it is more efficient to compute all the inner products in one triply-nested loop rather than one at a time when needed. The savings accrue from reading A(i) once for all B(j) vectors and from loop unrolling. for (i = 0; i < N; i+=8) { for (j = 0; j < M; j++) { sum0 = 0.0; sum1 = 0.0; sum2 = 0.0; sum3 = 0.0; sum4 = 0.0; sum5 = 0.0; sum6 = 0.0; sum7 = 0.0; for (k = 0; k < K; k++) { sum0 += A[i+0][k] * B[j][k]; sum1 += A[i+1][k] * B[j][k]; sum2 += A[i+2][k] * B[j][k]; sum3 += A[i+3][k] * B[j][k]; sum4 += A[i+4][k] * B[j][k]; sum5 += A[i+5][k] * B[j][k]; sum6 += A[i+6][k] * B[j][k]; sum7 += A[i+7][k] * B[j][k]; } C[i+0][j] = sum0; C[i+1][j] = sum1; C[i+2][j] = sum2; C[i+3][j] = sum3; C[i+4][j] = sum4; C[i+5][j] = sum5; C[i+6][j] = sum6; C[i+7][j] = sum7; }} This change requires knowledge of a typical run; i.e., that most inner products are computed. The reasons for the change, however, derive from basic optimization concepts. It is the type of change easily made at development time by a knowledgeable programmer. In code 5, we have the data version of the index optimization in code 6. Here a very expensive computation is a function of the loop indices and so cannot be hoisted out of the loop; however, the computation is invariant with respect to an outer iterative loop over time. We can compute its value for each iteration of the computation loop prior to entering the time loop and save the values in an array. The increase in memory required to store the values is small in comparison to the large savings in time. The main loop in Code 8 is doubly nested. The inner loop includes a series of guarded computations; some are a function of the inner loop index but not the outer loop index while others are a function of the outer loop index but not the inner loop index for (j = 0; j < N; j++) { for (i = 0; i < M; i++) { r = i * hrmax; R = A[j]; temp = (PRM[3] == 0.0) ? 1.0 : pow(r, PRM[3]); high = temp * kcoeff * B[j] * PRM[2] * PRM[4]; low = high * PRM[6] * PRM[6] / (1.0 + pow(PRM[4] * PRM[6], 2.0)); kap = (R > PRM[6]) ? high * R * R / (1.0 + pow(PRM[4]*r, 2.0) : low * pow(R/PRM[6], PRM[5]); < rest of loop omitted > }} Note that the value of temp is invariant to j. Thus, we can hoist the computation for temp out of the loop and save its values in an array. for (i = 0; i < M; i++) { r = i * hrmax; TEMP[i] = pow(r, PRM[3]); } [N.B. – the case for PRM[3] = 0 is omitted and will be reintroduced later.] We now hoist out of the inner loop the computations invariant to i. Since the conditional guarding the value of kap is invariant to i, it behooves us to hoist the computation out of the inner loop, thereby executing the guard once rather than M times. The final version of the code is for (j = 0; j < N; j++) { R = rig[j] / 1000.; tmp1 = kcoeff * par[2] * beta[j] * par[4]; tmp2 = 1.0 + (par[4] * par[4] * par[6] * par[6]); tmp3 = 1.0 + (par[4] * par[4] * R * R); tmp4 = par[6] * par[6] / tmp2; tmp5 = R * R / tmp3; tmp6 = pow(R / par[6], par[5]); if ((par[3] == 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp5; } else if ((par[3] == 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * tmp4 * tmp6; } else if ((par[3] != 0.0) && (R > par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp5; } else if ((par[3] != 0.0) && (R <= par[6])) { for (i = 1; i <= imax1; i++) KAP[i] = tmp1 * TEMP[i] * tmp4 * tmp6; } for (i = 0; i < M; i++) { kap = KAP[i]; r = i * hrmax; < rest of loop omitted > } } Maybe not the prettiest piece of code, but certainly much more efficient than the original loop, Copy operations Several programs unnecessarily copy data from one data structure to another. This problem occurs in both Fortran and C programs, although it manifests itself differently in the two languages. Code 1 declares two arrays—one for old values and one for new values. At the end of each iteration, the array of new values is copied to the array of old values to reset the data structures for the next iteration. This problem occurs in Fortran programs not included in this study and in both Fortran 77 and Fortran 90 code. Introducing pointers to the arrays and swapping pointer values is an obvious way to eliminate the copying; but pointers is not a feature that many Fortran programmers know well or are comfortable using. An easy solution not involving pointers is to extend the dimension of the value array by 1 and use the last dimension to differentiate between arrays at different times. For example, if the data space is N x N, declare the array (N, N, 2). Then store the problem’s initial values in (_, _, 2) and define the scalar names new = 2 and old = 1. At the start of each iteration, swap old and new to reset the arrays. The old–new copy problem did not appear in any C program. In programs that had new and old values, the code swapped pointers to reset data structures. Where unnecessary coping did occur is in structure assignment and parameter passing. Structures in C are handled much like scalars. Assignment causes the data space of the right-hand name to be copied to the data space of the left-hand name. Similarly, when a structure is passed to a function, the data space of the actual parameter is copied to the data space of the formal parameter. If the structure is large and the assignment or function call is in an inner loop, then copying costs can grow quite large. While none of the ten programs considered here manifested this problem, it did occur in programs not included in the study. A simple fix is always to refer to structures via pointers. Optimizing loop structures Since scientific programs spend almost all their time in loops, efficient loops are the key to good performance. Conditionals, function calls, little instruction level parallelism, and large numbers of temporary values make it difficult for the compiler to generate tightly packed, highly efficient code. Conditionals and function calls introduce jumps that disrupt code flow. Users should eliminate or isolate conditionls to their own loops as much as possible. Often logical expressions can be substituted for if-then-else statements. For example, code 2 includes the following snippet MaxDelta = 0.0 do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) if (Delta > MaxDelta) MaxDelta = Delta enddo enddo if (MaxDelta .gt. 0.001) goto 200 Since the only use of MaxDelta is to control the jump to 200 and all that matters is whether or not it is greater than 0.001, I made MaxDelta a boolean and rewrote the snippet as MaxDelta = .false. do J = 1, N do I = 1, M < code omitted > Delta = abs(OldValue ? NewValue) MaxDelta = MaxDelta .or. (Delta .gt. 0.001) enddo enddo if (MaxDelta) goto 200 thereby, eliminating the conditional expression from the inner loop. A microprocessor can execute many instructions per instruction cycle. Typically, it can execute one or more memory, floating point, integer, and jump operations. To be executed simultaneously, the operations must be independent. Thick loops tend to have more instruction level parallelism than thin loops. Moreover, they reduce memory traffice by maximizing data reuse. Loop unrolling and loop fusion are two techniques to increase the size of loop bodies. Several of the codes studied benefitted from loop unrolling, but none benefitted from loop fusion. This observation is not too surpising since it is the general tendency of programmers to write thick loops. As loops become thicker, the number of temporary values grows, increasing register pressure. If registers spill, then memory traffic increases and code flow is disrupted. A thick loop with many temporary values may execute slower than an equivalent series of thin loops. The biggest gain will be achieved if the thick loop can be split into a series of independent loops eliminating the need to write and read temporary arrays. I found such an occasion in code 10 where I split the loop do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do into two disjoint loops do i = 1, n do j = 1, m A24(j,i)= S24(j,i) * T24(j,i) + S25(j,i) * U25(j,i) B24(j,i)= S24(j,i) * T25(j,i) + S25(j,i) * U24(j,i) A25(j,i)= S24(j,i) * C24(j,i) + S25(j,i) * V24(j,i) B25(j,i)= S24(j,i) * U25(j,i) + S25(j,i) * V25(j,i) end do end do do i = 1, n do j = 1, m C24(j,i)= S26(j,i) * T26(j,i) + S27(j,i) * U26(j,i) D24(j,i)= S26(j,i) * T27(j,i) + S27(j,i) * V26(j,i) C25(j,i)= S27(j,i) * S28(j,i) + S26(j,i) * U28(j,i) D25(j,i)= S27(j,i) * T28(j,i) + S26(j,i) * V28(j,i) end do end do Conclusions Over the course of the last year, I have had the opportunity to work with over two dozen academic scientific programmers at leading research universities. Their research interests span a broad range of scientific fields. Except for two programs that relied almost exclusively on library routines (matrix multiply and fast Fourier transform), I was able to improve significantly the single processor performance of all codes. Improvements range from 2x to 15.5x with a simple average of 4.75x. Changes to the source code were at a very high level. I did not use sophisticated techniques or programming tools to discover inefficiencies or effect the changes. Only one code was parallel despite the availability of parallel systems to all developers. Clearly, we have a problem—personal scientific research codes are highly inefficient and not running parallel. The developers are unaware of simple optimization techniques to make programs run faster. They lack education in the art of code optimization and parallel programming. I do not believe we can fix the problem by publishing additional books or training manuals. To date, the developers in questions have not studied the books or manual available, and are unlikely to do so in the future. Short courses are a possible solution, but I believe they are too concentrated to be much use. The general concepts can be taught in a three or four day course, but that is not enough time for students to practice what they learn and acquire the experience to apply and extend the concepts to their codes. Practice is the key to becoming proficient at optimization. I recommend that graduate students be required to take a semester length course in optimization and parallel programming. We would never give someone access to state-of-the-art scientific equipment costing hundreds of thousands of dollars without first requiring them to demonstrate that they know how to use the equipment. Yet the criterion for time on state-of-the-art supercomputers is at most an interesting project. Requestors are never asked to demonstrate that they know how to use the system, or can use the system effectively. A semester course would teach them the required skills. Government agencies that fund academic scientific research pay for most of the computer systems supporting scientific research as well as the development of most personal scientific codes. These agencies should require graduate schools to offer a course in optimization and parallel programming as a requirement for funding. About the Author John Feo received his Ph.D. in Computer Science from The University of Texas at Austin in 1986. After graduate school, Dr. Feo worked at Lawrence Livermore National Laboratory where he was the Group Leader of the Computer Research Group and principal investigator of the Sisal Language Project. In 1997, Dr. Feo joined Tera Computer Company where he was project manager for the MTA, and oversaw the programming and evaluation of the MTA at the San Diego Supercomputer Center. In 2000, Dr. Feo joined Sun Microsystems as an HPC application specialist. He works with university research groups to optimize and parallelize scientific codes. Dr. Feo has published over two dozen research articles in the areas of parallel parallel programming, parallel programming languages, and application performance.

    Read the article

  • How Do You Profile & Optimize CUDA Kernels?

    - by John Dibling
    I am somewhat familiar with the CUDA visual profiler and the occupancy spreadsheet, although I am probably not leveraging them as well as I could. Profiling & optimizing CUDA code is not like profiling & optimizing code that runs on a CPU. So I am hoping to learn from your experiences about how to get the most out of my code. There was a post recently looking for the fastest possible code to identify self numbers, and I provided a CUDA implementation. I'm not satisfied that this code is as fast as it can be, but I'm at a loss as to figure out both what the right questions are and what tool I can get the answers from. How do you identify ways to make your CUDA kernels perform faster?

    Read the article

  • jQuery Optimizations

    - by aepheus
    I've just come to the end of a large development project. We were on a tight timeline, so a lot of optimization was "deferred". Now that we met our deadline, we're going back and trying to optimize things. My questions is this: What are some of the most important things you look for when optimizing jQuery web sites. Alternately I'd love to hear of sites/lists that have particularly good advise for optimizing jQuery. I've already read a few articles, http://www.tvidesign.co.uk/blog/improve-your-jquery-25-excellent-tips.aspx was an especially good read.

    Read the article

  • SEO Courses - Tips and Techniques to Enhance Visibility

    Search Engine Optimization (SEO) is the process of optimizing a web site to achieve a higher ranking when a search engine looks for web pages based on particular word or phrase. The SEO courses empower you with the wherewithal to successfully set up and run an optimization program.

    Read the article

  • Solaris 11 Resources

    - by user12618891
    .. Oracle Solaris 11 (November, 2011) Oracle Solaris 11 Landing Page Download Oracle Solaris 11 Oracle Solaris 11 Documentation Solaris 11 End-of-Life Notices What's New in Oracle Solaris 11 (blog) Oracle Solaris 11 Feature Demo Videos (blog) Solaris 11 Developer Resources (November, 2011) Oracle Solaris 11 ISV Adoption Guide Oracle Solaris 11 Preflight Checker The IPS System Repository (blog) Packaging and Delivering Software with the Image Packaging System - A Developer's Guide How to Create and Publish packages to an IPS Repository on Solaris 11 Solaris 10 Branded zone VM Templates for Solaris 11 (blog) Oracle Solaris 11 Security: What's New For Developers Optimizing Application with Oracle Solaris Studio Tools and CompilersOther Solaris 11 Technology Spotlights (Landing Page)

    Read the article

  • Key Techniques For Search Engine Optimization in 2010

    The art of creating web pages which will rank high in search engine returns is called Search Engine Optimization or SEO. By optimizing certain elements or sections in the HTML code of each page, SEO can be accomplished. The search engines specifically read these sections. The level of optimization can help determine the amount free referral traffic.

    Read the article

  • Daily tech links for .net and related technologies - May 13-16, 2010

    - by SanjeevAgarwal
    Daily tech links for .net and related technologies - May 13-16, 2010 Web Development Integrating Twitter Into An ASP.NET Website Using OAuth - Scott Mitchell T4MVC Extensions for MVC Partials - Evan Building a Data Grid in ASP.NET MVC - Ali Bastani Introducing the MVC Music Store - MVC 2 Sample Application and Tutorial - Jon Galloway Announcing the RTM of MvcExtensions - kazimanzurrashid Optimizing Your Website For Speed Web Design Validation with the jQuery UI Tabs Widget - Chris Love A Brief History...(read more)

    Read the article

  • Create Keyword Dense Content For Better SEO

    Briefly touching on this in the introduction this is important to do and be aware of but do not make this a massive part of your efforts to achieve better search engine ranking. This basically means optimizing your content to be more keyword dense so that the search engine robots will pick up your site as being relevant for a certain search phrase or keyword.

    Read the article

  • search with mobile : does Google just look in the mobile-optimized websites?

    - by Alireza Fallah
    When someone searches a keyword by mobile, does Google search in desktop version of all websites and find the proper result and then prioritize them according to the responsiveness or mobile-optimizing stuff etc OR it just search in the mobile version of the mobile-optimized websites ? I want to create a website with a responsive design, I was wondering that if I should care about SEO in mobile version of the website, or just try to optimize the desktop version for search engines and just care about the design of the mobile version ?

    Read the article

< Previous Page | 5 6 7 8 9 10 11 12 13 14 15 16  | Next Page >