Search Results

Search found 29588 results on 1184 pages for 'really'.

Page 2/1184 | < Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • Bluetooth not really "turning on" on a Sony Vaio VPCEG

    - by marcioAlmada
    On a Sony Vaio VPCEG, bluetooth is not working properly: The gnome-shell bluetooth indicator (at the main bar) shows the bluetooth is on, but when I try to switch visibility on it simply doesn't turn on. When I open the bluetooth configuration dialog, the on/off indicator shows me the bluetooth is off. And when I try to turn it on it simply does not turn on. The following is outputted by lsusb command: Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 001 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 002 Device 002: ID 8087:0024 Intel Corp. Integrated Rate Matching Hub Bus 001 Device 003: ID 0c45:64be Microdia Bus 001 Device 004: ID 0489:e027 Foxconn / Hon Hai Bus 002 Device 014: ID 4971:ce22 SimpleTech ====== Update: Output from rfkill list: 0: phy0: Wireless LAN Soft blocked: no Hard blocked: no 1: sony-wifi: Wireless LAN Soft blocked: no Hard blocked: no 2: sony-bluetooth: Bluetooth Soft blocked: no Hard blocked: no 3: hci0: Bluetooth Soft blocked: no Hard blocked: no What should I do to really activate the bluetooth?

    Read the article

  • Any really decent way to get three monitors?

    - by JohnCC
    What satisfactory solutions are there to achieve three monitors on Ubuntu? I know some ATI cards (Eyefinity) can support 3 monitors from a single card, but I don't know how well this is supported under Linux and besides, I've never had much luck with ATI on Linux. The alternative is to try two cards, but there seem to be problems there too. It looks to me like xrandr cannot support 2 GPUs. I believe you'll end up with two separate "Screens" across which you cannot move applications or windows, unless you enable Xinerama which as I understand it disables some acceleration and probably compositing too. I've found so much conflicting information on this online, I'm really confused. Please advise!

    Read the article

  • Firefox really slows down my computer

    - by user199208
    I love Firefox; it has a special place in my heart. However, using the default Firefox in Ubuntu eventually causes my computer to freeze and crash. Nothing more to the problem, just the performance of Firefox really sucks, therefor causing a system-wide freeze. How do I know this? I use Google Chrome mostly, and other programs. None of them ever made my computer freeze. Is it an Update problem? If so, can I install a different version of Firefox from Mozilla.com? What's the deal here?

    Read the article

  • Sorting a REALLY BIG delimited text file in UNIX / VMS [closed]

    - by gunbuster363
    Hi everyone, I am going to sort a REALLY BIG delimited text file, say 250Mb (or a bunch of files of more or less than 250Mb) . It have 37 fields, and I need to sort it by 5 fields, for example 1st, 4th, 5th, 6th 7th fields. Under Unix / VMS, do I have a good option to do this FAST? I can write COBOL program. Now I am trying to sort them using the below command, but it already run for a long time and just not going to finished. Thank you. The command I used: time sort -t ',' -o sorted.txt +0 -1 +4 -5 +5 -6 +6 -7 +22 -23 *.DAT_gprscdr_ftpd

    Read the article

  • how to really master a programming language

    - by cprogcr
    I know that learning a language, you can simply buy a book, follow the examples, and whenever possible try the exercises. But what I'm really looking is how to master the language once you've learned it. Now I know that experience is one major factor, but what about learning the internals of the language, what is the underlying structure, etc. There are articles out there saying read this book, read that book, make this game and that game. But to me this doesn't mean to master a language. I want to be able to read other people's code and understand it, no matter how hard that is. To understand when to use a function and when another, etc etc. The list could go on and on but I believe I've made the point. :) And finally, take whatever language as an example if needed, though best would be if C was taken as an example.

    Read the article

  • How can I really master a programming language?

    - by cprogcr
    I know that learning a language, you can simply buy a book, follow the examples, and whenever possible try the exercises. But what I'm really looking is how to master the language once you've learned it. Now I know that experience is one major factor, but what about learning the internals of the language, what is the underlying structure, etc. There are articles out there saying read this book, read that book, make this game and that game. But to me this doesn't mean to master a language. I want to be able to read other people's code and understand it, no matter how hard that is. To understand when to use a function and when another, etc etc. The list could go on and on but I believe I've made the point. :) And finally, take whatever language as an example if needed, though best would be if C was taken as an example.

    Read the article

  • is OpenID really that bad?

    - by DoPPler
    I have seen this question on Quora where lots of people seem to agree that OpenID is bad, even going as far as stating that: OpenID is the worst possible "solution" I have ever seen in my entire life to a problem that most people don't really have Then I've seen articles and tweets referencing that question saying that OpenID has lost, and Facebook won. It's sad to read as I quite like the OpenID (or at least idea behind it). I literally hate getting yet another login/password for page (I'll forget it anyway) - it's a pretty serious issue for me and I know lots of people with the same problem. Thus I thought that OpenId is a great solution but I'm not sure anymore. So the question is should I still bother to implement OpenID or it's not worth it? What is the most robust and convenient (from the user perspective) way to identify and authenticate an user?

    Read the article

  • Ubuntu 13.04 running really slow and Hanging

    - by CAM
    Up till recently I have been running 13.04 on my laptop very happily. This morning however, I turned on my laptop to find it running really slow. Takes 5 min to load a program and even then the program freezes and I have had 3 system hangs this morning already. The Unity Desktop appears to run ok but programs do not. Things I have tried so far: Checking for Propitiatory graphics drivers - none shown available (I have bumblebee running already). Using the recovery boot options from Grub to repair broken packages. Recent changes - Updated computer, Installed some indicator applets which have worked fine for me before. System Specs: Asus U36s, Intel Core i5-2450M 2.5GHz, 4GB RAM, Nvidia Geforce 610M-1GB, Dual boot Win7 & Ubuntu 13.04 I'm a bit of a noob with Ubuntu but am happy enough running stuff in terminal if you will advise me on what to run. I'm just a bit stuck on what do to fix this without a reinstall. Thanks a lot for your help.

    Read the article

  • How to Upload Really Large Files to SkyDrive, Dropbox, or Email

    - by Matthew Guay
    Do you need to upload a very large file to store online or email to a friend? Unfortunately, whether you’re emailing a file or using online storage sites like SkyDrive, there’s a limit on the size of files you can use. Here’s how to get around the limits. Skydrive only lets you add files up to 50 MB, and while the Dropbox desktop client lets you add really large files, the web interface has a 300 MB limit, so if you were on another PC and wanted to add giant files to your Dropbox, you’d need to split them. This same technique also works for any file sharing service—even if you were sending files through email. There’s two ways that you can get around the limits—first, by just compressing the files if you’re close to the limit, but the second and more interesting way is to split up the files into smaller chunks. Keep reading for how to do both. Latest Features How-To Geek ETC The How-To Geek Guide to Learning Photoshop, Part 8: Filters Get the Complete Android Guide eBook for Only 99 Cents [Update: Expired] Improve Digital Photography by Calibrating Your Monitor The How-To Geek Guide to Learning Photoshop, Part 7: Design and Typography How to Choose What to Back Up on Your Linux Home Server How To Harmonize Your Dual-Boot Setup for Windows and Ubuntu Hang in There Scrat! – Ice Age Wallpaper How Do You Know When You’ve Passed Geek and Headed to Nerd? On The Tip – A Lamborghini Theme for Chrome and Iron What if Wile E. Coyote and the Road Runner were Human? [Video] Peaceful Winter Cabin Wallpaper Store Tabs for Later Viewing in Opera with Tab Vault

    Read the article

  • Does TDD really work for complex projects?

    - by Amir Rezaei
    I’m asking this question regarding problems I have experienced during TDD projects. I have noticed the following challenges when creating unit tests. Generating and maintaining mock data It’s hard and unrealistic to maintain large mock data. It’s is even harder when database structure undergoes changes. Testing GUI Even with MVVM and ability to test GUI, it takes a lot of code to reproduce the GUI scenario. Testing the business I have experience that TDD works well if you limit it to simple business logic. However complex business logic is hard to test since the number of combinations of tests (test space) is very large. Contradiction in requirements In reality it’s hard to capture all requirements under analysis and design. Many times one note requirements lead to contradiction because the project is complex. The contradiction is found late under implementation phase. TDD requires that requirements are 100% correct. In such cases one could expect that conflicting requirements would be captured during creating of tests. But the problem is that this isn’t the case in complex scenarios. I have read this question: Why does TDD work? Does TDD really work for complex enterprise projects, or is it practically limit to project type?

    Read the article

  • Are Intel compilers really better than Microsoft ones?

    - by Rocket Surgeon
    Years ago I was surprised when discovered that Intel sells Studio compatible compilers. I tried it in particular for C/C++ as well as fantastic diagnostic tools. But the code was simply not that computationally intensive to notice the difference. The only impression was: did Intel really did it for me just now, Wow, amazing tools with nanoseconds resolution, unbeleivable. But the trial ended and team never seriously considered a purchase. From your experience, if license cost does not matter, which vendor is a winner ? It is not broad or vague question or attemt to spark a holy war. This sort of question about 2 very visible tools. Nobody likes when tools have any mysteries or surprises. And choices between best and best are always the pain. I also understand the "grass greener" argument. I want to hear all "what ifs" stories. What if Intel just locally optimizes it for the chip stepping of the month, and not every hardware target will actually work as well as Microsoft compiled ? What if AMD hardware is the target and everything will slow down for no reason ? Or on other hand, what if Intel's hardware has so many unnoticable opportunities, that Microsoft compiler writers are too slow to adopt and never implement in the compiler ? What if both are the same exactly, actually a single codebase just wrapped into 2 different boxes and licensed to both vendors by some 3rd party shop? And so on. But someone knows some answers.

    Read the article

  • Who is a CMS really for?

    - by Eirc man
    I have started lately discovering Content Management Systems, and I was wondering, who is really CMS for? What I mean by that: is it only for companies, small businesses or individuals, that pays a contractor to make a website that it's users can just upload content through a easy interface. Or is it used also by programmers, to build their own websites, projects? Would a Facebook, Tweeter, StackExhange ever started by using a CMS, a very powerful one for example. Would you as a programmer build your own "fancy" website on top of a CMS, for example like Typo3, or you would build it from scratch? P.S To be more clear is a summary: What I mean to begin with is, would I as a developer choose a CMS to develop a website that can be scaled with a big base of users, be stuck if I choose to start with a CMS system. What if I build a website using CMS, and the website explodes in popularity, and then I wanted to add much more functionality that I have planed, is it possible that the CMS will limit the growth, because it might have not been build for that kind of scale?

    Read the article

  • Is SugarCRM really adequate for custom development?

    - by dukeofgaming
    I used SugarCRM for a project about two years ago, I ran into errors from the very installation, having to hack the actual installation file to deploy the software in the server... and other erros that I can't recall now. Two years after, I'm picking it up for a project once again... oh dear, I'm feeling like I should have developed the whole thing from scratch myself. Some examples: I couldn't install it in the server (again)... I had to install it locally, then copy the files and database over to the server and manually edit the config file. Constantly getting deployment errors from the module builder... one reason is SugarCRM keeps creating a record in the upgrade_history table for a file that does not exist, I keep deleting such record and it keeps coming back corrupt. I get other deployment errors, but have not figured them out... then I have to rollback all files and database to try again. I deleted a custom module with relationships, the relationships stayed in the other modules and cannot be deleted anymore, PHP warnings all over the place. Quick create for custom modules does not appear, hack needed. Its whole cache directory is a joke, permanent data/files are stored there. The module builder interface disappears required fields. Edit the wrong thing, module builder won't deploy again... then pray Quick Repair and/or Rebuild Relationships do the trick. My impression of SugarCRM now is that, regardless of its pretty exterior and apparent functionality, it is a very low quality piece of software. This even scared me more: http://amplicate.com/hate/sugarcrm; a quote: I wis this info had been available when I tried to implement it 2 years ago... I searched high and low and the only info I found was positive. Yes, it's a piece of crap. The community edition was full of bugs... nothing worked. Essentially I got fired for implementing it. I'm glad though, because now I work for myself, am much happier and make more money... so, I should really thank SugarCRM for sucking so much I guess! I figured that perhaps some of you have had similar experiences, and have either sticked with SugarCRM or moved on to another solution. I'm very interested in knowing what your resolutions were -or your current situations are- to make up my own mind, since the project I'm working on is long term and I'm feeling SugarCRM will be more an obstacle than an aid. Thanks in advance.

    Read the article

  • My new laptop - with a really nice battery option

    - by Rob Farley
    It was about time I got a new laptop, and so I made a phone-call to Dell to discuss my options. I decided not to get an SSD from them, because I’d rather choose one myself – the sales guy tells me that changing the HD doesn’t void my warranty, so that’s good (incidentally, I’d love to hear people’s recommendations for which SSD to get for my laptop). Unfortunately this machine only has one HD slot, but I figure that I’ll put lots of stuff onto external disks anyway. The machine I got was a Dell Studio XPS 16. It’s red (which suits my company), but also has the Intel® Core™ i7-820QM Processor, which is 4 Cores/8 Threads. Makes for a pretty Task Manager, but nothing like the one I saw at SQLBits last year (at 96 cores), or the one that my good friend James Rowland-Jones writes about here. But the reason for this post is actually something in the software that comes with the machine – you know, the stuff that most people uninstall at the earliest opportunity. I had just reinstalled the operating system, and was going through the utilities to get the drivers up-to-date, when I noticed that one of Dell applications included an option to disable battery charging. So I installed it. And sure enough, I can tell the battery not to charge now. Clearly Dell see it as a temporary option, and one that’s designed for when you’re on a plane. But for me, I most often use my laptop with the power plugged in, which means I don’t need to have my battery continually topping itself up. So I really love this option, but I feel like it could go a little further. I’d like “Not Charging” to be the default option, and let me set it when I want to charge it (which should theoretically make my battery last longer). I also intend to work out how this option works, so that I can script it and put it into my StartUp options (so it can be the Default setting). Actually – if someone has already worked this out and can tell me what it does, then please feel free to let me know. Even better would be an external switch. I had a switch on my old laptop (a Dell Latitude) for WiFi, so that I could turn that off before I turned on the computer (this laptop doesn’t give me that option – no physical switch for flight mode). I guess it just means I’ll get used to leaving the WiFi off by default, and turning it on when I want it – might save myself some battery power that way too. Soon I’ll need to take the plunge and sync my iPhone with the new laptop. I’m a little worried that I might lose something – Apple’s messages about how my stuff will be wiped and replaced with what’s on the PC doesn’t fill me with confidence, as it’s a new PC that doesn’t have stuff on it. But having a new machine is definitely a nice experience, and one that I can recommend. I’m sure when I get around to buying an SSD I’ll feel like it’s shiny and new all over again! Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Building KPIs to monitor your business Its not really about the Technology

    When I have discussions with people about Business Intelligence, one of the questions the inevitably come up is about building KPIs and how to accomplish that. From a technical level the concept of a KPI is very simple, almost too simple in that it is like the tip of an iceberg floating above the water. The key to that iceberg is not really the tip, but the mass of the iceberg that is hidden beneath the surface upon which the tip sits. The analogy of the iceberg is not meant to indicate that the foundation of the KPI is overly difficult or complex. The disparity in size in meant to indicate that the larger thing that needs to be defined is not the technical tip, but the underlying business definition of what the KPI means. From a technical perspective the KPI consists of primarily the following items: Actual Value This is the actual value data point that is being measured. An example would be something like the amount of sales. Target Value This is the target goal for the KPI. This is a number that can be measured against Actual Value. An example would be $10,000 in monthly sales. Target Indicator Range This is the definition of ranges that define what type of indicator the user will see comparing the Actual Value to the Target Value. Most often this is defined by stoplight, but can be any indicator that is going to show a status in a quick fashion to the user. Typically this would be something like: Red Light = Actual Value more than 5% below target; Yellow Light = Within 5% of target either direction; Green Light = More than 5% higher than Target Value Status\Trend Indicator This is an optional attribute of a KPI that is typically used to show some kind of trend. The vast majority of these indicators are used to show some type of progress against a previous period. As an example, the status indicator might be used to show how the monthly sales compare to last month. With this type of indicator there needs to be not only a definition of what the ranges are for your status indictor, but then also what value the number needs to be compared against. So now we have an idea of what data points a KPI consists of from a technical perspective lets talk a bit about tools. As you can see technically there is not a whole lot to them and the choice of technology is not as important as the definition of the KPIs, which we will get to in a minute. There are many different types of tools in the Microsoft BI stack that you can use to expose your KPI to the business. These include Performance Point, SharePoint, Excel, and SQL Reporting Services. There are pluses and minuses to each technology and the right technology is based a lot on your goals and how you want to deliver the information to the users. Additionally, there are other non-Microsoft tools that can be used to expose KPI indicators to your business users. Regardless of the technology used as your front end, the heavy lifting of KPI is in the business definition of the values and benchmarks for that KPI. The discussion about KPIs is very dependent on the history of an organization and how much they are exposed to the attributes of a KPI. Often times when discussing KPIs with a business contact who has not been exposed to KPIs the discussion tends to also be a session educating the business user about what a KPI is and what goes into the definition of a KPI. The majority of times the business user has an idea of what their actual values are and they have been tracking those numbers for some time, generally in Excel and all manually. So they will know the amount of sales last month along with sales two years ago in the same month. Where the conversation tends to get stuck is when you start discussing what the target value should be. The actual value is answering the What and How much questions. When you are talking about the Target values you are asking the question Is this number good or bad. Typically, the user will know whether or not the value is good or bad, but most of the time they are not able to quantify what is good or bad. Their response is usually something like I just know. Because they have been watching the sales quantity for years now, they can tell you that a 5% decrease in sales this month might actually be a good thing, maybe because the salespeople are all waiting until next month when the new versions come out. It can sometimes be very hard to break the business people of this habit. One of the fears generally is that the status indicator is not subjective. Thus, in the scenario above, the business user is going to be fearful that their boss, just looking at a negative red indicator, is going to haul them out to the woodshed for a bad month. But, on the flip side, if all you are displaying is the amount of sales, only a person with knowledge of last month sales and the target amount for this month would have any idea if $10,000 in sales is good or not. Here is where a key point about KPIs needs to be communicated to both the business user and any user who might be viewing the results of that KPI. The KPI is just one tool that is used to report on business performance. The KPI is meant as a quick indicator of one business statistic. It is not meant to tell the entire story. It does not answer the question Why. Its primary purpose is to objectively and quickly expose an area of the business that might warrant more review. There is always going to be the need to do further analysis on any potential negative or neutral KPI. So, hopefully, once you have convinced your business user to come up with some target numbers and ranges for status indicators, you then need to take the next step and help them answer the Why question. The main question here to ask is, Okay, you see the indicator and you need to discover why the number is what is, where do you go?. The answer is usually a combination of sources. A sales manager might have some of the following items at their disposal (Marketing report showing a decrease in the promotional discounts for the month, Pricing Report showing the reduction of prices of older models, an Inventory Report showing the discontinuation of a particular product line, or a memo showing the ending of a large affiliate partnership. The answers to the question Why are never as simple as a single indicator value. Bring able to quickly get to this information is all about designing how a user accesses the KPIs and then also how easily they can get to the additional information they need. This is where a Dashboard mentality can come in handy. For example, the business user can have a dashboard that shows their KPIs, but also has links to some of the common reports that they run regarding Sales Data. The users boss may have the same KPIs on their dashboard, but instead of links to individual reports they are going to have a link to a status report that was created by the user that pulls together all the data about the KPI in a summary format the users boss can review. So some of the key things to think about when building or evaluating KPIs for your organization: Technology should not be the driving factor KPIs are of little value without some indicator for whether a value is good, bad or neutral. KPIs only give an answer to the Is this number good\bad? question Make sure the ability to drill into the Why of a KPI is close at hand and relevant to the user who is viewing the KPI. The KPI is a key business tool when defined properly to help monitor business performance across the enterprise in an objective and consistent manner. At times it might feel like the process of defining the business aspects of a KPI can sometimes be arduous, the payoff in the end can far outweigh the costs. Some of the benefits of going through this process are a better understanding of the key metrics for an organization and the measure of those metrics and a consistent snapshot of business performance that can be utilized across the organization. And I think that these are benefits to any organization regardless of the technology or the implementation.Did you know that DotNetSlackers also publishes .net articles written by top known .net Authors? We already have over 80 articles in several categories including Silverlight. Take a look: here.

    Read the article

  • Why do we (really) program to interfaces?

    - by Kyle Burns
    One of the earliest lessons I was taught in Enterprise development was "always program against an interface".  This was back in the VB6 days and I quickly learned that no code would be allowed to move to the QA server unless my business objects and data access objects each are defined as an interface and have a matching implementation class.  Why?  "It's more reusable" was one answer.  "It doesn't tie you to a specific implementation" a slightly more knowing answer.  And let's not forget the discussion ending "it's a standard".  The problem with these responses was that senior people didn't really understand the reason we were doing the things we were doing and because of that, we were entirely unable to realize the intent behind the practice - we simply used interfaces and had a bunch of extra code to maintain to show for it. It wasn't until a few years later that I finally heard the term "Inversion of Control".  Simply put, "Inversion of Control" takes the creation of objects that used to be within the control (and therefore a responsibility of) of your component and moves it to some outside force.  For example, consider the following code which follows the old "always program against an interface" rule in the manner of many corporate development shops: 1: ICatalog catalog = new Catalog(); 2: Category[] categories = catalog.GetCategories(); In this example, I met the requirement of the rule by declaring the variable as ICatalog, but I didn't hit "it doesn't tie you to a specific implementation" because I explicitly created an instance of the concrete Catalog object.  If I want to test the functionality of the code I just wrote I have to have an environment in which Catalog can be created along with any of the resources upon which it depends (e.g. configuration files, database connections, etc) in order to test my functionality.  That's a lot of setup work and one of the things that I think ultimately discourages real buy-in of unit testing in many development shops. So how do I test my code without needing Catalog to work?  A very primitive approach I've seen is to change the line the instantiates catalog to read: 1: ICatalog catalog = new FakeCatalog();   once the test is run and passes, the code is switched back to the real thing.  This obviously poses a huge risk for introducing test code into production and in my opinion is worse than just keeping the dependency and its associated setup work.  Another popular approach is to make use of Factory methods which use an object whose "job" is to know how to obtain a valid instance of the object.  Using this approach, the code may look something like this: 1: ICatalog catalog = CatalogFactory.GetCatalog();   The code inside the factory is responsible for deciding "what kind" of catalog is needed.  This is a far better approach than the previous one, but it does make projects grow considerably because now in addition to the interface, the real implementation, and the fake implementation(s) for testing you have added a minimum of one factory (or at least a factory method) for each of your interfaces.  Once again, developers say "that's too complicated and has me writing a bunch of useless code" and quietly slip back into just creating a new Catalog and chalking any test failures up to "it will probably work on the server". This is where software intended specifically to facilitate Inversion of Control comes into play.  There are many libraries that take on the Inversion of Control responsibilities in .Net and most of them have many pros and cons.  From this point forward I'll discuss concepts from the standpoint of the Unity framework produced by Microsoft's Patterns and Practices team.  I'm primarily focusing on this library because it questions about it inspired this posting. At Unity's core and that of most any IoC framework is a catalog or registry of components.  This registry can be configured either through code or using the application's configuration file and in the most simple terms says "interface X maps to concrete implementation Y".  It can get much more complicated, but I want to keep things at the "what does it do" level instead of "how does it do it".  The object that exposes most of the Unity functionality is the UnityContainer.  This object exposes methods to configure the catalog as well as the Resolve<T> method which is used to obtain an instance of the type represented by T.  When using the Resolve<T> method, Unity does not necessarily have to just "new up" the requested object, but also can track dependencies of that object and ensure that the entire dependency chain is satisfied. There are three basic ways that I have seen Unity used within projects.  Those are through classes directly using the Unity container, classes requiring injection of dependencies, and classes making use of the Service Locator pattern. The first usage of Unity is when classes are aware of the Unity container and directly call its Resolve method whenever they need the services advertised by an interface.  The up side of this approach is that IoC is utilized, but the down side is that every class has to be aware that Unity is being used and tied directly to that implementation. Many developers don't like the idea of as close a tie to specific IoC implementation as is represented by using Unity within all of your classes and for the most part I agree that this isn't a good idea.  As an alternative, classes can be designed for Dependency Injection.  Dependency Injection is where a force outside the class itself manipulates the object to provide implementations of the interfaces that the class needs to interact with the outside world.  This is typically done either through constructor injection where the object has a constructor that accepts an instance of each interface it requires or through property setters accepting the service providers.  When using dependency, I lean toward the use of constructor injection because I view the constructor as being a much better way to "discover" what is required for the instance to be ready for use.  During resolution, Unity looks for an injection constructor and will attempt to resolve instances of each interface required by the constructor, throwing an exception of unable to meet the advertised needs of the class.  The up side of this approach is that the needs of the class are very clearly advertised and the class is unaware of which IoC container (if any) is being used.  The down side of this approach is that you're required to maintain the objects passed to the constructor as instance variables throughout the life of your object and that objects which coordinate with many external services require a lot of additional constructor arguments (this gets ugly and may indicate a need for refactoring). The final way that I've seen and used Unity is to make use of the ServiceLocator pattern, of which the Patterns and Practices team has also provided a Unity-compatible implementation.  When using the ServiceLocator, your class calls ServiceLocator.Retrieve in places where it would have called Resolve on the Unity container.  Like using Unity directly, it does tie you directly to the ServiceLocator implementation and makes your code aware that dependency injection is taking place, but it does have the up side of giving you the freedom to swap out the underlying IoC container if necessary.  I'm not hugely concerned with hiding IoC entirely from the class (I view this as a "nice to have"), so the single biggest problem that I see with the ServiceLocator approach is that it provides no way to proactively advertise needs in the way that constructor injection does, allowing more opportunity for difficult to track runtime errors. This blog entry has not been intended in any way to be a definitive work on IoC, but rather as something to spur thought about why we program to interfaces and some ways to reach the intended value of the practice instead of having it just complicate your code.  I hope that it helps somebody begin or continue a journey away from being a "Cargo Cult Programmer".

    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

  • What is JavaScript, really?

    - by Lord Loh.
    All this started when I was looking for a way to test my webpage for JavaScript conformance like the W3C HTML Validator. I have not found one yet. So let me know if you know of any... I looked for the official JavaScript page and find ECMA Script. These people have standardized a scripting language (I do not feel like calling it JavaScript anymore!) and called it ECMA-262 (Wikipedia). Their latest work is Edition 5.1 JavaScript was developed my Mozilla Corporation and their last stable version is 1.8.5 (see this) which is based on the ECMA's edition 5.1 The Wikipedia page linked mentions dialects. Mozilla's JavaScript 1.8.5 is listed as a dialect along with JScript 9 (IE) and JavaScript (Chrome's V8[Wiki]) and a lot others. Am I to understand that JavaScript 1.8.5 is a derivative of the ECMA-262 and SpiderMonkey[Wiki] is an engine that runs it? And Chrome has its own dialect and V8 engine is the program that runs it? With all these dialects based off ECMA-262, what I can no longer understand is "What is JavaScript"? Are there any truly cross browser scripting languages? Do the various implementers come together to agree on the dialect cross compatibility? Is this effort ECMA?

    Read the article

  • JD Edwards in the Cloud…Really Already!

    - by user709270
    Yes, there is a lot of conversation about Oracle and the cloud.  Many of you may assume that Oracle applications in the cloud  only apply to Oracle Fusion Applications.  And JD Edwards customers are curious about if, when and how JD Edwards might be offered to them as a subscription offering.  The truth of the matter is that Oracle partners today are providing a JD Edwards subscription offering.  In order to help you understand what’s available, please read on for the reader’s digest version! Let’s start with a definition.  JD Edwards EnterpriseOne is available as an Accelerate subscription.  Oracle “Accelerate” subscription is Oracle's approach for providing simple to deploy, packaged, enterprise-class software solutions to growing midsize organizations through its network of expert partners. The partners that offer Oracle  JD Edwards Accelerate Subscriptions do so via their Partner Private Clouds (PPC).  The Oracle JD Edwards cloud solutions are offered only by qualified Oracle JD Edwards partners and they provide customers a complete Oracle solution that includes license software, maintenance, hosting and other services on a monthly subscription basis.  Qualified partners must be members of Oracle PartnerNetwork, be an Oracle Accelerate solutions provider and be enabled to deliver JD Edwards applications via Oracle Business Accelerator rapid implementation technology.  Currently we have many JD Edwards partners around the globe that offer the JD Edwards Accelerate Subscription model.  To access a list of Oracle JD Edwards partners currently in this program click here.  To learn more about Oracle JD Edwards Cloud Computing read this recently published white paper:   Oracle JD Edwards Cloud Computing. Choosing a deployment strategy that fits

    Read the article

  • When a restore isn’t really complete

    - by John Paul Cook
    This week I discovered that restoring from a full backup doesn’t always restore SQL Server to the same state it was in when the backup was made. There are three settings that, if enabled, are not restored after a database restore. Thanks to Greg Low for pointing out that the list of affected settings is found in the SQL Server 2008 Upgrade Technical Reference Guide from which I quote: · is_broker_enabled · is_honor_broker_priority_on · is_trustworthy_on Detaching and attaching a database will also...(read more)

    Read the article

  • Is Learning C++ Through The Qt Framework Really Learning C++

    - by user866190
    The problem I have, is that most of the C++ books I read spend almost forever on syntax and the basics of the language, e.g. for and loops while, arrays, lists, pointers, etc. But they never seem to build anything that is simple enough to use for learning, yet practical enough to get you to understand the philosophy and power of the language. Then I stumbled upon QT which is an amazing library! But working through the demos they have, it seems like I am now in the reverse dilemma. I feel like the rich man's son driving round in a sports car subsidized by the father. Like I could build fantastic software, but have no clue what's going on under the hood. As an example of my dilemma take the task of building a simple web browser. In pure C++, I wouldn't even know where to start, yet with the Qt library it can be done within a few lines on code. I am not complaining about this. I am just wondering how to fill the knowledge void between the basic structure of the language and the high level interface that the Qt framework provides?

    Read the article

  • Is it really free to deploy SharpDevelop apps?

    - by Gabriel
    I'm trying SharpDevelop to develop c# applications. Regardless of the language and the IDE, is it free to deploy applications that use WinForms? I've been developing with MonoDevelop just because it has a designer for Gtk# (client doesn't want to pay more for licences as to use VS...), but it's too buggy and it's making us lose lots of time. SharpDevelop looks great at first sight, but I wouldn't like the user or us to have problems with legal software. Thank you for the data! Warm regards.

    Read the article

  • Is recursion really bad?

    - by dotneteer
    After my previous post about the stack space, it appears that there is perception from the feedback that recursion is bad and we should avoid deep recursion. After writing a compiler, I know that the modern computer and compiler are complex enough and one cannot automatically assume that a hand crafted code would out-perform the compiler optimization. The only way is to do some prototype to find out. So why recursive code may not perform as well? Compilers place frames on a stack. In additional to arguments and local variables, compiles also need to place frame and program pointers on the frame, resulting in overheads. So why hand-crafted code may not performance as well? The stack used by a compiler is a simpler data structure and can grow and shrink cleanly. To replace recursion with out own stack, our stack is allocated in the heap that is far more complicated to manage. There could be overhead as well if the compiler needs to mark objects for garbage collection. Compiler also needs to worry about the memory fragmentation. Then there is additional complexity: CPUs have registers and multiple levels of cache. Register access is a few times faster than in-CPU cache access and is a few 10s times than on-board memory access. So it is up to the OS and compiler to maximize the use of register and in-CPU cache. For my particular problem, I did an experiment to rewrite my c# version of recursive code with a loop and stack approach. So here are the outcomes of the two approaches:   Recursive call Loop and Stack Lines of code for the algorithm 17 46 Speed Baseline 3% faster Readability Clean Far more complex So at the end, I was able to achieve 3% better performance with other drawbacks. My message is never assuming your sophisticated approach would automatically work out better than a simpler approach with a modern computer and compiler. Gage carefully before committing to a more complex approach.

    Read the article

  • PonyEdit: It’s really fast

    - by Gary Pendergast
    Over the past few months, a friend and I have been hard at work on a new breed of text editor that we call PonyEdit. If you’ve ever found yourself cursing over the lag of working on remote cloud servers, this is the editor for you.It’s not just another SFTP editor…Reading and writing files over SFTP is nothing new; dozens of text editors can do it. But it’s always slow, clunky and feels like the feature was bolted on as an afterthought. You’ll find yourself using separate shortcuts to open files locally vs remotely, and dealing with sometimes painful save times with every edit, no matter how minor.PonyEdit gets rid of this terribly slow method of working by connecting over SSH, and using edit streaming to push changes to the server in the background as-you-type.Head on over to PonyEdit.com to download a free trial, and let me know what you think! Oh, and…Stand by to have your mind blown.

    Read the article

  • Have you really fixed that problem?

    - by DavidWimbush
    The day before yesterday I saw our main live server's CPU go up to constantly 100% with just the occasional short drop to a lower level. The exact opposite of what you'd want to see. We're log shipping every 15 minutes and part of that involves calling WinRAR to compress the log backups before copying them over. (We're on SQL2005 so there's no native compression and we have bandwidth issues with the connection to our remote site.) I realised the log shipping jobs were taking about 10 minutes and that most of that was spent shipping a 'live' reporting database that is completely rebuilt every 20 minutes. (I'm just trying to keep this stuff alive until I can improve it.) We can rebuild this database in minutes if we have to fail over so I disabled log shipping of that database. The log shipping went down to less than 2 minutes and I went off to the SQL Social evening in London feeling quite pleased with myself. It was a great evening - fun, educational and thought-provoking. Thanks to Simon Sabin & co for laying that on, and thanks too to the guests for making the effort when they must have been pretty worn out after doing DevWeek all day first. The next morning I came down to earth with a bump: CPU still at 100%. WTF? I looked in the activity monitor but it was confusing because some sessions have been running for a long time so it's not a good guide what's using the CPU now. I tried the standard reports showing queries by CPU (average and total) but they only show the top 10 so they just show my big overnight archiving and data cleaning stuff. But the Profiler showed it was four queries used by our new website usage tracking system. Four simple indexes later the CPU was back where it should be: about 20% with occasional short spikes. So the moral is: even when you're convinced you've found the cause and fixed the problem, you HAVE to go back and confirm that the problem has gone. And, yes, I have checked the CPU again today and it's still looking sweet.

    Read the article

< Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >