Search Results

Search found 18191 results on 728 pages for 'single board'.

Page 113/728 | < Previous Page | 109 110 111 112 113 114 115 116 117 118 119 120  | Next Page >

  • Loading Dimension Tables - Methodologies

    - by Nev_Rahd
    Hello, Recently I been working on project, where need to populated Dim Tables from EDW Tables. EDW Tables are of type II which does maintain historical data. When comes to load Dim Table, for which source may be multiple EDW Tables or would be single table with multi level pivoting (on attributes). Mean: There would be 10 records - one for each attribute which need to be pivoted on domain_code to make a single row in Dim. Out of these 10 records there would be some attributes with same domain_code but with different sub_domain_code, which needs further pivoting on subdomain code. Ex: if i got domain code: 01,02, 03 = which are straight pivot on domain code I would also have domain code: 10 with subdomain code / version as 2006,2007,2008,2009 That means I need to split my source table with above attributes into two = one for domain code and other for domain_code + version. so far so good. When it comes to load Dim Table: As per design specs for Dimensions (originally written by third party), what they want is: for every single change in EDW (attribute), it should assemble all the related records (for that NK) mean new one with other attribute values which are current = process them to create a new dim record and insert it. That mean if a single extract contains 100 records updated (one for each NK), it should assemble 100 + (100*9) records to insert / update dim table. How good is this approach. Other way I tried to do is just do a lookup into dim table for that NK get the value's of recent records (attributes which not changed) and insert it and update the current one. What would be the better approach assembling records at source side for one attribute change or looking into dim table's recent record and process it. If this doesn't make sense, would like to elaborate it further. Thanks

    Read the article

  • Can someone confirm how Microsoft Excel 2007 internally represents numbers?

    - by Jon
    I know the IEEE 754 floating point standard by heart as I had to learn it for an exam. I know exactly how floating point numbers are used and the problems that they can have. I can manually do any operation on the binary representation of floating point numbers. However, I have not found a single source which unambiguously states that excel uses 64 bit floating point numbers to internally represent every single cell "type" in excel except for text. I have no idea whether some of the types use signed or unsigned integers and some use 64 bit floating point. I have found literally trillions of articles which 1) describe floating point numbers and then 2) talk about being careful with excel because of floating point numbers. I have not found a single statement saying "all types are 64 bit floating point numbers except text". I have not found a single statement which says "changing the type of a cell only changes its visual representation and not its internal representation, unless you change the type from text to some other type which is not text or you change some other type which is not text to text". This is literally all I want to know, and it's so simple and axiomatic that I am amazed that I can find trillions of articles and pages which talk around these statements but do not state them directly.

    Read the article

  • ASP.NET Web Application: use 1 or multiple virtual directories

    - by tster
    I am working on a (largish) internal web application which has multiple modules (security, execution, features, reports, etc.). All the pages in the app share navigation, CSS, JS, controls, etc. I want to make a single "Web Application" project, which includes all the pages for the app, then references various projects which will have the database and business logic in them. However, some of the people on the project want to have separate projects for the pages of each module. To make this more clear, this is what I'm advocating to be the projects. /WebInterface* /SecurityLib /ExecutionLib etc... And here is what they are advocating: /SecurityInterface* /SecutiryLib /ExecutionInterface* /ExecutionLib etc... *project will be published to a virtual directory of IIS Basically What I'm looking for is the advantages of both approaches. Here is what I can think of so far: Single Virtual Directory Pros Modules can share a single MasterPage Modules can share UserControls (this will be common) Links to other modules are within the same Virtual directory, and thus don't need to be fully qualified. Less chance of having incompatible module versions together. Multiple Virtual Directories Pros Can publish a new version of a single module without disrupting other modules Module is more compartmentalized. Less likely that changes will break other modules. I don't buy those arguments though. First, using load balanced servers (which we will have) we should be able to publish new versions of the project with zero downtime assuming there are no breaking database changes. Second, If something "breaks" another module, then there is either an improper dependency or the break will show up eventually in the other module, when the developers copy over the latest version of the UserControl, MasterPage or dll. As a point of reference, there are about 10 developers on the project for about 50% of their time. The initial development will be about 9 months.

    Read the article

  • Blank Swf after printing from another app.

    - by Carlos Barbosa
    Ok, so i am having a problem, where i developed an app that uses the webcam, adds an image and then takes a picture, well, i implemented textbook style the printjob, and well everytime after printing my swf goes blank. http://stackoverflow.com/questions/3021557/as3-printing-problem-blanks-swf-after-print-or-cancel But now i have been continuing testing, and i found out, that if i print from another application and then comeback, my video stream from the webcam is blank, this has no sense at all, since i am printing from another app. Please i need help with this, i have found one additional case: http://board.flashkit.com/board/showthread.php?t=781089 What is going on here i mean, this must be a flash player issue right.

    Read the article

  • Should frontend and backend be handled by different controllers?

    - by DR
    In my previous learning projects I always used a single controller, but now I wonder if that is good practice or even always possible. In all RESTful Rails tutorials the controllers have a show, an edit and an index view. If an authorized user is logged on, the edit view becomes available and the index view shows additional data manipulation controls, like a delete button or a link to the edit view. Now I have a Rails application which falls exactly into this pattern, but the index view is not reusable: The normal user sees a flashy index page with lots of pictures, complex layout, no Javascript requirement, ... The Admin user index has a completly different minimalistic design, jQuery table and lots of additional data, ... Now I'm not sure how to handle this case. I can think of the following: Single controller, single view: The view is split into two large blocks/partials using an if statement. Single controller, two views: index and index_admin. Two different controllers: BookController and BookAdminController None of these solutions seems perfect, but for now I'm inclined to use the 3rd option. What's the preferred way to do this?

    Read the article

  • Out-of-memory algorithms for addressing large arrays

    - by reve_etrange
    I am trying to deal with a very large dataset. I have k = ~4200 matrices (varying sizes) which must be compared combinatorially, skipping non-unique and self comparisons. Each of k(k-1)/2 comparisons produces a matrix, which must be indexed against its parents (i.e. can find out where it came from). The convenient way to do this is to (triangularly) fill a k-by-k cell array with the result of each comparison. These are ~100 X ~100 matrices, on average. Using single precision floats, it works out to 400 GB overall. I need to 1) generate the cell array or pieces of it without trying to place the whole thing in memory and 2) access its elements (and their elements) in like fashion. My attempts have been inefficient due to reliance on MATLAB's eval() as well as save and clear occurring in loops. for i=1:k [~,m] = size(data{i}); cur_var = ['H' int2str(i)]; %# if i == 1; save('FileName'); end; %# If using a single MAT file and need to create it. eval([cur_var ' = cell(1,k-i);']); for j=i+1:k [~,n] = size(data{j}); eval([cur_var '{i,j} = zeros(m,n,''single'');']); eval([cur_var '{i,j} = compare(data{i},data{j});']); end save(cur_var,cur_var); %# Add '-append' when using a single MAT file. clear(cur_var); end The other thing I have done is to perform the split when mod((i+j-1)/2,max(factor(k(k-1)/2))) == 0. This divides the result into the largest number of same-size pieces, which seems logical. The indexing is a little more complicated, but not too bad because a linear index could be used. Does anyone know/see a better way?

    Read the article

  • How to change the image by click keyboard arrows?

    - by Nitz
    Hey Guys, I have made one project in java. Now in this particular module, i am showing the user , so i am showing them a stock inventory. Now for doing this. As you will see in the screen shot [ i have put screenshot so you can understand well ]. To show the images , i have made one table and i have put labels in that and then i am setting the icon at the label. Now as soon as user click on the any image, then that image i put on big label. But this works perfect on mouse click but client want that , if user navigate the by keyboard arrow keys then in same should be happen. Means : User navigate by key board then it should reflect on that table [user should know that which is currently seleted ] and then as soon as the user navigate by key board arrow events and then click enter then that image should reflect on big label. Here is the Link for that screenshot.

    Read the article

  • Java - understanding servlets

    - by Trup
    I am working on a homework project that should implement a board game between 2 clients over an HttpServlet. I have couple of questions: 1) I read that HttpServlets must be stateless, however, for the sake of the game, I have to keep a lot of state(whose turn it is, the state of the board, etc). Do I have to keep this in the clients? Does the HttpServlet indeed have to be stateless, i.e. have no fields that track state? 2) I know that the clients will talk to the servlet via the doGet/doPost methods, but how can the servlet talk to the clients(for example, if player 1 just made a move and sent it to the servlet, the servlet has to tell client 2 what the move was). Thank you Also, if you can point me to a useful, simple example of a similar code online, I would be very grateful

    Read the article

  • C++ USB Programming

    - by JB_SO
    Hi, I am new to hardware programming(especially USB) so please bear with me and my questions. I am using C++ and I need to send/receive some data (a byte array) to/from a USB port on a microprocessor board. Now, I have done some serial port programming before and I know that for a serial port you have to open a port, setup, perform i/o and finally close the port. I am guessing to use a USB port, it is not as simple as what I mentioned above. I do know that I want to use Microsoft standard drivers and implement standard Windows IO commands to accomplish this, since I believe there are no drivers for the microprocessor board for me to interact with. If somebody can point me in the right direction as to the steps needed to "talk" to a USB port (open, setup, i/o) via standard Windows IO commands, I would truly and greatly appreciate it. Thanks you so much!!

    Read the article

  • c++ recursion with arrays

    - by Sam
    I have this project that I'm working on for a class, and I'm not looking for the answer, but maybe just a few tips since I don't feel like I'm using true recursion. The problem involves solving the game of Hi Q or "peg solitaire" where you want the end result to be one peg remaining (it's played with a triangular board and one space is open at the start.) I've represented the board with a simple array, each index being a spot and having a value of 1 if it has a peg, and 0 if it doesn't and also the set of valid moves with a 2 dimensional array that is 36, 3 (each move set contains 3 numbers; the peg you're moving, the peg it hops over, and the destination index.) So my problem is that in my recursive function, I'm using a lot of iteration to determine things like which space is open (or has a value of 0) and which move to use based on which space is open by looping through the arrays. Secondly, I don't understand how you would then backtrack with recursion, in the event that an incorrect move was made and you wanted to take that move back in order to choose a different one. This is what I have so far; the "a" array represents the board, the "b" array represents the moves, and the "c" array was my idea of a reminder as to which moves I used already. The functions used are helper functions that basically just loop through the arrays to find an empty space and corresponding move. : void solveGame(int a[15], int b[36][3], int c[15][3]){ int empSpace; int moveIndex; int count = 0; if(pegCount(a) < 2){ return; } else{ empSpace = findEmpty(a); moveIndex = chooseMove( a, b, empSpace); a[b[moveIndex][0]] = 0; a[b[moveIndex][1]] = 0; a[b[moveIndex][2]] = 1; c[count][0] = b[moveIndex][0]; c[count][1] = b[moveIndex][1]; c[count][2] = b[moveIndex][2]; solveGame(a,b,c); } }

    Read the article

  • "Othello" game needs some clarification

    - by pappu
    I am trying to see if my understanding of "othello" fame is correct or not. According to the rules, we flip the dark/light sides if we get some sequence like X000X = XXXXX. The question I have is if in the process of flipping 0-X or X- 0, do we also need to consider the rows/columns/diagonals of newly flipped elements? e.g. consider board state as shown in above image(New element X is placed @ 2,3) When we update board, we mark elements from 2,3 to 6,3 as Xs but in this process elements like horizontal 4,3 to 4,5 and diagonal 2,3 to 4,5 are also eligible for update? so do we update those elements as well? or just the elements which have starting as 2,3 (i.e update rows/column/diagonal whose starting point is the element we are dealing with, in our case 2,3?) Please help me understand it

    Read the article

  • Starting Game dev on iPhone:iPad - learning path ??

    - by Tibi
    Hi there, I'm beginning in iPhone/iPad game dev and I'm searching to set up my learning path. The basic features I would like to learn (after the basic SDK iphone components programming) are : using a board like interface where I can move pawn with my fingers detect where the pawn was moved and triggers events in the game The board will be constitute by 6 tiles that may be organised randomly when starting the game : may i use an sdk component with a delegate and datasource to determine where the pawn was left and on which tile ? need to use dices (which kind of library may I use ?) ... Do you have any idea about where to start ? ;-) Many thanks, Tib.

    Read the article

  • I want to create a simple function to reset all input values without target input. (Javascript, Jque

    - by question_about_the_problem
    Hi, I want to create a simple function to reset all input values without target input. I don't know how i can do it. Thanks. Here is my sample codes: <script src="http://ajax.googleapis.com/ajax/libs/jquery/1.4.2/jquery.min.js"></script> <script> function reset_other_inputs(room) { $("input[name^='check_in_date_']").each(function () { if ( $("input[name!='check_in_date_'+room]") ) { this.val(""); } }); $("input[name^='check_out_date_']").each(function () { if ( $("input[name!='check_out_date_'+room]") ) { this.val(""); } }); } </script> <input type="text" name="check_in_date_single" value="single" onClick="reset_other_inputs('single');"> <input type="text" name="check_out_date_single" value="single" onClick="reset_other_inputs('single');"> <input type="text" name="check_in_date_double" value="double" onClick="reset_other_inputs('double');"> <input type="text" name="check_out_date_double" value="double" onClick="reset_other_inputs('double');"> <input type="text" name="check_in_date_triple" value="triple" onClick="reset_other_inputs('triple');"> <input type="text" name="check_out_date_triple" value="triple" onClick="reset_other_inputs('triple');"> <input type="text" name="check_in_date_suite" value="suite" onClick="reset_other_inputs('suite');"> <input type="text" name="check_out_date_suite" value="suite" onClick="reset_other_inputs('suite');">

    Read the article

  • How to efficiently use LOCK_ESCALATION mssql 2008

    - by Avias
    I'm currently having troubles with frequent deadlocks with a specific user table in MS SQL 2008. Here are some facts about this particular table: Has a large amount of rows (1 to 2 million) All the indexes used on this table only has "use row lock" ticked on its option rows are frequently updated by multiple transactions but are unique (e.g. probably a thousand or more update statements are executed to different unique rows every hour) the table does not use partitions. Upon checking the table on sys.tables, I found that the lock_escalation is set to TABLE I'm very tempted to turn the lock_escalation for this table to DISABLE but I'm not really sure what side effect this would incur. From What I understand, using DISABLE will minimize escalating locks to TABLE level which if combined with the row lock settings of the indexes should theoretically minimize the deadlocks I am encountering.. From what I have read in Determining threshold for lock escalation it seems that locking automatically escalates when a single transaction fetches 5000 rows.. What does a single transaction mean in this sense? A single session/connection getting 5000 rows thru individual update/select statements? Or is it a single sql update/select statement that fetches 5000 or more rows? Any insight is appreciated, btw, n00b DBA here Thanks

    Read the article

  • Need help INSERT record(s) MySQL DB

    - by JM4
    I have an online form which collects member(s) information and stores it into a very long MySQL database. We allow up to 16 members to enroll at a single time and originally structured the DB to allow such. For example: If 1 Member enrolls, his personal information (first name, last name, address, phone, email) are stored on a single row. If 15 Members enroll (all at once), their personal information are stored in the same single row. The row has information housing columns for all 'possible' inputs. I am trying to consolidate this code and having every nth member that enrolls put onto a new record within the database. I have seen sugestions before for inserting multiple records as such: INSERT INTO tablename VALUES (('$f1name', '$f1address', '$f1phone'), ('$f2name', '$f2address', '$f2phone')... The issue with this is two fold: I do not know how many records are being enrolled from person to person so the only way to make the statement above is to use a loop The information collected from the forms is NOT a single array so I can't loop through one array and have it parse out. My information is collected as individual input fields like such: Member1FirstName, Member1LastName, Member1Phone, Member2Firstname, Member2LastName, Member2Phone... and so on Is it possible to store information in separate rows WITHOUT using a loop (and therefore having to go back and completely restructure my form field names and such (which can't happen due to the way the validation rules are built.)

    Read the article

  • How to approach socket programming between C# -> Java (Android)

    - by Alex
    I've recently knocked up a server/client app for Windows & Android that allows one to send a file from Windows to an android phone over a socket connection. It works great for a single file but trying to send multiple files over in a single stream is causing me problems. I've also realised that aside from the binary data, I will need to send messages over the socket to indicate error states and other application messages. I have little experience with network programming and and wondering what is the best way forward. Basically the C# server side of the app just goes into a listening state and uses Socket.SendFile to transmit the file. On Android I use the standard Java Socket.getInputStream() to receive the file. That works great for a single file transfer, but how should I handle multiple files and error/messaging information? Do I need to use a different socket for each file? Should I be using a higher level framework to handle this or can I send everything over the single socket? Any other suggestions for frameworks or learning materials?

    Read the article

  • How do i change everything behind a certain point in a Jagged array?

    - by Jack Null
    Say I have a jagged array, and position 2,3 is taken by int 3. Every other spot is filled with int 0. How would I fill all the positions behind 2,3 with a 4? 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 to this: 4 4 4 4 4 4 4 4 4 4 4 4 4 3 0 0 0 0 0 0 0 Ive tried variations of this: int a = 2; int b = 3; for (int x = 0; x < a; x++) { for (int y = 0; y < board.space[b].Length; y++) { board.space[x][y] = 4; } }

    Read the article

  • Sort List based on dynamically generated numbers in C++

    - by user367322
    I have a list of objects ("Move"'s in this case) that I want to sort based on their calculated evaluation. So, I have the List, and a bunch of numbers that are "associated" with an element in the list. I now want to sort the List elements with the first element having the lowest associated number, and the last having the highest. Once the items are order I can discard the associated number. How do I do this? This is what my code looks like (kind've): list<Move> moves = board.getLegalMoves(board.turn); for(i = moves.begin(); i != moves.end(); ++i) { //... a = max; // <-- number associated with current Move }

    Read the article

  • Is Berkeley DB a NoSQL solution?

    - by Gregory Burd
    Berkeley DB is a library. To use it to store data you must link the library into your application. You can use most programming languages to access the API, the calls across these APIs generally mimic the Berkeley DB C-API which makes perfect sense because Berkeley DB is written in C. The inspiration for Berkeley DB was the DBM library, a part of the earliest versions of UNIX written by AT&T's Ken Thompson in 1979. DBM was a simple key/value hashtable-based storage library. In the early 1990s as BSD UNIX was transitioning from version 4.3 to 4.4 and retrofitting commercial code owned by AT&T with unencumbered code, it was the future founders of Sleepycat Software who wrote libdb (aka Berkeley DB) as the replacement for DBM. The problem it addressed was fast, reliable local key/value storage. At that time databases almost always lived on a single node, even the most sophisticated databases only had simple fail-over two node solutions. If you had a lot of data to store you would choose between the few commercial RDBMS solutions or to write your own custom solution. Berkeley DB took the headache out of the custom approach. These basic market forces inspired other DBM implementations. There was the "New DBM" (ndbm) and the "GNU DBM" (GDBM) and a few others, but the theme was the same. Even today TokyoCabinet calls itself "a modern implementation of DBM" mimicking, and improving on, something first created over thirty years ago. In the mid-1990s, DBM was the name for what you needed if you were looking for fast, reliable local storage. Fast forward to today. What's changed? Systems are connected over fast, very reliable networks. Disks are cheep, fast, and capable of storing huge amounts of data. CPUs continued to follow Moore's Law, processing power that filled a room in 1990 now fits in your pocket. PCs, servers, and other computers proliferated both in business and the personal markets. In addition to the new hardware entire markets, social systems, and new modes of interpersonal communication moved onto the web and started evolving rapidly. These changes cause a massive explosion of data and a need to analyze and understand that data. Taken together this resulted in an entirely different landscape for database storage, new solutions were needed. A number of novel solutions stepped up and eventually a category called NoSQL emerged. The new market forces inspired the CAP theorem and the heated debate of BASE vs. ACID. But in essence this was simply the market looking at what to trade off to meet these new demands. These new database systems shared many qualities in common. There were designed to address massive amounts of data, millions of requests per second, and scale out across multiple systems. The first large-scale and successful solution was Dynamo, Amazon's distributed key/value database. Dynamo essentially took the next logical step and added a twist. Dynamo was to be the database of record, it would be distributed, data would be partitioned across many nodes, and it would tolerate failure by avoiding single points of failure. Amazon did this because they recognized that the majority of the dynamic content they provided to customers visiting their web store front didn't require the services of an RDBMS. The queries were simple, key/value look-ups or simple range queries with only a few queries that required more complex joins. They set about to use relational technology only in places where it was the best solution for the task, places like accounting and order fulfillment, but not in the myriad of other situations. The success of Dynamo, and it's design, inspired the next generation of Non-SQL, distributed database solutions including Cassandra, Riak and Voldemort. The problem their designers set out to solve was, "reliability at massive scale" so the first focal point was distributed database algorithms. Underneath Dynamo there is a local transactional database; either Berkeley DB, Berkeley DB Java Edition, MySQL or an in-memory key/value data structure. Dynamo was an evolution of local key/value storage onto networks. Cassandra, Riak, and Voldemort all faced similar design decisions and one, Voldemort, choose Berkeley DB Java Edition for it's node-local storage. Riak at first was entirely in-memory, but has recently added write-once, append-only log-based on-disk storage similar type of storage as Berkeley DB except that it is based on a hash table which must reside entirely in-memory rather than a btree which can live in-memory or on disk. Berkeley DB evolved too, we added high availability (HA) and a replication manager that makes it easy to setup replica groups. Berkeley DB's replication doesn't partitioned the data, every node keeps an entire copy of the database. For consistency, there is a single node where writes are committed first - a master - then those changes are delivered to the replica nodes as log records. Applications can choose to wait until all nodes are consistent, or fire and forget allowing Berkeley DB to eventually become consistent. Berkeley DB's HA scales-out quite well for read-intensive applications and also effectively eliminates the central point of failure by allowing replica nodes to be elected (using a PAXOS algorithm) to mastership if the master should fail. This implementation covers a wide variety of use cases. MemcacheDB is a server that implements the Memcache network protocol but uses Berkeley DB for storage and HA to replicate the cache state across all the nodes in the cache group. Google Accounts, the user authentication layer for all Google properties, was until recently running Berkeley DB HA. That scaled to a globally distributed system. That said, most NoSQL solutions try to partition (shard) data across nodes in the replication group and some allow writes as well as reads at any node, Berkeley DB HA does not. So, is Berkeley DB a "NoSQL" solution? Not really, but it certainly is a component of many of the existing NoSQL solutions out there. Forgetting all the noise about how NoSQL solutions are complex distributed databases when you boil them down to a single node you still have to store the data to some form of stable local storage. DBMs solved that problem a long time ago. NoSQL has more to do with the layers on top of the DBM; the distributed, sometimes-consistent, partitioned, scale-out storage that manage key/value or document sets and generally have some form of simple HTTP/REST-style network API. Does Berkeley DB do that? Not really. Is Berkeley DB a "NoSQL" solution today? Nope, but it's the most robust solution on which to build such a system. Re-inventing the node-local data storage isn't easy. A lot of people are starting to come to appreciate the sophisticated features found in Berkeley DB, even mimic them in some cases. Could Berkeley DB grow into a NoSQL solution? Absolutely. Our key/value API could be extended over the net using any of a number of existing network protocols such as memcache or HTTP/REST. We could adapt our node-local data partitioning out over replicated nodes. We even have a nice query language and cost-based query optimizer in our BDB XML product that we could reuse were we to build out a document-based NoSQL-style product. XML and JSON are not so different that we couldn't adapt one to work with the other interchangeably. Without too much effort we could add what's missing, we could jump into this No SQL market withing a single product development cycle. Why isn't Berkeley DB already a NoSQL solution? Why aren't we working on it? Why indeed...

    Read the article

  • MySQL – Scalability on Amazon RDS: Scale out to multiple RDS instances

    - by Pinal Dave
    Today, I’d like to discuss getting better MySQL scalability on Amazon RDS. The question of the day: “What can you do when a MySQL database needs to scale write-intensive workloads beyond the capabilities of the largest available machine on Amazon RDS?” Let’s take a look. In a typical EC2/RDS set-up, users connect to app servers from their mobile devices and tablets, computers, browsers, etc.  Then app servers connect to an RDS instance (web/cloud services) and in some cases they might leverage some read-only replicas.   Figure 1. A typical RDS instance is a single-instance database, with read replicas.  This is not very good at handling high write-based throughput. As your application becomes more popular you can expect an increasing number of users, more transactions, and more accumulated data.  User interactions can become more challenging as the application adds more sophisticated capabilities. The result of all this positive activity: your MySQL database will inevitably begin to experience scalability pressures. What can you do? Broadly speaking, there are four options available to improve MySQL scalability on RDS. 1. Larger RDS Instances – If you’re not already using the maximum available RDS instance, you can always scale up – to larger hardware.  Bigger CPUs, more compute power, more memory et cetera. But the largest available RDS instance is still limited.  And they get expensive. “High-Memory Quadruple Extra Large DB Instance”: 68 GB of memory 26 ECUs (8 virtual cores with 3.25 ECUs each) 64-bit platform High I/O Capacity Provisioned IOPS Optimized: 1000Mbps 2. Provisioned IOPs – You can get provisioned IOPs and higher throughput on the I/O level. However, there is a hard limit with a maximum instance size and maximum number of provisioned IOPs you can buy from Amazon and you simply cannot scale beyond these hardware specifications. 3. Leverage Read Replicas – If your application permits, you can leverage read replicas to offload some reads from the master databases. But there are a limited number of replicas you can utilize and Amazon generally requires some modifications to your existing application. And read-replicas don’t help with write-intensive applications. 4. Multiple Database Instances – Amazon offers a fourth option: “You can implement partitioning,thereby spreading your data across multiple database Instances” (Link) However, Amazon does not offer any guidance or facilities to help you with this. “Multiple database instances” is not an RDS feature.  And Amazon doesn’t explain how to implement this idea. In fact, when asked, this is the response on an Amazon forum: Q: Is there any documents that describe the partition DB across multiple RDS? I need to use DB with more 1TB but exist a limitation during the create process, but I read in the any FAQ that you need to partition database, but I don’t find any documents that describe it. A: “DB partitioning/sharding is not an official feature of Amazon RDS or MySQL, but a technique to scale out database by using multiple database instances. The appropriate way to split data depends on the characteristics of the application or data set. Therefore, there is no concrete and specific guidance.” So now what? The answer is to scale out with ScaleBase. Amazon RDS with ScaleBase: What you get – MySQL Scalability! ScaleBase is specifically designed to scale out a single MySQL RDS instance into multiple MySQL instances. Critically, this is accomplished with no changes to your application code.  Your application continues to “see” one database.   ScaleBase does all the work of managing and enforcing an optimized data distribution policy to create multiple MySQL instances. With ScaleBase, data distribution, transactions, concurrency control, and two-phase commit are all 100% transparent and 100% ACID-compliant, so applications, services and tooling continue to interact with your distributed RDS as if it were a single MySQL instance. The result: now you can cost-effectively leverage multiple MySQL RDS instance to scale out write-intensive workloads to an unlimited number of users, transactions, and data. Amazon RDS with ScaleBase: What you keep – Everything! And how does this change your Amazon environment? 1. Keep your application, unchanged – There is no change your application development life-cycle at all.  You still use your existing development tools, frameworks and libraries.  Application quality assurance and testing cycles stay the same. And, critically, you stay with an ACID-compliant MySQL environment. 2. Keep your RDS value-added services – The value-added services that you rely on are all still available. Amazon will continue to handle database maintenance and updates for you. You can still leverage High Availability via Multi A-Z.  And, if it benefits youra application throughput, you can still use read replicas. 3. Keep your RDS administration – Finally the RDS monitoring and provisioning tools you rely on still work as they did before. With your one large MySQL instance, now split into multiple instances, you can actually use less expensive, smallersmaller available RDS hardware and continue to see better database performance. Conclusion Amazon RDS is a tremendous service, but it doesn’t offer solutions to scale beyond a single MySQL instance. Larger RDS instances get more expensive.  And when you max-out on the available hardware, you’re stuck.  Amazon recommends scaling out your single instance into multiple instances for transaction-intensive apps, but offers no services or guidance to help you. This is where ScaleBase comes in to save the day. It gives you a simple and effective way to create multiple MySQL RDS instances, while removing all the complexities typically caused by “DIY” sharding andwith no changes to your applications . With ScaleBase you continue to leverage the AWS/RDS ecosystem: commodity hardware and value added services like read replicas, multi A-Z, maintenance/updates and administration with monitoring tools and provisioning. SCALEBASE ON AMAZON If you’re curious to try ScaleBase on Amazon, it can be found here – Download NOW. Reference: Pinal Dave (http://blog.sqlauthority.com)Filed under: MySQL, PostADay, SQL, SQL Authority, SQL Optimization, SQL Performance, SQL Query, SQL Server, SQL Tips and Tricks, T SQL

    Read the article

  • Using Table-Valued Parameters in SQL Server

    - by Jesse
    I work with stored procedures in SQL Server pretty frequently and have often found myself with a need to pass in a list of values at run-time. Quite often this list contains a set of ids on which the stored procedure needs to operate the size and contents of which are not known at design time. In the past I’ve taken the collection of ids (which are usually integers), converted them to a string representation where each value is separated by a comma and passed that string into a VARCHAR parameter of a stored procedure. The body of the stored procedure would then need to parse that string into a table variable which could be easily consumed with set-based logic within the rest of the stored procedure. This approach works pretty well but the VARCHAR variable has always felt like an un-wanted “middle man” in this scenario. Of course, I could use a BULK INSERT operation to load the list of ids into a temporary table that the stored procedure could use, but that approach seems heavy-handed in situations where the list of values is usually going to contain only a few dozen values. Fortunately SQL Server 2008 introduced the concept of table-valued parameters which effectively eliminates the need for the clumsy middle man VARCHAR parameter. Example: Customer Transaction Summary Report Let’s say we have a report that can summarize the the transactions that we’ve conducted with customers over a period of time. The report returns a pretty simple dataset containing one row per customer with some key metrics about how much business that customer has conducted over the date range for which the report is being run. Sometimes the report is run for a single customer, sometimes it’s run for all customers, and sometimes it’s run for a handful of customers (i.e. a salesman runs it for the customers that fall into his sales territory). This report can be invoked from a website on-demand, or it can be scheduled for periodic delivery to certain users via SQL Server Reporting Services. Because the report can be created from different places and the query to generate the report is complex it’s been packed into a stored procedure that accepts three parameters: @startDate – The beginning of the date range for which the report should be run. @endDate – The end of the date range for which the report should be run. @customerIds – The customer Ids for which the report should be run. Obviously, the @startDate and @endDate parameters are DATETIME variables. The @customerIds parameter, however, needs to contain a list of the identity values (primary key) from the Customers table representing the customers that were selected for this particular run of the report. In prior versions of SQL Server we might have made this parameter a VARCHAR variable, but with SQL Server 2008 we can make it into a table-valued parameter. Defining And Using The Table Type In order to use a table-valued parameter, we first need to tell SQL Server about what the table will look like. We do this by creating a user defined type. For the purposes of this stored procedure we need a very simple type to model a table variable with a single integer column. We can create a generic type called ‘IntegerListTableType’ like this: CREATE TYPE IntegerListTableType AS TABLE (Value INT NOT NULL) Once defined, we can use this new type to define the @customerIds parameter in the signature of our stored procedure. The parameter list for the stored procedure definition might look like: 1: CREATE PROCEDURE dbo.rpt_CustomerTransactionSummary 2: @starDate datetime, 3: @endDate datetime, 4: @customerIds IntegerListTableTableType READONLY   Note the ‘READONLY’ statement following the declaration of the @customerIds parameter. SQL Server requires any table-valued parameter be marked as ‘READONLY’ and no DML (INSERT/UPDATE/DELETE) statements can be performed on a table-valued parameter within the routine in which it’s used. Aside from the DML restriction, however, you can do pretty much anything with a table-valued parameter as you could with a normal TABLE variable. With the user defined type and stored procedure defined as above, we could invoke like this: 1: DECLARE @cusomterIdList IntegerListTableType 2: INSERT @customerIdList VALUES (1) 3: INSERT @customerIdList VALUES (2) 4: INSERT @customerIdList VALUES (3) 5:  6: EXEC dbo.rpt_CustomerTransationSummary 7: @startDate = '2012-05-01', 8: @endDate = '2012-06-01' 9: @customerIds = @customerIdList   Note that we can simply declare a variable of type ‘IntegerListTableType’ just like any other normal variable and insert values into it just like a TABLE variable. We could also populate the variable with a SELECT … INTO or INSERT … SELECT statement if desired. Using The Table-Valued Parameter With ADO .NET Invoking a stored procedure with a table-valued parameter from ADO .NET is as simple as building a DataTable and passing it in as the Value of a SqlParameter. Here’s some example code for how we would construct the SqlParameter for the @customerIds parameter in our stored procedure: 1: var customerIdsParameter = new SqlParameter(); 2: customerIdParameter.Direction = ParameterDirection.Input; 3: customerIdParameter.TypeName = "IntegerListTableType"; 4: customerIdParameter.Value = selectedCustomerIds.ToIntegerListDataTable("Value");   All we’re doing here is new’ing up an instance of SqlParameter, setting the pamameters direction, specifying the name of the User Defined Type that this parameter uses, and setting its value. We’re assuming here that we have an IEnumerable<int> variable called ‘selectedCustomerIds’ containing all of the customer Ids for which the report should be run. The ‘ToIntegerListDataTable’ method is an extension method of the IEnumerable<int> type that looks like this: 1: public static DataTable ToIntegerListDataTable(this IEnumerable<int> intValues, string columnName) 2: { 3: var intergerListDataTable = new DataTable(); 4: intergerListDataTable.Columns.Add(columnName); 5: foreach(var intValue in intValues) 6: { 7: var nextRow = intergerListDataTable.NewRow(); 8: nextRow[columnName] = intValue; 9: intergerListDataTable.Rows.Add(nextRow); 10: } 11:  12: return intergerListDataTable; 13: }   Since the ‘IntegerListTableType’ has a single int column called ‘Value’, we pass that in for the ‘columnName’ parameter to the extension method. The method creates a new single-columned DataTable using the provided column name then iterates over the items in the IEnumerable<int> instance adding one row for each value. We can then use this SqlParameter instance when invoking the stored procedure just like we would use any other parameter. Advanced Functionality Using passing a list of integers into a stored procedure is a very simple usage scenario for the table-valued parameters feature, but I’ve found that it covers the majority of situations where I’ve needed to pass a collection of data for use in a query at run-time. I should note that BULK INSERT feature still makes sense for passing large amounts of data to SQL Server for processing. MSDN seems to suggest that 1000 rows of data is the tipping point where the overhead of a BULK INSERT operation can pay dividends. I should also note here that table-valued parameters can be used to deal with more complex data structures than single-columned tables of integers. A User Defined Type that backs a table-valued parameter can use things like identities and computed columns. That said, using some of these more advanced features might require the use the SqlDataRecord and SqlMetaData classes instead of a simple DataTable. Erland Sommarskog has a great article on his website that describes when and how to use these classes for table-valued parameters. What About Reporting Services? Earlier in the post I referenced the fact that our example stored procedure would be called from both a web application and a SQL Server Reporting Services report. Unfortunately, using table-valued parameters from SSRS reports can be a bit tricky and warrants its own blog post which I’ll be putting together and posting sometime in the near future.

    Read the article

  • SQL SERVER – Weekly Series – Memory Lane – #032

    - by Pinal Dave
    Here is the list of selected articles of SQLAuthority.com across all these years. Instead of just listing all the articles I have selected a few of my most favorite articles and have listed them here with additional notes below it. Let me know which one of the following is your favorite article from memory lane. 2007 Complete Series of Database Coding Standards and Guidelines SQL SERVER Database Coding Standards and Guidelines – Introduction SQL SERVER – Database Coding Standards and Guidelines – Part 1 SQL SERVER – Database Coding Standards and Guidelines – Part 2 SQL SERVER Database Coding Standards and Guidelines Complete List Download Explanation and Example – SELF JOIN When all of the data you require is contained within a single table, but data needed to extract is related to each other in the table itself. Examples of this type of data relate to Employee information, where the table may have both an Employee’s ID number for each record and also a field that displays the ID number of an Employee’s supervisor or manager. To retrieve the data tables are required to relate/join to itself. Insert Multiple Records Using One Insert Statement – Use of UNION ALL This is very interesting question I have received from new developer. How can I insert multiple values in table using only one insert? Now this is interesting question. When there are multiple records are to be inserted in the table following is the common way using T-SQL. Function to Display Current Week Date and Day – Weekly Calendar Straight blog post with script to find current week date and day based on the parameters passed in the function.  2008 In my beginning years, I have almost same confusion as many of the developer had in their earlier years. Here are two of the interesting question which I have attempted to answer in my early year. Even if you are experienced developer may be you will still like to read following two questions: Order Of Column In Index Order of Conditions in WHERE Clauses Example of DISTINCT in Aggregate Functions Have you ever used DISTINCT with the Aggregation Function? Here is a simple example about how users can do it. Create a Comma Delimited List Using SELECT Clause From Table Column Straight to script example where I explained how to do something easy and quickly. Compound Assignment Operators SQL SERVER 2008 has introduced new concept of Compound Assignment Operators. Compound Assignment Operators are available in many other programming languages for quite some time. Compound Assignment Operators is operator where variables are operated upon and assigned on the same line. PIVOT and UNPIVOT Table Examples Here is a very interesting question – the answer to the question can be YES or NO both. “If we PIVOT any table and UNPIVOT that table do we get our original table?” Read the blog post to get the explanation of the question above. 2009 What is Interim Table – Simple Definition of Interim Table The interim table is a table that is generated by joining two tables and not the final result table. In other words, when two tables are joined they create an interim table as resultset but the resultset is not final yet. It may be possible that more tables are about to join on the interim table, and more operations are still to be applied on that table (e.g. Order By, Having etc). Besides, it may be possible that there is no interim table; sometimes final table is what is generated when the query is run. 2010 Stored Procedure and Transactions If Stored Procedure is transactional then, it should roll back complete transactions when it encounters any errors. Well, that does not happen in this case, which proves that Stored Procedure does not only provide just the transactional feature to a batch of T-SQL. Generate Database Script for SQL Azure When talking about SQL Azure the most common complaint I hear is that the script generated from stand-along SQL Server database is not compatible with SQL Azure. This was true for some time for sure but not any more. If you have SQL Server 2008 R2 installed you can follow the guideline below to generate a script which is compatible with SQL Azure. Convert IN to EXISTS – Performance Talk It is NOT necessary that every time when IN is replaced by EXISTS it gives better performance. However, in our case listed above it does for sure give better performance. You can read about this subject in the associated blog post. Subquery or Join – Various Options – SQL Server Engine Knows the Best Every single time whenever there is a performance tuning exercise, I hear the conversation from developer where some prefer subquery and some prefer join. In this two part blog post, I explain the same in the detail with examples. Part 1 | Part 2 Merge Operations – Insert, Update, Delete in Single Execution MERGE is a new feature that provides an efficient way to do multiple DML operations. In earlier versions of SQL Server, we had to write separate statements to INSERT, UPDATE, or DELETE data based on certain conditions; however, at present, by using the MERGE statement, we can include the logic of such data changes in one statement that even checks when the data is matched and then just update it, and similarly, when the data is unmatched, it is inserted. 2011 Puzzle – Statistics are not updated but are Created Once Here is the quick scenario about my setup. Create Table Insert 1000 Records Check the Statistics Now insert 10 times more 10,000 indexes Check the Statistics – it will be NOT updated – WHY? Question to You – When to use Function and When to use Stored Procedure Personally, I believe that they are both different things - they cannot be compared. I can say, it will be like comparing apples and oranges. Each has its own unique use. However, they can be used interchangeably at many times and in real life (i.e., production environment). I have personally seen both of these being used interchangeably many times. This is the precise reason for asking this question. 2012 In year 2012 I had two interesting series ran on the blog. If there is no fun in learning, the learning becomes a burden. For the same reason, I had decided to build a three part quiz around SEQUENCE. The quiz was to identify the next value of the sequence. I encourage all of you to take part in this fun quiz. Guess the Next Value – Puzzle 1 Guess the Next Value – Puzzle 2 Guess the Next Value – Puzzle 3 Guess the Next Value – Puzzle 4 Simple Example to Configure Resource Governor – Introduction to Resource Governor Resource Governor is a feature which can manage SQL Server Workload and System Resource Consumption. We can limit the amount of CPU and memory consumption by limiting /governing /throttling on the SQL Server. If there are different workloads running on SQL Server and each of the workload needs different resources or when workloads are competing for resources with each other and affecting the performance of the whole server resource governor is a very important task. Tricks to Replace SELECT * with Column Names – SQL in Sixty Seconds #017 – Video  Retrieves unnecessary columns and increases network traffic When a new columns are added views needs to be refreshed manually Leads to usage of sub-optimal execution plan Uses clustered index in most of the cases instead of using optimal index It is difficult to debug SQL SERVER – Load Generator – Free Tool From CodePlex The best part of this SQL Server Load Generator is that users can run multiple simultaneous queries again SQL Server using different login account and different application name. The interface of the tool is extremely easy to use and very intuitive as well. A Puzzle – Swap Value of Column Without Case Statement Let us assume there is a single column in the table called Gender. The challenge is to write a single update statement which will flip or swap the value in the column. For example if the value in the gender column is ‘male’ swap it with ‘female’ and if the value is ‘female’ swap it with ‘male’. Reference: Pinal Dave (http://blog.sqlauthority.com) Filed under: Memory Lane, PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Integrating Oracle Hyperion Smart View Data Queries with MS Word and Power Point

    - by Andreea Vaduva
    Untitled Document table { border: thin solid; } Most Smart View users probably appreciate that they can use just one add-in to access data from the different sources they might work with, like Oracle Essbase, Oracle Hyperion Planning, Oracle Hyperion Financial Management and others. But not all of them are aware of the options to integrate data analyses not only in Excel, but also in MS Word or Power Point. While in the past, copying and pasting single numbers or tables from a recent analysis in Excel made the pasted content a static snapshot, copying so called Data Points now creates dynamic, updateable references to the data source. It also provides additional nice features, which can make life easier and less stressful for Smart View users. So, how does this option work: after building an ad-hoc analysis with Smart View as usual in an Excel worksheet, any area including data cells/numbers from the database can be highlighted in order to copy data points - even single data cells only.   TIP It is not necessary to highlight and copy the row or column descriptions   Next from the Smart View ribbon select Copy Data Point. Then transfer to the Word or Power Point document into which the selected content should be copied. Note that in these Office programs you will find a menu item Smart View;from it select the Paste Data Point icon. The copied details from the Excel report will be pasted, but showing #NEED_REFRESH in the data cells instead of the original numbers. =After clicking the Refresh icon on the Smart View menu the data will be retrieved and displayed. (Maybe at that moment a login window pops up and you need to provide your credentials.) It works in the same way if you just copy one single number without any row or column descriptions, for example in order to incorporate it into a continuous text: Before refresh: After refresh: From now on for any subsequent updates of the data shown in your documents you only need to refresh data by clicking the Refresh button on the Smart View menu, without copying and pasting the context or content again. As you might realize, trying out this feature on your own, there won’t be any Point of View shown in the Office document. Also you have seen in the example, where only a single data cell was copied, that there aren’t any member names or row/column descriptions copied, which are usually required in an ad-hoc report in order to exactly define where data comes from or how data is queried from the source. Well, these definitions are not visible, but they are transferred to the Word or Power Point document as well. They are stored in the background for each individual data cell copied and can be made visible by double-clicking the data cell as shown in the following screen shot (but which is taken from another context).   So for each cell/number the complete connection information is stored along with the exact member/cell intersection from the database. And that’s not all: you have the chance now to exchange the members originally selected in the Point of View (POV) in the Excel report. Remember, at that time we had the following selection:   By selecting the Manage POV option from the Smart View meny in Word or Power Point…   … the following POV Manager – Queries window opens:   You can now change your selection for each dimension from the original POV by either double-clicking the dimension member in the lower right box under POV: or by selecting the Member Selector icon on the top right hand side of the window. After confirming your changes you need to refresh your document again. Be aware, that this will update all (!) numbers taken from one and the same original Excel sheet, even if they appear in different locations in your Office document, reflecting your recent changes in the POV. TIP Build your original report already in a way that dimensions you might want to change from within Word or Power Point are placed in the POV. And there is another really nice feature I wouldn’t like to miss mentioning: Using Dynamic Data Points in the way described above, you will never miss or need to search again for your original Excel sheet from which values were taken and copied as data points into an Office document. Because from even only one single data cell Smart View is able to recreate the entire original report content with just a few clicks: Select one of the numbers from within your Word or Power Point document by double-clicking.   Then select the Visualize in Excel option from the Smart View menu. Excel will open and Smart View will rebuild the entire original report, including POV settings, and retrieve all data from the most recent actual state of the database. (It might be necessary to provide your credentials before data is displayed.) However, in order to make this work, an active online connection to your databases on the server is necessary and at least read access to the retrieved data. But apart from this, your newly built Excel report is fully functional for ad-hoc analysis and can be used in the common way for drilling, pivoting and all the other known functions and features. So far about embedding Dynamic Data Points into Office documents and linking them back into Excel worksheets. You can apply this in the described way with ad-hoc analyses directly on Essbase databases or using Hyperion Planning and Hyperion Financial Management ad-hoc web forms. If you are also interested in other new features and smart enhancements in Essbase or Hyperion Planning stay tuned for coming articles or check our training courses and web presentations. You can find general information about offerings for the Essbase and Planning curriculum or other Oracle-Hyperion products here (please make sure to select your country/region at the top of this page) or in the OU Learning paths section , where Planning, Essbase and other Hyperion products can be found under the Fusion Middleware heading (again, please select the right country/region). Or drop me a note directly: [email protected] . About the Author: Bernhard Kinkel started working for Hyperion Solutions as a Presales Consultant and Consultant in 1998 and moved to Hyperion Education Services in 1999. He joined Oracle University in 2007 where he is a Principal Education Consultant. Based on these many years of working with Hyperion products he has detailed product knowledge across several versions. He delivers both classroom and live virtual courses. His areas of expertise are Oracle/Hyperion Essbase, Oracle Hyperion Planning and Hyperion Web Analysis.  

    Read the article

  • Scheduling thread tiles with C++ AMP

    - by Daniel Moth
    This post assumes you are totally comfortable with, what some of us call, the simple model of C++ AMP, i.e. you could write your own matrix multiplication. We are now ready to explore the tiled model, which builds on top of the non-tiled one. Tiling the extent We know that when we pass a grid (which is just an extent under the covers) to the parallel_for_each call, it determines the number of threads to schedule and their index values (including dimensionality). For the single-, two-, and three- dimensional cases you can go a step further and subdivide the threads into what we call tiles of threads (others may call them thread groups). So here is a single-dimensional example: extent<1> e(20); // 20 units in a single dimension with indices from 0-19 grid<1> g(e);      // same as extent tiled_grid<4> tg = g.tile<4>(); …on the 3rd line we subdivided the single-dimensional space into 5 single-dimensional tiles each having 4 elements, and we captured that result in a concurrency::tiled_grid (a new class in amp.h). Let's move on swiftly to another example, in pictures, this time 2-dimensional: So we start on the left with a grid of a 2-dimensional extent which has 8*6=48 threads. We then have two different examples of tiling. In the first case, in the middle, we subdivide the 48 threads into tiles where each has 4*3=12 threads, hence we have 2*2=4 tiles. In the second example, on the right, we subdivide the original input into tiles where each has 2*2=4 threads, hence we have 4*3=12 tiles. Notice how you can play with the tile size and achieve different number of tiles. The numbers you pick must be such that the original total number of threads (in our example 48), remains the same, and every tile must have the same size. Of course, you still have no clue why you would do that, but stick with me. First, we should see how we can use this tiled_grid, since the parallel_for_each function that we know expects a grid. Tiled parallel_for_each and tiled_index It turns out that we have additional overloads of parallel_for_each that accept a tiled_grid instead of a grid. However, those overloads, also expect that the lambda you pass in accepts a concurrency::tiled_index (new in amp.h), not an index<N>. So how is a tiled_index different to an index? A tiled_index object, can have only 1 or 2 or 3 dimensions (matching exactly the tiled_grid), and consists of 4 index objects that are accessible via properties: global, local, tile_origin, and tile. The global index is the same as the index we know and love: the global thread ID. The local index is the local thread ID within the tile. The tile_origin index returns the global index of the thread that is at position 0,0 of this tile, and the tile index is the position of the tile in relation to the overall grid. Confused? Here is an example accompanied by a picture that hopefully clarifies things: array_view<int, 2> data(8, 6, p_my_data); parallel_for_each(data.grid.tile<2,2>(), [=] (tiled_index<2,2> t_idx) restrict(direct3d) { /* todo */ }); Given the code above and the picture on the right, what are the values of each of the 4 index objects that the t_idx variables exposes, when the lambda is executed by T (highlighted in the picture on the right)? If you can't work it out yourselves, the solution follows: t_idx.global       = index<2> (6,3) t_idx.local          = index<2> (0,1) t_idx.tile_origin = index<2> (6,2) t_idx.tile             = index<2> (3,1) Don't move on until you are comfortable with this… the picture really helps, so use it. Tiled Matrix Multiplication Example – part 1 Let's paste here the C++ AMP matrix multiplication example, bolding the lines we are going to change (can you guess what the changes will be?) 01: void MatrixMultiplyTiled_Part1(vector<float>& vC, const vector<float>& vA, const vector<float>& vB, int M, int N, int W) 02: { 03: 04: array_view<const float,2> a(M, W, vA); 05: array_view<const float,2> b(W, N, vB); 06: array_view<writeonly<float>,2> c(M, N, vC); 07: parallel_for_each(c.grid, 08: [=](index<2> idx) restrict(direct3d) { 09: 10: int row = idx[0]; int col = idx[1]; 11: float sum = 0.0f; 12: for(int i = 0; i < W; i++) 13: sum += a(row, i) * b(i, col); 14: c[idx] = sum; 15: }); 16: } To turn this into a tiled example, first we need to decide our tile size. Let's say we want each tile to be 16*16 (which assumes that we'll have at least 256 threads to process, and that c.grid.extent.size() is divisible by 256, and moreover that c.grid.extent[0] and c.grid.extent[1] are divisible by 16). So we insert at line 03 the tile size (which must be a compile time constant). 03: static const int TS = 16; ...then we need to tile the grid to have tiles where each one has 16*16 threads, so we change line 07 to be as follows 07: parallel_for_each(c.grid.tile<TS,TS>(), ...that means that our index now has to be a tiled_index with the same characteristics as the tiled_grid, so we change line 08 08: [=](tiled_index<TS, TS> t_idx) restrict(direct3d) { ...which means, without changing our core algorithm, we need to be using the global index that the tiled_index gives us access to, so we insert line 09 as follows 09: index<2> idx = t_idx.global; ...and now this code just works and it is tiled! Closing thoughts on part 1 The process we followed just shows the mechanical transformation that can take place from the simple model to the tiled model (think of this as step 1). In fact, when we wrote the matrix multiplication example originally, the compiler was doing this mechanical transformation under the covers for us (and it has additional smarts to deal with the cases where the total number of threads scheduled cannot be divisible by the tile size). The point is that the thread scheduling is always tiled, even when you use the non-tiled model. But with this mechanical transformation, we haven't gained anything… Hint: our goal with explicitly using the tiled model is to gain even more performance. In the next post, we'll evolve this further (beyond what the compiler can automatically do for us, in this first release), so you can see the full usage of the tiled model and its benefits… Comments about this post by Daniel Moth welcome at the original blog.

    Read the article

< Previous Page | 109 110 111 112 113 114 115 116 117 118 119 120  | Next Page >