Search Results

Search found 3481 results on 140 pages for 'convex optimization'.

Page 35/140 | < Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42  | Next Page >

  • SQL SERVER – Fundamentals of Columnstore Index

    - by pinaldave
    There are two kind of storage in database. Row Store and Column Store. Row store does exactly as the name suggests – stores rows of data on a page – and column store stores all the data in a column on the same page. These columns are much easier to search – instead of a query searching all the data in an entire row whether the data is relevant or not, column store queries need only to search much lesser number of the columns. This means major increases in search speed and hard drive use. Additionally, the column store indexes are heavily compressed, which translates to even greater memory and faster searches. I am sure this looks very exciting and it does not mean that you convert every single index from row store to column store index. One has to understand the proper places where to use row store or column store indexes. Let us understand in this article what is the difference in Columnstore type of index. Column store indexes are run by Microsoft’s VertiPaq technology. However, all you really need to know is that this method of storing data is columns on a single page is much faster and more efficient. Creating a column store index is very easy, and you don’t have to learn new syntax to create them. You just need to specify the keyword “COLUMNSTORE” and enter the data as you normally would. Keep in mind that once you add a column store to a table, though, you cannot delete, insert or update the data – it is READ ONLY. However, since column store will be mainly used for data warehousing, this should not be a big problem. You can always use partitioning to avoid rebuilding the index. A columnstore index stores each column in a separate set of disk pages, rather than storing multiple rows per page as data traditionally has been stored. The difference between column store and row store approaches is illustrated below: In case of the row store indexes multiple pages will contain multiple rows of the columns spanning across multiple pages. In case of column store indexes multiple pages will contain multiple single columns. This will lead only the columns needed to solve a query will be fetched from disk. Additionally there is good chance that there will be redundant data in a single column which will further help to compress the data, this will have positive effect on buffer hit rate as most of the data will be in memory and due to same it will not need to be retrieved. Let us see small example of how columnstore index improves the performance of the query on a large table. As a first step let us create databaseset which is large enough to show performance impact of columnstore index. The time taken to create sample database may vary on different computer based on the resources. USE AdventureWorks GO -- Create New Table CREATE TABLE [dbo].[MySalesOrderDetail]( [SalesOrderID] [int] NOT NULL, [SalesOrderDetailID] [int] NOT NULL, [CarrierTrackingNumber] [nvarchar](25) NULL, [OrderQty] [smallint] NOT NULL, [ProductID] [int] NOT NULL, [SpecialOfferID] [int] NOT NULL, [UnitPrice] [money] NOT NULL, [UnitPriceDiscount] [money] NOT NULL, [LineTotal] [numeric](38, 6) NOT NULL, [rowguid] [uniqueidentifier] NOT NULL, [ModifiedDate] [datetime] NOT NULL ) ON [PRIMARY] GO -- Create clustered index CREATE CLUSTERED INDEX [CL_MySalesOrderDetail] ON [dbo].[MySalesOrderDetail] ( [SalesOrderDetailID]) GO -- Create Sample Data Table -- WARNING: This Query may run upto 2-10 minutes based on your systems resources INSERT INTO [dbo].[MySalesOrderDetail] SELECT S1.* FROM Sales.SalesOrderDetail S1 GO 100 Now let us do quick performance test. I have kept STATISTICS IO ON for measuring how much IO following queries take. In my test first I will run query which will use regular index. We will note the IO usage of the query. After that we will create columnstore index and will measure the IO of the same. -- Performance Test -- Comparing Regular Index with ColumnStore Index USE AdventureWorks GO SET STATISTICS IO ON GO -- Select Table with regular Index SELECT ProductID, SUM(UnitPrice) SumUnitPrice, AVG(UnitPrice) AvgUnitPrice, SUM(OrderQty) SumOrderQty, AVG(OrderQty) AvgOrderQty FROM [dbo].[MySalesOrderDetail] GROUP BY ProductID ORDER BY ProductID GO -- Table 'MySalesOrderDetail'. Scan count 1, logical reads 342261, physical reads 0, read-ahead reads 0. -- Create ColumnStore Index CREATE NONCLUSTERED COLUMNSTORE INDEX [IX_MySalesOrderDetail_ColumnStore] ON [MySalesOrderDetail] (UnitPrice, OrderQty, ProductID) GO -- Select Table with Columnstore Index SELECT ProductID, SUM(UnitPrice) SumUnitPrice, AVG(UnitPrice) AvgUnitPrice, SUM(OrderQty) SumOrderQty, AVG(OrderQty) AvgOrderQty FROM [dbo].[MySalesOrderDetail] GROUP BY ProductID ORDER BY ProductID GO It is very clear from the results that query is performance extremely fast after creating ColumnStore Index. The amount of the pages it has to read to run query is drastically reduced as the column which are needed in the query are stored in the same page and query does not have to go through every single page to read those columns. If we enable execution plan and compare we can see that column store index performance way better than regular index in this case. Let us clean up the database. -- Cleanup DROP INDEX [IX_MySalesOrderDetail_ColumnStore] ON [dbo].[MySalesOrderDetail] GO TRUNCATE TABLE dbo.MySalesOrderDetail GO DROP TABLE dbo.MySalesOrderDetail GO In future posts we will see cases where Columnstore index is not appropriate solution as well few other tricks and tips of the columnstore index. Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: Pinal Dave, PostADay, SQL, SQL Authority, SQL Index, SQL Optimization, SQL Performance, SQL Query, SQL Scripts, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Optimizing collision engine bottleneck

    - by Vittorio Romeo
    Foreword: I'm aware that optimizing this bottleneck is not a necessity - the engine is already very fast. I, however, for fun and educational purposes, would love to find a way to make the engine even faster. I'm creating a general-purpose C++ 2D collision detection/response engine, with an emphasis on flexibility and speed. Here's a very basic diagram of its architecture: Basically, the main class is World, which owns (manages memory) of a ResolverBase*, a SpatialBase* and a vector<Body*>. SpatialBase is a pure virtual class which deals with broad-phase collision detection. ResolverBase is a pure virtual class which deals with collision resolution. The bodies communicate to the World::SpatialBase* with SpatialInfo objects, owned by the bodies themselves. There currenly is one spatial class: Grid : SpatialBase, which is a basic fixed 2D grid. It has it's own info class, GridInfo : SpatialInfo. Here's how its architecture looks: The Grid class owns a 2D array of Cell*. The Cell class contains two collection of (not owned) Body*: a vector<Body*> which contains all the bodies that are in the cell, and a map<int, vector<Body*>> which contains all the bodies that are in the cell, divided in groups. Bodies, in fact, have a groupId int that is used for collision groups. GridInfo objects also contain non-owning pointers to the cells the body is in. As I previously said, the engine is based on groups. Body::getGroups() returns a vector<int> of all the groups the body is part of. Body::getGroupsToCheck() returns a vector<int> of all the groups the body has to check collision against. Bodies can occupy more than a single cell. GridInfo always stores non-owning pointers to the occupied cells. After the bodies move, collision detection happens. We assume that all bodies are axis-aligned bounding boxes. How broad-phase collision detection works: Part 1: spatial info update For each Body body: Top-leftmost occupied cell and bottom-rightmost occupied cells are calculated. If they differ from the previous cells, body.gridInfo.cells is cleared, and filled with all the cells the body occupies (2D for loop from the top-leftmost cell to the bottom-rightmost cell). body is now guaranteed to know what cells it occupies. For a performance boost, it stores a pointer to every map<int, vector<Body*>> of every cell it occupies where the int is a group of body->getGroupsToCheck(). These pointers get stored in gridInfo->queries, which is simply a vector<map<int, vector<Body*>>*>. body is now guaranteed to have a pointer to every vector<Body*> of bodies of groups it needs to check collision against. These pointers are stored in gridInfo->queries. Part 2: actual collision checks For each Body body: body clears and fills a vector<Body*> bodiesToCheck, which contains all the bodies it needs to check against. Duplicates are avoided (bodies can belong to more than one group) by checking if bodiesToCheck already contains the body we're trying to add. const vector<Body*>& GridInfo::getBodiesToCheck() { bodiesToCheck.clear(); for(const auto& q : queries) for(const auto& b : *q) if(!contains(bodiesToCheck, b)) bodiesToCheck.push_back(b); return bodiesToCheck; } The GridInfo::getBodiesToCheck() method IS THE BOTTLENECK. The bodiesToCheck vector must be filled for every body update because bodies could have moved meanwhile. It also needs to prevent duplicate collision checks. The contains function simply checks if the vector already contains a body with std::find. Collision is checked and resolved for every body in bodiesToCheck. That's it. So, I've been trying to optimize this broad-phase collision detection for quite a while now. Every time I try something else than the current architecture/setup, something doesn't go as planned or I make assumption about the simulation that later are proven to be false. My question is: how can I optimize the broad-phase of my collision engine maintaining the grouped bodies approach? Is there some kind of magic C++ optimization that can be applied here? Can the architecture be redesigned in order to allow for more performance? Actual implementation: SSVSCollsion Body.h, Body.cpp World.h, World.cpp Grid.h, Grid.cpp Cell.h, Cell.cpp GridInfo.h, GridInfo.cpp

    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

  • SQL SERVER – A Quick Look at Logging and Ideas around Logging

    - by pinaldave
    This blog post is written in response to the T-SQL Tuesday post on Logging. When someone talks about logging, personally I get lots of ideas about it. I have seen logging as a very generic term. Let me ask you this question first before I continue writing about logging. What is the first thing comes to your mind when you hear word “Logging”? Now ask the same question to the guy standing next to you. I am pretty confident that you will get  a different answer from different people. I decided to do this activity and asked 5 SQL Server person the same question. Question: What is the first thing comes to your mind when you hear the word “Logging”? Strange enough I got a different answer every single time. Let me just list what answer I got from my friends. Let us go over them one by one. Output Clause The very first person replied output clause. Pretty interesting answer to start with. I see what exactly he was thinking. SQL Server 2005 has introduced a new OUTPUT clause. OUTPUT clause has access to inserted and deleted tables (virtual tables) just like triggers. OUTPUT clause can be used to return values to client clause. OUTPUT clause can be used with INSERT, UPDATE, or DELETE to identify the actual rows affected by these statements. Here are some references for Output Clause: OUTPUT Clause Example and Explanation with INSERT, UPDATE, DELETE Reasons for Using Output Clause – Quiz Tips from the SQL Joes 2 Pros Development Series – Output Clause in Simple Examples Error Logs I was expecting someone to mention Error logs when it is about logging. The error log is the most looked place when there is any error either with the application or there is an error with the operating system. I have kept the policy to check my server’s error log every day. The reason is simple – enough time in my career I have figured out that when I am looking at error logs I find something which I was not expecting. There are cases, when I noticed errors in the error log and I fixed them before end user notices it. Other common practices I always tell my DBA friends to do is that when any error happens they should find relevant entries in the error logs and document the same. It is quite possible that they will see the same error in the error log  and able to fix the error based on the knowledge base which they have created. There can be many different kinds of error log files exists in SQL Server as well – 1) SQL Server Error Logs 2) Windows Event Log 3) SQL Server Agent Log 4) SQL Server Profile Log 5) SQL Server Setup Log etc. Here are some references for Error Logs: Recycle Error Log – Create New Log file without Server Restart SQL Error Messages Change Data Capture I got surprised with this answer. I think more than the answer I was surprised by the person who had answered me this one. I always thought he was expert in HTML, JavaScript but I guess, one should never assume about others. Indeed one of the cool logging feature is Change Data Capture. Change Data Capture records INSERTs, UPDATEs, and DELETEs applied to SQL Server tables, and makes a record available of what changed, where, and when, in simple relational ‘change tables’ rather than in an esoteric chopped salad of XML. These change tables contain columns that reflect the column structure of the source table you have chosen to track, along with the metadata needed to understand the changes that have been made. Here are some references for Change Data Capture: Introduction to Change Data Capture (CDC) in SQL Server 2008 Tuning the Performance of Change Data Capture in SQL Server 2008 Download Script of Change Data Capture (CDC) CDC and TRUNCATE – Cannot truncate table because it is published for replication or enabled for Change Data Capture Dynamic Management View (DMV) I like this answer. If asked I would have not come up with DMV right away but in the spirit of the original question, I think DMV does log the data. DMV logs or stores or records the various data and activity on the SQL Server. Dynamic management views return server state information that can be used to monitor the health of a server instance, diagnose problems, and tune performance. One can get plethero of information from DMVs – High Availability Status, Query Executions Details, SQL Server Resources Status etc. Here are some references for Dynamic Management View (DMV): SQL SERVER – Denali – DMV Enhancement – sys.dm_exec_query_stats – New Columns DMV – sys.dm_os_windows_info – Information about Operating System DMV – sys.dm_os_wait_stats Explanation – Wait Type – Day 3 of 28 DMV sys.dm_exec_describe_first_result_set_for_object – Describes the First Result Metadata for the Module Transaction Log Impact Detection Using DMV – dm_tran_database_transactions Log Files I almost flipped with this final answer from my friend. This should be probably the first answer. Yes, indeed log file logs the SQL Server activities. One can write infinite things about log file. SQL Server uses log file with the extension .ldf to manage transactions and maintain database integrity. Log file ensures that valid data is written out to database and system is in a consistent state. Log files are extremely useful in case of the database failures as with the help of full backup file database can be brought in the desired state (point in time recovery is also possible). SQL Server database has three recovery models – 1) Simple, 2) Full and 3) Bulk Logged. Each of the model uses the .ldf file for performing various activities. It is very important to take the backup of the log files (along with full backup) as one never knows when backup of the log file come into the action and save the day! How to Stop Growing Log File Too Big Reduce the Virtual Log Files (VLFs) from LDF file Log File Growing for Model Database – model Database Log File Grew Too Big master Database Log File Grew Too Big SHRINKFILE and TRUNCATE Log File in SQL Server 2008 Can I just say I loved this month’s T-SQL Tuesday Question. It really provoked very interesting conversation around me. Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: Pinal Dave, PostADay, SQL, SQL Authority, SQL Optimization, SQL Query, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Does armcc optimizes non-volatile variables with -O0 ?

    - by Dor
    int* Register = 0x00FF0000; // Address of micro-seconds timer while(*Register != 0); Should I declare *Register as volatile while using armcc compiler and -O0 optimization ? In other words: Does -O0 optimization requires qualifying that sort of variables as volatile ? (which is probably required in -O2 optimization)

    Read the article

  • Timer a usage in msp430 in high compiler optimization mode

    - by Vishal
    Hi, I have used timer A in MSP430 with high compiler optimization, but found that my timer code is failing when high compiler optimization used. When none optimization is used code works fine. This code is used to achieve 1 ms timer tick. timeOutCNT is increamented in interrupt. Following is the code, //Disable interrupt and clear CCR0 TIMER_A_TACTL = TIMER_A_TASSEL | // set the clock source as SMCLK TIMER_A_ID | // set the divider to 8 TACLR | // clear the timer MC_1; // continuous mode TIMER_A_TACTL &= ~TIMER_A_TAIE; // timer interrupt disabled TIMER_A_TACTL &= 0; // timer interrupt flag disabled CCTL0 = CCIE; // CCR0 interrupt enabled CCR0 = 500; TIMER_A_TACTL &= TIMER_A_TAIE; //enable timer interrupt TIMER_A_TACTL &= TIMER_A_TAIFG; //enable timer interrupt TACTL = TIMER_A_TASSEL + MC_1 + ID_3; // SMCLK, upmode timeOutCNT = 0; //timeOutCNT is increased in timer interrupt while(timeOutCNT <= 1); //delay of 1 milisecond TIMER_A_TACTL = TIMER_A_TASSEL | // set the clock source as SMCLK TIMER_A_ID | // set the divider to 8 TACLR | // clear the timer MC_1; // continuous mode TIMER_A_TACTL &= ~TIMER_A_TAIE; // timer interrupt disabled TIMER_A_TACTL &= 0x00; // timer interrupt flag disabled Can anybody help me here to resolve this issue? Is there any other way we can use timer A so it works fine in optimization modes? Or do I have used is wrongly to achieve 1 ms interrupt? Thanks in advanced. Vishal N

    Read the article

  • My timer code is failing when IAR is configured to do max optimization

    - by Vishal
    Hi, I have used timer A in MSP430 with high compiler optimization, but found that my timer code is failing when high compiler optimization used. When none optimization is used code works fine. This code is used to achieve 1 ms timer tick. timeOutCNT is increamented in interrupt. Following is the code [Code] //Disable interrupt and clear CCR0 TIMER_A_TACTL = TIMER_A_TASSEL | // set the clock source as SMCLK TIMER_A_ID | // set the divider to 8 TACLR | // clear the timer MC_1; // continuous mode TIMER_A_TACTL &= ~TIMER_A_TAIE; // timer interrupt disabled TIMER_A_TACTL &= 0; // timer interrupt flag disabled CCTL0 = CCIE; // CCR0 interrupt enabled CCR0 = 500; TIMER_A_TACTL &= TIMER_A_TAIE; //enable timer interrupt TIMER_A_TACTL &= TIMER_A_TAIFG; //enable timer interrupt TACTL = TIMER_A_TASSEL + MC_1 + ID_3; // SMCLK, upmode timeOutCNT = 0; //timeOutCNT is increased in timer interrupt while(timeOutCNT <= 1); //delay of 1 milisecond TIMER_A_TACTL = TIMER_A_TASSEL | // set the clock source as SMCLK TIMER_A_ID | // set the divider to 8 TACLR | // clear the timer MC_1; // continuous mode TIMER_A_TACTL &= ~TIMER_A_TAIE; // timer interrupt disabled TIMER_A_TACTL &= 0x00; // timer interrupt flag disabled [/code] Can anybody help me here to resolve this issue? Is there any other way we can use timer A so it works fine in optimization modes? Or do I have used is wrongly to achieve 1 ms interrupt? Thanks in advanced. Vishal N

    Read the article

  • Webserver optimization

    - by f-aminov
    Hi guys! I have a website hosted on a VPS (512Mb - minimum guranteed memory, 510Mhz proccessor, Debian 5.0 Lenny, Apache 2.2.9 with nginx 0.7.65 as a frontend to serve static content, MySQL 5.1.44, PHP 5.3.2 with APC caching). I'm a web developer, so I'm not very good at optimizing servers, but I've managed to install and setup all those neccessary components (LAMP, nginx, etc.). After that I decided to stress test my website (which uses Drupal 6.16 with caching and all possible optimization enabled) using a utility called "Webserver Stress Tool 7". And it seems to me that the results aren't any good - here is a graph (sorry, as a new user I'm not allowed to post images) As you can see the response time depending on amount of simultaneous users increases very quickly. With 10 simultaneous users the time is about 1000ms, with 100 simultaneous users it's about 15000ms (15s!). The question is do you think this is normal behavior for such a server or something is wrong with the settings and optimization? If you think something is wrong what particulary could be wrong? Any other suggestion how to speed this a little bit up?

    Read the article

  • No optimization causes wrong search result

    - by KailZhang
    I just took over our solr/lucene stuff from my ex-colleague. But there is a weird bug. If there is no optimization after dataimport, actually if there are multiple segment files, the search result then will be wrong. We are using a customized solr searchComponent. As far as I know about lucene, optimization is only optimization which could improve the speed of indexing and should not affect search result. I doubt this may be related to multithreading or unclosed searcher/reader or something. Anybody can help? Thank you.

    Read the article

  • How to cache dynamic javascript/jquery/ajax/json content with Akamai

    - by Starfs
    Trying to wrap my head around how things are cached on a CDN and it is new territory for me. In the document we received about sending in environment requests, it says "Dynamically-generated content will not benefit much from EdgeSuite". I feel like this is a simplified statement and there has to be a way to make it so you cache dynamically generated content if the tools are configured correctly. The site we are working with runs off a wordpress database, and uses javascript and ajax to build the pages, based on the json objects that php scripts have generated. The process - user's browser this URL, browser talks to edgesuite tools which will have cached certain pre-defined elements, and then requests from the host web server anything that is not cached, once edgesuite has compiled a combination of the two, it sends that information back to the browser. Can we not simply cache all json objects (and of course images, js, css) and therefore the web browser never has to hit the host server's database, at which point in essence, we have cached our dynamic content? Does anyone have any pointers on the most efficient configuration for this type of system -- Akamai/CDN -- to served javascript/ajax/json generated pages that ideally already hit pre-cached json data? Any and all feedback is welcome!

    Read the article

  • SQL SERVER – Introduction to Rollup Clause

    - by pinaldave
    In this article we will go over basic understanding of Rollup clause in SQL Server. ROLLUP clause is used to do aggregate operation on multiple levels in hierarchy. Let us understand how it works by using an example. Consider a table with the following structure and data: CREATE TABLE tblPopulation ( Country VARCHAR(100), [State] VARCHAR(100), City VARCHAR(100), [Population (in Millions)] INT ) GO INSERT INTO tblPopulation VALUES('India', 'Delhi','East Delhi',9 [...]

    Read the article

  • Page Speed and it&rsquo;s affect on your business

    - by ihaynes
    Originally posted on: http://geekswithblogs.net/ihaynes/archive/2014/05/29/page-speed-and-itrsquos-affect-on-your-business.aspxPage speed was an important issue 10 years ago, when we all had slow modems, but became less so with the advent of fast broadband connections that even seemed to make Ajax unnecessary.  Then along came the mobile internet and we’re back to a world where page speed and asset optimisation are critical again. If you doubt this an article on SitePoint discussing ’Page Speed and Business Metrics’ may change your mind. http://www.sitepoint.com/page-speed-business-metrics/ Here are some of the figures it quotes: Walmart – saw a 2% increase in conversions for every second of improvement in page load time. Put another way, accumulated growth of revenues went up 1% for every 100 milliseconds of load time improvement. Yahoo – for every 400 milliseconds of improvement, the site traffic increased by 9%. The bottom line is that if people have to wait more than a few seconds for a page to fully render, particularly on a mobile device, they’ll probably go elsewhere. Ignore this at your peril.   For two previous posts on the subject see: Page Weight: 10 easy fixes Xat.com Image Optimiser – Useful for RWD/Mobile

    Read the article

  • Python — Time complexity of built-in functions versus manually-built functions in finite fields

    - by stackuser
    Generally, I'm wondering about the advantages versus disadvantages of using the built-in arithmetic functions versus rolling your own in Python. Specifically, I'm taking in GF(2) finite field polynomials in string format, converting to base 2 values, performing arithmetic, then output back into polynomials as string format. So a small example of this is in multiplication: Rolling my own: def multiply(a,b): bitsa = reversed("{0:b}".format(a)) g = [(b<<i)*int(bit) for i,bit in enumerate(bitsa)] return reduce(lambda x,y: x+y,g) Versus the built-in: def multiply(a,b): # a,b are GF(2) polynomials in binary form .... return a*b #returns product of 2 polynomials in gf2 Currently, operations like multiplicative inverse (with for example 20 bit exponents) take a long time to run in my program as it's using all of Python's built-in mathematical operations like // floor division and % modulus, etc. as opposed to making my own division, remainder, etc. I'm wondering how much of a gain in efficiency and performance I can get by building these manually (as shown above). I realize the gains are dependent on how well the manual versions are built, that's not the question. I'd like to find out 'basically' how much advantage there is over the built-in's. So for instance, if multiplication (as in the example above) is well-suited for base 10 (decimal) arithmetic but has to jump through more hoops to change bases to binary and then even more hoops in operating (so it's lower efficiency), that's what I'm wondering. Like, I'm wondering if it's possible to bring the time down significantly by building them myself in ways that maybe some professionals here have already come across.

    Read the article

  • OpenGL's matrix stack vs Hand multiplying

    - by deft_code
    Which is more efficient using OpenGL's transformation stack or applying the transformations by hand. I've often heard that you should minimize the number of state transitions in your graphics pipeline. Pushing and popping translation matrices seem like a big change. However, I wonder if the graphics card might be able to more than make up for pipeline hiccup by using its parallel execution hardware to bulk multiply the vertices. My specific case. I have font rendered to a sprite sheet. The coordinates of each character or a string are calculated and added to a vertex buffer. Now I need to move that string. Would it be better to iterate through the vertex buffer and adjust each of the vertices by hand or temporarily push a new translation matrix?

    Read the article

  • Minecraft Style Chunk building problem

    - by David Torrey
    I'm having some problems with speed in my chunk engine. I timed it out, and in its current state it takes a total ~5 seconds per chunk to fill each face's list. I have a check to see if each face of a block is visible and if it is not visible, it skips it and moves on. I'm using a dictionary (unordered map) because it makes sense memorywise to just not have an entry if there is no block. I've tracked my problem down to testing if there is an entry, and accessing an entry if it does exist. If I remove the tests to see if there is an entry in the dictionary for an adjacent block, or if the block type itself is seethrough, it runs within about 2-4 milliseconds. so here's my question: Is there a faster way to check for an entry in a dictionary than .ContainsKey()? As an aside, I tried TryGetValue() and it doesn't really help with the speed that much. If I remove the ContainsKey() and keep the test where it does the IsSeeThrough for each block, it halves the time, but it's still about 2-3 seconds. It only drops to 2-4ms if I remove BOTH checks. Here is my code: using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Threading.Tasks; using System.Runtime.InteropServices; using OpenTK; using OpenTK.Graphics.OpenGL; using System.Drawing; namespace Anabelle_Lee { public enum BlockEnum { air = 0, dirt = 1, } [StructLayout(LayoutKind.Sequential,Pack=1)] public struct Coordinates<T1> { public T1 x; public T1 y; public T1 z; public override string ToString() { return "(" + x + "," + y + "," + z + ") : " + typeof(T1); } } public struct Sides<T1> { public T1 left; public T1 right; public T1 top; public T1 bottom; public T1 front; public T1 back; } public class Block { public int blockType; public bool SeeThrough() { switch (blockType) { case 0: return true; } return false ; } public override string ToString() { return ((BlockEnum)(blockType)).ToString(); } } class Chunk { private Dictionary<Coordinates<byte>, Block> mChunkData; //stores the block data private Sides<List<Coordinates<byte>>> mVBOVertexBuffer; private Sides<int> mVBOHandle; //private bool mIsChanged; private const byte mCHUNKSIZE = 16; public Chunk() { } public void InitializeChunk() { //create VBO references #if DEBUG Console.WriteLine ("Initializing Chunk"); #endif mChunkData = new Dictionary<Coordinates<byte> , Block>(); //mIsChanged = true; GL.GenBuffers(1, out mVBOHandle.left); GL.GenBuffers(1, out mVBOHandle.right); GL.GenBuffers(1, out mVBOHandle.top); GL.GenBuffers(1, out mVBOHandle.bottom); GL.GenBuffers(1, out mVBOHandle.front); GL.GenBuffers(1, out mVBOHandle.back); //make new list of vertexes for each face mVBOVertexBuffer.top = new List<Coordinates<byte>>(); mVBOVertexBuffer.bottom = new List<Coordinates<byte>>(); mVBOVertexBuffer.left = new List<Coordinates<byte>>(); mVBOVertexBuffer.right = new List<Coordinates<byte>>(); mVBOVertexBuffer.front = new List<Coordinates<byte>>(); mVBOVertexBuffer.back = new List<Coordinates<byte>>(); #if DEBUG Console.WriteLine("Chunk Initialized"); #endif } public void GenerateChunk() { #if DEBUG Console.WriteLine("Generating Chunk"); #endif for (byte i = 0; i < mCHUNKSIZE; i++) { for (byte j = 0; j < mCHUNKSIZE; j++) { for (byte k = 0; k < mCHUNKSIZE; k++) { Random blockLoc = new Random(); Coordinates<byte> randChunk = new Coordinates<byte> { x = i, y = j, z = k }; mChunkData.Add(randChunk, new Block()); mChunkData[randChunk].blockType = blockLoc.Next(0, 1); } } } #if DEBUG Console.WriteLine("Chunk Generated"); #endif } public void DeleteChunk() { //delete VBO references #if DEBUG Console.WriteLine("Deleting Chunk"); #endif GL.DeleteBuffers(1, ref mVBOHandle.left); GL.DeleteBuffers(1, ref mVBOHandle.right); GL.DeleteBuffers(1, ref mVBOHandle.top); GL.DeleteBuffers(1, ref mVBOHandle.bottom); GL.DeleteBuffers(1, ref mVBOHandle.front); GL.DeleteBuffers(1, ref mVBOHandle.back); //clear all vertex buffers ClearPolyLists(); #if DEBUG Console.WriteLine("Chunk Deleted"); #endif } public void UpdateChunk() { #if DEBUG Console.WriteLine("Updating Chunk"); #endif ClearPolyLists(); //prepare buffers //for every entry in mChunkData map foreach(KeyValuePair<Coordinates<byte>,Block> feBlockData in mChunkData) { Coordinates<byte> checkBlock = new Coordinates<byte> { x = feBlockData.Key.x, y = feBlockData.Key.y, z = feBlockData.Key.z }; //check for polygonson the left side of the cube if (checkBlock.x > 0) { //check to see if there is a key for current x - 1. if not, add the vector if (!IsVisible(checkBlock.x - 1, checkBlock.y, checkBlock.z)) { //add polygon AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.left); } } else { //polygon is far left and should be added AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.left); } //check for polygons on the right side of the cube if (checkBlock.x < mCHUNKSIZE - 1) { if (!IsVisible(checkBlock.x + 1, checkBlock.y, checkBlock.z)) { //add poly AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.right); } } else { //poly for right add AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.right); } if (checkBlock.y > 0) { //check to see if there is a key for current x - 1. if not, add the vector if (!IsVisible(checkBlock.x, checkBlock.y - 1, checkBlock.z)) { //add polygon AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.bottom); } } else { //polygon is far left and should be added AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.bottom); } //check for polygons on the right side of the cube if (checkBlock.y < mCHUNKSIZE - 1) { if (!IsVisible(checkBlock.x, checkBlock.y + 1, checkBlock.z)) { //add poly AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.top); } } else { //poly for right add AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.top); } if (checkBlock.z > 0) { //check to see if there is a key for current x - 1. if not, add the vector if (!IsVisible(checkBlock.x, checkBlock.y, checkBlock.z - 1)) { //add polygon AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.back); } } else { //polygon is far left and should be added AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.back); } //check for polygons on the right side of the cube if (checkBlock.z < mCHUNKSIZE - 1) { if (!IsVisible(checkBlock.x, checkBlock.y, checkBlock.z + 1)) { //add poly AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.front); } } else { //poly for right add AddPoly(checkBlock.x, checkBlock.y, checkBlock.z, mVBOHandle.front); } } BuildBuffers(); #if DEBUG Console.WriteLine("Chunk Updated"); #endif } public void RenderChunk() { } public void LoadChunk() { #if DEBUG Console.WriteLine("Loading Chunk"); #endif #if DEBUG Console.WriteLine("Chunk Deleted"); #endif } public void SaveChunk() { #if DEBUG Console.WriteLine("Saving Chunk"); #endif #if DEBUG Console.WriteLine("Chunk Saved"); #endif } private bool IsVisible(int pX,int pY,int pZ) { Block testBlock; Coordinates<byte> checkBlock = new Coordinates<byte> { x = Convert.ToByte(pX), y = Convert.ToByte(pY), z = Convert.ToByte(pZ) }; if (mChunkData.TryGetValue(checkBlock,out testBlock )) //if data exists { if (testBlock.SeeThrough() == true) //if existing data is not seethrough { return true; } } return true; } private void AddPoly(byte pX, byte pY, byte pZ, int BufferSide) { //create temp array GL.BindBuffer(BufferTarget.ArrayBuffer, BufferSide); if (BufferSide == mVBOHandle.front) { //front face mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.front.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ + 1) }); } else if (BufferSide == mVBOHandle.right) { //back face mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.back.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ) }); } else if (BufferSide == mVBOHandle.top) { //left face mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.left.Add(new Coordinates<byte> { x = (byte)(pX), y = (byte)(pY + 1), z = (byte)(pZ) }); } else if (BufferSide == mVBOHandle.bottom) { //right face mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ + 1) }); mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY) , z = (byte)(pZ) }); mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.right.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); } else if (BufferSide == mVBOHandle.front) { //top face mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ + 1) }); mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY + 1), z = (byte)(pZ) }); mVBOVertexBuffer.top.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY + 1), z = (byte)(pZ) }); } else if (BufferSide == mVBOHandle.back) { //bottom face mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY), z = (byte)(pZ + 1) }); mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY), z = (byte)(pZ) }); mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY), z = (byte)(pZ) }); mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY), z = (byte)(pZ) }); mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX + 1), y = (byte)(pY), z = (byte)(pZ + 1) }); mVBOVertexBuffer.bottom.Add(new Coordinates<byte> { x = (byte)(pX) , y = (byte)(pY), z = (byte)(pZ + 1) }); } } private void BuildBuffers() { #if DEBUG Console.WriteLine("Building Chunk Buffers"); #endif GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.front); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.front.Count), mVBOVertexBuffer.front.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.back); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.back.Count), mVBOVertexBuffer.back.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.left); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.left.Count), mVBOVertexBuffer.left.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.right); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.right.Count), mVBOVertexBuffer.right.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.top); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.top.Count), mVBOVertexBuffer.top.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer, mVBOHandle.bottom); GL.BufferData(BufferTarget.ArrayBuffer, (IntPtr)(Marshal.SizeOf(new Coordinates<byte>()) * mVBOVertexBuffer.bottom.Count), mVBOVertexBuffer.bottom.ToArray(), BufferUsageHint.StaticDraw); GL.BindBuffer(BufferTarget.ArrayBuffer,0); #if DEBUG Console.WriteLine("Chunk Buffers Built"); #endif } private void ClearPolyLists() { #if DEBUG Console.WriteLine("Clearing Polygon Lists"); #endif mVBOVertexBuffer.top.Clear(); mVBOVertexBuffer.bottom.Clear(); mVBOVertexBuffer.left.Clear(); mVBOVertexBuffer.right.Clear(); mVBOVertexBuffer.front.Clear(); mVBOVertexBuffer.back.Clear(); #if DEBUG Console.WriteLine("Polygon Lists Cleared"); #endif } }//END CLASS }//END NAMESPACE

    Read the article

  • How often is software speed evident in the eyes of customers?

    - by rwong
    In theory, customers should be able to feel the software performance improvements from first-hand experience. In practice, sometimes the improvements are not noticible enough, such that in order to monetize from the improvements, it is necessary to use quotable performance figures in marketing in order to attract customers. We already know the difference between perceived performance (GUI latency, etc) and server-side performance (machines, networks, infrastructure, etc). How often is it that programmers need to go the extra length to "write up" performance analyses for which the audience is not fellow programmers, but managers and customers?

    Read the article

  • CPU Architecture and floating-point math

    - by Jo-Herman Haugholt
    I'm trying to wrap my head around some details about how floating point math is performed on the CPU, trying to better understand what data types to use etc. I think I have a fairly good understanding of how integer math is performed. If I've understood correctly, and disregarding SIMD, a 32-bit CPU will generally perform integer math at at least 32-bit precision etc. Is it correct that floating-point math is dependent on the presence of a FPU? And that the FPU on the x86 is 80-bit, so floating point math is performed at this precision unless using SIMD? What about ARM?

    Read the article

  • "Optimal" game loop for 2D side-scroller

    - by MrDatabase
    Is it possible to describe an "optimal" (in terms of performance) layout for a 2D side-scroller's game loop? In this context the "game loop" takes user input, updates the states of game objects and draws the game objects. For example having a GameObject base class with a deep inheritance hierarchy could be good for maintenance... you can do something like the following: foreach(GameObject g in gameObjects) g.update(); However I think this approach can create performance issues. On the other hand all game objects' data and functions could be global. Which would be a maintenance headache but might be closer to an optimally performing game loop. Any thoughts? I'm interested in practical applications of near optimal game loop structure... even if I get a maintenance headache in exchange for great performance.

    Read the article

< Previous Page | 31 32 33 34 35 36 37 38 39 40 41 42  | Next Page >