Search Results

Search found 11380 results on 456 pages for 'cpu speed'.

facebook twitter digg google delicious technorati stumbleupon myspace wordpress linkedin gmail igoogle windows live tumblr viadeo yahoo buzz yahoo mail yahoo bookmarks favorites email print

Page 103/456 | < Previous Page | 99 100 101 102 103 104 105 106 107 108 109 110  | Next Page >

  • How can I store large amount of data from a database to XML (speed problem, part three)?

    - by Andrija
    After getting some responses, the current situation is that I'm using this tip: http://www.ibm.com/developerworks/xml/library/x-tipbigdoc5.html (Listing 1. Turning ResultSets into XML), and XMLWriter for Java from http://www.megginson.com/downloads/ . Basically, it reads date from the database and writes them to a file as characters, using column names to create opening and closing tags. While doing so, I need to make two changes to the input stream, namely to the dates and numbers. // Iterate over the set while (rs.next()) { w.startElement("row"); for (int i = 0; i < count; i++) { Object ob = rs.getObject(i + 1); if (rs.wasNull()) { ob = null; } String colName = meta.getColumnLabel(i + 1); if (ob != null ) { if (ob instanceof Timestamp) { w.dataElement(colName, Util.formatDate((Timestamp)ob, dateFormat)); } else if (ob instanceof BigDecimal){ w.dataElement(colName, Util.transformToHTML(new Integer(((BigDecimal)ob).intValue()))); } else { w.dataElement(colName, ob.toString()); } } else { w.emptyElement(colName); } } w.endElement("row"); } The SQL that gets the results has the to_number command (e.g. to_number(sif.ID) ID ) and the to_date command (e.g. TO_DATE (sif.datum_do, 'DD.MM.RRRR') datum_do). The problems are that the returning date is a timestamp, meaning I don't get 14.02.2010 but rather 14.02.2010 00:00:000 so I have to format it to the dd.mm.yyyy format. The second problem are the numbers; for some reason, they are in database as varchar2 and can have leading zeroes that need to be stripped; I'm guessing I could do that in my SQL with the trim function so the Util.transformToHTML is unnecessary (for clarification, here's the method): public static String transformToHTML(Integer number) { String result = ""; try { result = number.toString(); } catch (Exception e) {} return result; } What I'd like to know is a) Can I get the date in the format I want and skip additional processing thus shortening the processing time? b) Is there a better way to do this? We're talking about XML files that are in the 50 MB - 250 MB filesize category.

    Read the article

  • Fast inter-process (inter-threaded) communications IPC on large multi-cpu system.

    - by IPC
    What would be the fastest portable bi-directional communication mechanism for inter-process communication where threads from one application need to communicate to multiple threads in another application on the same computer, and the communicating threads can be on different physical CPUs). I assume that it would involve a shared memory and a circular buffer and shared synchronization mechanisms. But shared mutexes are very expensive (and there are limited number of them too) to synchronize when threads are running on different physical CPUs.

    Read the article

  • Does allocation speed depend on the garbage collector being used?

    - by jkff
    My app is allocating a ton of objects (1mln per second; most objects are byte arrays of size ~80-100 and strings of the same size) and I think it might be the source of its poor performance. The app's working set is only tens of megabytes. Profiling the app shows that GC time is negligibly small. However, I suspect that perhaps the allocation procedure depends on which GC is being used, and some settings might make allocation faster or perhaps make a positive influence on cache hit rate, etc. Is that so? Or is allocation performance independent on GC settings under the assumption that garbage collection itself takes little time?

    Read the article

  • Does the Java Memory Model (JSR-133) imply that entering a monitor flushes the CPU data cache(s)?

    - by Durandal
    There is something that bugs me with the Java memory model (if i even understand everything correctly). If there are two threads A and B, there are no guarantees that B will ever see a value written by A, unless both A and B synchronize on the same monitor. For any system architecture that guarantees cache coherency between threads, there is no problem. But if the architecture does not support cache coherency in hardware, this essentially means that whenever a thread enters a monitor, all memory changes made before must be commited to main memory, and the cache must be invalidated. And it needs to be the entire data cache, not just a few lines, since the monitor has no information which variables in memory it guards. But that would surely impact performance of any application that needs to synchronize frequently (especially things like job queues with short running jobs). So can Java work reasonably well on architectures without hardware cache-coherency? If not, why doesn't the memory model make stronger guarantees about visibility? Wouldn't it be more efficient if the language would require information what is guarded by a monitor? As i see it the memory model gives us the worst of both worlds, the absolute need to synchronize, even if cache coherency is guaranteed in hardware, and on the other hand bad performance on incoherent architectures (full cache flushes). So shouldn't it be more strict (require information what is guarded by a monitor) or more lose and restrict potential platforms to cache-coherent architectures? As it is now, it doesn't make too much sense to me. Can somebody clear up why this specific memory model was choosen? EDIT: My use of strict and lose was a bad choice in retrospect. I used "strict" for the case where less guarantees are made and "lose" for the opposite. To avoid confusion, its probably better to speak in terms of stronger or weaker guarantees.

    Read the article

  • Is learning ed worth it to boost my speed in VIM?

    - by Ksiresh
    I've learned the basic/intermediate levels of VIM ( it's to vast to list ). I often find that I slip back to my old ways and start using the mouse, holding down keys to get somewhere, and doing other stupid things that could be spead up. Would it be worth spending time to learn ed to break the habits learned from years in Windoze? Does using ed cultivate the right type of thinking that will transfer to VIM???

    Read the article

  • How to speed up saving a UIImagePickerController image from the camera to the filesystem via UIImagePNGRepresentation()?

    - by kazuhito0000
    I'm making an applications that let users take a photo and show them both in thumbnail and photo viewer. I have NSManagedObject class called photo and photo has a method that takes UIImage and converts it to PNG using UIImagePNGRepresentation() and saves it to filesystem. After this operation, resize the image to thumbnail size and save it. The problem here is UIImagePNGRepresentation() and conversion of image size seems to be really slow and I don't know if this is a right way to do it. Tell me if anyone know the best way to accomplish what I want to do. Thank you in advance.

    Read the article

  • How do I test the speed of a mySQL query?

    - by Chris
    I have a select and query like below... $sql = "SELECT * FROM notifications WHERE to_id='".$userid."' AND (alert_read != '1' OR user_read != '1') ORDER BY alert_time DESC"; $result = mysql_query($sql); how do I test how long the query took to run?

    Read the article

  • How to copy bytes from buffer into the managed struct?

    - by Chupo_cro
    I have a problem with getting the code to work in a managed environment (VS2008 C++/CLI Win Forms App). The problem is I cannot declare the unmanaged struct (is that even possible?) inside the managed code, so I've declared a managed struct but now I have a problem how to copy bytes from buffer into that struct. Here is the pure C++ code that obviously works as expected: typedef struct GPS_point { float point_unknown_1; float latitude; float longitude; float altitude; // x10000 float time; int point_unknown_2; int speed; // x100 int manually_logged_point; // flag (1 --> point logged manually) } track_point; int offset = 0; int filesize = 256; // simulates filesize int point_num = 10; // simulates number of records int main () { char *buffer_dyn = new char[filesize]; // allocate RAM // here, the file would have been read into the buffer buffer_dyn[0xa8] = 0x1e; // simulates the speed data (1e 00 00 00) buffer_dyn[0xa9] = 0x00; buffer_dyn[0xaa] = 0x00; buffer_dyn[0xab] = 0x00; offset = 0x90; // if the data with this offset is transfered trom buffer // to struct, int speed is alligned with the buffer at the // offset of 0xa8 track_point *points = new track_point[point_num]; points[0].speed = 0xff; // (debug) it should change into 0x1e memcpy(&points[0],buffer_dyn+offset,32); cout << "offset: " << offset << "\r\n"; //cout << "speed: " << points[0].speed << "\r\n"; printf ("speed : 0x%x\r\n",points[0].speed); printf("byte at offset 0xa8: 0x%x\r\n",(unsigned char)buffer_dyn[0xa8]); // should be 0x1e delete[] buffer_dyn; // release RAM delete[] points; /* What I need is to rewrite the lines 29 and 31 to work in the managed code (VS2008 Win Forms C++/CLI) What should I have after: array<track_point^>^ points = gcnew array<track_point^>(point_num); so I can copy 32 bytes from buffer_dyn to the managed struct declared as typedef ref struct GPS_point { float point_unknown_1; float latitude; float longitude; float altitude; // x10000 float time; int point_unknown_2; int speed; // x100 int manually_logged_point; // flag (1 --> point logged manually) } track_point; */ return 0; } Here is the paste to codepad.org so it can be seen the code is OK. What I need is to rewrite these two lines: track_point *points = new track_point[point_num]; memcpy(&points[0],buffer_dyn+offset,32); to something that will work in a managed application. I wrote: array<track_point^>^ points = gcnew array<track_point^>(point_num); and now trying to reproduce the described copying of the data from buffer over the struct, but haven't any idea how it should be done. Alternatively, if there is a way to use an unmanaged struct in the same way shown in my code, then I would like to avoid working with managed struct.

    Read the article

  • Speed of interpolation algorithms, C# and C++ working together.

    - by Kaminari
    Hello. I need fast implementation of popular interpolation algorithms. I figured it out that C# in such simple algorithms will be much slower than C++ so i think of writing some native code and using it in my C# GUI. First of all i run some tests and few operations on 1024x1024x3 matrix took 32ms in C# and 4ms in C++ and that's what i basicly need. Interpolation however is not a good word because i need them only for downscaling. But the question is: Will it be faster than C# methods in Drawing2D Image outputImage = new Bitmap(destWidth, destHeight, PixelFormat.Format24bppRgb); Graphics grPhoto = Graphics.FromImage(outputImage); grPhoto.InterpolationMode = InterpolationMode.*; //all of them grPhoto.DrawImage(bmp, new Rectangle(0, 0, destWidth, destHeight), Rectangle(0, 0, sourceWidth, sourceHeight), GraphicsUnit.Pixel); grPhoto.Dispose(); Some of these method run in 20ms and some in 80. Is there a way to do it faster?

    Read the article

  • Slow loop in R, any suggestion to speed it up?

    - by cfceric
    I have a data frame "m" as shown below: I am trying to find each account's most active month (with most number of V1). for example for account "2", it will be "month 6", for account 3 it will be "month 1", .... I wrote the below loop, it works fine but just takes a long time even I only used 8000 rows, the whole data set has 250,000 rows, so the code below is not usable. Does any one can suggest a better way to achieve this? Thanks a lot.

    Read the article

  • What's up with LDoms: Part 1 - Introduction & Basic Concepts

    - by Stefan Hinker
    LDoms - the correct name is Oracle VM Server for SPARC - have been around for quite a while now.  But to my surprise, I get more and more requests to explain how they work or to give advise on how to make good use of them.  This made me think that writing up a few articles discussing the different features would be a good idea.  Now - I don't intend to rewrite the LDoms Admin Guide or to copy and reformat the (hopefully) well known "Beginners Guide to LDoms" by Tony Shoumack from 2007.  Those documents are very recommendable - especially the Beginners Guide, although based on LDoms 1.0, is still a good place to begin with.  However, LDoms have come a long way since then, and I hope to contribute to their adoption by discussing how they work and what features there are today.  In this and the following posts, I will use the term "LDoms" as a common abbreviation for Oracle VM Server for SPARC, just because it's a lot shorter and easier to type (and presumably, read). So, just to get everyone on the same baseline, lets briefly discuss the basic concepts of virtualization with LDoms.  LDoms make use of a hypervisor as a layer of abstraction between real, physical hardware and virtual hardware.  This virtual hardware is then used to create a number of guest systems which each behave very similar to a system running on bare metal:  Each has its own OBP, each will install its own copy of the Solaris OS and each will see a certain amount of CPU, memory, disk and network resources available to it.  Unlike some other type 1 hypervisors running on x86 hardware, the SPARC hypervisor is embedded in the system firmware and makes use both of supporting functions in the sun4v SPARC instruction set as well as the overall CPU architecture to fulfill its function. The CMT architecture of the supporting CPUs (T1 through T4) provide a large number of cores and threads to the OS.  For example, the current T4 CPU has eight cores, each running 8 threads, for a total of 64 threads per socket.  To the OS, this looks like 64 CPUs.  The SPARC hypervisor, when creating guest systems, simply assigns a certain number of these threads exclusively to one guest, thus avoiding the overhead of having to schedule OS threads to CPUs, as do typical x86 hypervisors.  The hypervisor only assigns CPUs and then steps aside.  It is not involved in the actual work being dispatched from the OS to the CPU, all it does is maintain isolation between different guests. Likewise, memory is assigned exclusively to individual guests.  Here,  the hypervisor provides generic mappings between the physical hardware addresses and the guest's views on memory.  Again, the hypervisor is not involved in the actual memory access, it only maintains isolation between guests. During the inital setup of a system with LDoms, you start with one special domain, called the Control Domain.  Initially, this domain owns all the hardware available in the system, including all CPUs, all RAM and all IO resources.  If you'd be running the system un-virtualized, this would be what you'd be working with.  To allow for guests, you first resize this initial domain (also called a primary domain in LDoms speak), assigning it a small amount of CPU and memory.  This frees up most of the available CPU and memory resources for guest domains.  IO is a little more complex, but very straightforward.  When LDoms 1.0 first came out, the only way to provide IO to guest systems was to create virtual disk and network services and attach guests to these services.  In the meantime, several different ways to connect guest domains to IO have been developed, the most recent one being SR-IOV support for network devices released in version 2.2 of Oracle VM Server for SPARC. I will cover these more advanced features in detail later.  For now, lets have a short look at the initial way IO was virtualized in LDoms: For virtualized IO, you create two services, one "Virtual Disk Service" or vds, and one "Virtual Switch" or vswitch.  You can, of course, also create more of these, but that's more advanced than I want to cover in this introduction.  These IO services now connect real, physical IO resources like a disk LUN or a networt port to the virtual devices that are assigned to guest domains.  For disk IO, the normal case would be to connect a physical LUN (or some other storage option that I'll discuss later) to one specific guest.  That guest would be assigned a virtual disk, which would appear to be just like a real LUN to the guest, while the IO is actually routed through the virtual disk service down to the physical device.  For network, the vswitch acts very much like a real, physical ethernet switch - you connect one physical port to it for outside connectivity and define one or more connections per guest, just like you would plug cables between a real switch and a real system. For completeness, there is another service that provides console access to guest domains which mimics the behavior of serial terminal servers. The connections between the virtual devices on the guest's side and the virtual IO services in the primary domain are created by the hypervisor.  It uses so called "Logical Domain Channels" or LDCs to create point-to-point connections between all of these devices and services.  These LDCs work very similar to high speed serial connections and are configured automatically whenever the Control Domain adds or removes virtual IO. To see all this in action, now lets look at a first example.  I will start with a newly installed machine and configure the control domain so that it's ready to create guest systems. In a first step, after we've installed the software, let's start the virtual console service and downsize the primary domain.  root@sun # ldm list NAME STATE FLAGS CONS VCPU MEMORY UTIL UPTIME primary active -n-c-- UART 512 261632M 0.3% 2d 13h 58m root@sun # ldm add-vconscon port-range=5000-5100 \ primary-console primary root@sun # svcadm enable vntsd root@sun # svcs vntsd STATE STIME FMRI online 9:53:21 svc:/ldoms/vntsd:default root@sun # ldm set-vcpu 16 primary root@sun # ldm set-mau 1 primary root@sun # ldm start-reconf primary root@sun # ldm set-memory 7680m primary root@sun # ldm add-config initial root@sun # shutdown -y -g0 -i6 So what have I done: I've defined a range of ports (5000-5100) for the virtual network terminal service and then started that service.  The vnts will later provide console connections to guest systems, very much like serial NTS's do in the physical world. Next, I assigned 16 vCPUs (on this platform, a T3-4, that's two cores) to the primary domain, freeing the rest up for future guest systems.  I also assigned one MAU to this domain.  A MAU is a crypto unit in the T3 CPU.  These need to be explicitly assigned to domains, just like CPU or memory.  (This is no longer the case with T4 systems, where crypto is always available everywhere.) Before I reassigned the memory, I started what's called a "delayed reconfiguration" session.  That avoids actually doing the change right away, which would take a considerable amount of time in this case.  Instead, I'll need to reboot once I'm all done.  I've assigned 7680MB of RAM to the primary.  That's 8GB less the 512MB which the hypervisor uses for it's own private purposes.  You can, depending on your needs, work with less.  I'll spend a dedicated article on sizing, discussing the pros and cons in detail. Finally, just before the reboot, I saved my work on the ILOM, to make this configuration available after a powercycle of the box.  (It'll always be available after a simple reboot, but the ILOM needs to know the configuration of the hypervisor after a power-cycle, before the primary domain is booted.) Now, lets create a first disk service and a first virtual switch which is connected to the physical network device igb2. We will later use these to connect virtual disks and virtual network ports of our guest systems to real world storage and network. root@sun # ldm add-vds primary-vds root@sun # ldm add-vswitch net-dev=igb2 switch-primary primary You are free to choose whatever names you like for the virtual disk service and the virtual switch.  I strongly recommend that you choose names that make sense to you and describe the function of each service in the context of your implementation.  For the vswitch, for example, you could choose names like "admin-vswitch" or "production-network" etc. This already concludes the configuration of the control domain.  We've freed up considerable amounts of CPU and RAM for guest systems and created the necessary infrastructure - console, vts and vswitch - so that guests systems can actually interact with the outside world.  The system is now ready to create guests, which I'll describe in the next section. For further reading, here are some recommendable links: The LDoms 2.2 Admin Guide The "Beginners Guide to LDoms" The LDoms Information Center on MOS LDoms on OTN

    Read the article

  • 10 Windows Tweaking Myths Debunked

    - by Chris Hoffman
    Windows is big, complicated, and misunderstood. You’ll still stumble across bad advice from time to time when browsing the web. These Windows tweaking, performance, and system maintenance tips are mostly just useless, but some are actively harmful. Luckily, most of these myths have been stomped out on mainstream sites and forums. However, if you start searching the web, you’ll still find websites that recommend you do these things. Erase Cache Files Regularly to Speed Things Up You can free up disk space by running an application like CCleaner, another temporary-file-cleaning utility, or even the Windows Disk Cleanup tool. In some cases, you may even see an old computer speed up when you erase a large amount of useless files. However, running CCleaner or similar utilities every day to erase your browser’s cache won’t actually speed things up. It will slow down your web browsing as your web browser is forced to redownload the files all over again, and reconstruct the cache you regularly delete. If you’ve installed CCleaner or a similar program and run it every day with the default settings, you’re actually slowing down your web browsing. Consider at least preventing the program from wiping out your web browser cache. Enable ReadyBoost to Speed Up Modern PCs Windows still prompts you to enable ReadyBoost when you insert a USB stick or memory card. On modern computers, this is completely pointless — ReadyBoost won’t actually speed up your computer if you have at least 1 GB of RAM. If you have a very old computer with a tiny amount of RAM — think 512 MB — ReadyBoost may help a bit. Otherwise, don’t bother. Open the Disk Defragmenter and Manually Defragment On Windows 98, users had to manually open the defragmentation tool and run it, ensuring no other applications were using the hard drive while it did its work. Modern versions of Windows are capable of defragmenting your file system while other programs are using it, and they automatically defragment your disks for you. If you’re still opening the Disk Defragmenter every week and clicking the Defragment button, you don’t need to do this — Windows is doing it for you unless you’ve told it not to run on a schedule. Modern computers with solid-state drives don’t have to be defragmented at all. Disable Your Pagefile to Increase Performance When Windows runs out of empty space in RAM, it swaps out data from memory to a pagefile on your hard disk. If a computer doesn’t have much memory and it’s running slow, it’s probably moving data to the pagefile or reading data from it. Some Windows geeks seem to think that the pagefile is bad for system performance and disable it completely. The argument seems to be that Windows can’t be trusted to manage a pagefile and won’t use it intelligently, so the pagefile needs to be removed. As long as you have enough RAM, it’s true that you can get by without a pagefile. However, if you do have enough RAM, Windows will only use the pagefile rarely anyway. Tests have found that disabling the pagefile offers no performance benefit. Enable CPU Cores in MSConfig Some websites claim that Windows may not be using all of your CPU cores or that you can speed up your boot time by increasing the amount of cores used during boot. They direct you to the MSConfig application, where you can indeed select an option that appears to increase the amount of cores used. In reality, Windows always uses the maximum amount of processor cores your CPU has. (Technically, only one core is used at the beginning of the boot process, but the additional cores are quickly activated.) Leave this option unchecked. It’s just a debugging option that allows you to set a maximum number of cores, so it would be useful if you wanted to force Windows to only use a single core on a multi-core system — but all it can do is restrict the amount of cores used. Clean Your Prefetch To Increase Startup Speed Windows watches the programs you run and creates .pf files in its Prefetch folder for them. The Prefetch feature works as a sort of cache — when you open an application, Windows checks the Prefetch folder, looks at the application’s .pf file (if it exists), and uses that as a guide to start preloading data that the application will use. This helps your applications start faster. Some Windows geeks have misunderstood this feature. They believe that Windows loads these files at boot, so your boot time will slow down due to Windows preloading the data specified in the .pf files. They also argue you’ll build up useless files as you uninstall programs and .pf files will be left over. In reality, Windows only loads the data in these .pf files when you launch the associated application and only stores .pf files for the 128 most recently launched programs. If you were to regularly clean out the Prefetch folder, not only would programs take longer to open because they won’t be preloaded, Windows will have to waste time recreating all the .pf files. You could also modify the PrefetchParameters setting to disable Prefetch, but there’s no reason to do that. Let Windows manage Prefetch on its own. Disable QoS To Increase Network Bandwidth Quality of Service (QoS) is a feature that allows your computer to prioritize its traffic. For example, a time-critical application like Skype could choose to use QoS and prioritize its traffic over a file-downloading program so your voice conversation would work smoothly, even while you were downloading files. Some people incorrectly believe that QoS always reserves a certain amount of bandwidth and this bandwidth is unused until you disable it. This is untrue. In reality, 100% of bandwidth is normally available to all applications unless a program chooses to use QoS. Even if a program does choose to use QoS, the reserved space will be available to other programs unless the program is actively using it. No bandwidth is ever set aside and left empty. Set DisablePagingExecutive to Make Windows Faster The DisablePagingExecutive registry setting is set to 0 by default, which allows drivers and system code to be paged to the disk. When set to 1, drivers and system code will be forced to stay resident in memory. Once again, some people believe that Windows isn’t smart enough to manage the pagefile on its own and believe that changing this option will force Windows to keep important files in memory rather than stupidly paging them out. If you have more than enough memory, changing this won’t really do anything. If you have little memory, changing this setting may force Windows to push programs you’re using to the page file rather than push unused system files there — this would slow things down. This is an option that may be helpful for debugging in some situations, not a setting to change for more performance. Process Idle Tasks to Free Memory Windows does things, such as creating scheduled system restore points, when you step away from your computer. It waits until your computer is “idle” so it won’t slow your computer and waste your time while you’re using it. Running the “Rundll32.exe advapi32.dll,ProcessIdleTasks” command forces Windows to perform all of these tasks while you’re using the computer. This is completely pointless and won’t help free memory or anything like that — all you’re doing is forcing Windows to slow your computer down while you’re using it. This command only exists so benchmarking programs can force idle tasks to run before performing benchmarks, ensuring idle tasks don’t start running and interfere with the benchmark. Delay or Disable Windows Services There’s no real reason to disable Windows services anymore. There was a time when Windows was particularly heavy and computers had little memory — think Windows Vista and those “Vista Capable” PCs Microsoft was sued over. Modern versions of Windows like Windows 7 and 8 are lighter than Windows Vista and computers have more than enough memory, so you won’t see any improvements from disabling system services included with Windows. Some people argue for not disabling services, however — they recommend setting services from “Automatic” to “Automatic (Delayed Start)”. By default, the Delayed Start option just starts services two minutes after the last “Automatic” service starts. Setting services to Delayed Start won’t really speed up your boot time, as the services will still need to start — in fact, it may lengthen the time it takes to get a usable desktop as services will still be loading two minutes after booting. Most services can load in parallel, and loading the services as early as possible will result in a better experience. The “Delayed Start” feature is primarily useful for system administrators who need to ensure a specific service starts later than another service. If you ever find a guide that recommends you set a little-known registry setting to improve performance, take a closer look — the change is probably useless. Want to actually speed up your PC? Try disabling useless startup programs that run on boot, increasing your boot time and consuming memory in the background. This is a much better tip than doing any of the above, especially considering most Windows PCs come packed to the brim with bloatware.     

    Read the article

  • Real tortoises keep it slow and steady. How about the backups?

    - by Maria Zakourdaev
      … Four tortoises were playing in the backyard when they decided they needed hibiscus flower snacks. They pooled their money and sent the smallest tortoise out to fetch the snacks. Two days passed and there was no sign of the tortoise. "You know, she is taking a lot of time", said one of the tortoises. A little voice from just out side the fence said, "If you are going to talk that way about me I won't go." Is it too much to request from the quite expensive 3rd party backup tool to be a way faster than the SQL server native backup? Or at least save a respectable amount of storage by producing a really smaller backup files?  By saying “really smaller”, I mean at least getting a file in half size. After Googling the internet in an attempt to understand what other “sql people” are using for database backups, I see that most people are using one of three tools which are the main players in SQL backup area:  LiteSpeed by Quest SQL Backup by Red Gate SQL Safe by Idera The feedbacks about those tools are truly emotional and happy. However, while reading the forums and blogs I have wondered, is it possible that many are accustomed to using the above tools since SQL 2000 and 2005.  This can easily be understood due to the fact that a 300GB database backup for instance, using regular a SQL 2005 backup statement would have run for about 3 hours and have produced ~150GB file (depending on the content, of course).  Then you take a 3rd party tool which performs the same backup in 30 minutes resulting in a 30GB file leaving you speechless, you run to management persuading them to buy it due to the fact that it is definitely worth the price. In addition to the increased speed and disk space savings you would also get backup file encryption and virtual restore -  features that are still missing from the SQL server. But in case you, as well as me, don’t need these additional features and only want a tool that performs a full backup MUCH faster AND produces a far smaller backup file (like the gain you observed back in SQL 2005 days) you will be quite disappointed. SQL Server backup compression feature has totally changed the market picture. Medium size database. Take a look at the table below, check out how my SQL server 2008 R2 compares to other tools when backing up a 300GB database. It appears that when talking about the backup speed, SQL 2008 R2 compresses and performs backup in similar overall times as all three other tools. 3rd party tools maximum compression level takes twice longer. Backup file gain is not that impressive, except the highest compression levels but the price that you pay is very high cpu load and much longer time. Only SQL Safe by Idera was quite fast with it’s maximum compression level but most of the run time have used 95% cpu on the server. Note that I have used two types of destination storage, SATA 11 disks and FC 53 disks and, obviously, on faster storage have got my backup ready in half time. Looking at the above results, should we spend money, bother with another layer of complexity and software middle-man for the medium sized databases? I’m definitely not going to do so.  Very large database As a next phase of this benchmark, I have moved to a 6 terabyte database which was actually my main backup target. Note, how multiple files usage enables the SQL Server backup operation to use parallel I/O and remarkably increases it’s speed, especially when the backup device is heavily striped. SQL Server supports a maximum of 64 backup devices for a single backup operation but the most speed is gained when using one file per CPU, in the case above 8 files for a 2 Quad CPU server. The impact of additional files is minimal.  However, SQLsafe doesn’t show any speed improvement between 4 files and 8 files. Of course, with such huge databases every half percent of the compression transforms into the noticeable numbers. Saving almost 470GB of space may turn the backup tool into quite valuable purchase. Still, the backup speed and high CPU are the variables that should be taken into the consideration. As for us, the backup speed is more critical than the storage and we cannot allow a production server to sustain 95% cpu for such a long time. Bottomline, 3rd party backup tool developers, we are waiting for some breakthrough release. There are a few unanswered questions, like the restore speed comparison between different tools and the impact of multiple backup files on restore operation. Stay tuned for the next benchmarks.    Benchmark server: SQL Server 2008 R2 sp1 2 Quad CPU Database location: NetApp FC 15K Aggregate 53 discs Backup statements: No matter how good that UI is, we need to run the backup tasks from inside of SQL Server Agent to make sure they are covered by our monitoring systems. I have used extended stored procedures (command line execution also is an option, I haven’t noticed any impact on the backup performance). SQL backup LiteSpeed SQL Backup SQL safe backup database <DBNAME> to disk= '\\<networkpath>\par1.bak' , disk= '\\<networkpath>\par2.bak', disk= '\\<networkpath>\par3.bak' with format, compression EXECUTE master.dbo.xp_backup_database @database = N'<DBName>', @backupname= N'<DBName> full backup', @desc = N'Test', @compressionlevel=8, @filename= N'\\<networkpath>\par1.bak', @filename= N'\\<networkpath>\par2.bak', @filename= N'\\<networkpath>\par3.bak', @init = 1 EXECUTE master.dbo.sqlbackup '-SQL "BACKUP DATABASE <DBNAME> TO DISK= ''\\<networkpath>\par1.sqb'', DISK= ''\\<networkpath>\par2.sqb'', DISK= ''\\<networkpath>\par3.sqb'' WITH DISKRETRYINTERVAL = 30, DISKRETRYCOUNT = 10, COMPRESSION = 4, INIT"' EXECUTE master.dbo.xp_ss_backup @database = 'UCMSDB', @filename = '\\<networkpath>\par1.bak', @backuptype = 'Full', @compressionlevel = 4, @backupfile = '\\<networkpath>\par2.bak', @backupfile = '\\<networkpath>\par3.bak' If you still insist on using 3rd party tools for the backups in your production environment with maximum compression level, you will definitely need to consider limiting cpu usage which will increase the backup operation time even more: RedGate : use THREADPRIORITY option ( values 0 – 6 ) LiteSpeed : use  @throttle ( percentage, like 70%) SQL safe :  the only thing I have found was @Threads option.   Yours, Maria

    Read the article

  • Bouncing off a circular Boundary with multiple balls?

    - by Anarkie
    I am making a game like this : Yellow Smiley has to escape from red smileys, when yellow smiley hits the boundary game is over, when red smileys hit the boundary they should bounce back with the same angle they came, like shown below: Every 10 seconds a new red smiley comes in the big circle, when red smiley hits yellow, game is over, speed and starting angle of red smileys should be random. I control the yellow smiley with arrow keys. The biggest problem I have reflecting the red smileys from the boundary with the angle they came. I don't know how I can give a starting angle to a red smiley and bouncing it with the angle it came. I would be glad for any tips! My js source code : var canvas = document.getElementById("mycanvas"); var ctx = canvas.getContext("2d"); // Object containing some global Smiley properties. var SmileyApp = { radius: 15, xspeed: 0, yspeed: 0, xpos:200, // x-position of smiley ypos: 200 // y-position of smiley }; var SmileyRed = { radius: 15, xspeed: 0, yspeed: 0, xpos:350, // x-position of smiley ypos: 65 // y-position of smiley }; var SmileyReds = new Array(); for (var i=0; i<5; i++){ SmileyReds[i] = { radius: 15, xspeed: 0, yspeed: 0, xpos:350, // x-position of smiley ypos: 67 // y-position of smiley }; SmileyReds[i].xspeed = Math.floor((Math.random()*50)+1); SmileyReds[i].yspeed = Math.floor((Math.random()*50)+1); } function drawBigCircle() { var centerX = canvas.width / 2; var centerY = canvas.height / 2; var radiusBig = 300; ctx.beginPath(); ctx.arc(centerX, centerY, radiusBig, 0, 2 * Math.PI, false); // context.fillStyle = 'green'; // context.fill(); ctx.lineWidth = 5; // context.strokeStyle = '#003300'; // green ctx.stroke(); } function lineDistance( positionx, positiony ) { var xs = 0; var ys = 0; xs = positionx - 350; xs = xs * xs; ys = positiony - 350; ys = ys * ys; return Math.sqrt( xs + ys ); } function drawSmiley(x,y,r) { // outer border ctx.lineWidth = 3; ctx.beginPath(); ctx.arc(x,y,r, 0, 2*Math.PI); //red ctx.fillStyle="rgba(255,0,0, 0.5)"; ctx.fillStyle="rgba(255,255,0, 0.5)"; ctx.fill(); ctx.stroke(); // mouth ctx.beginPath(); ctx.moveTo(x+0.7*r, y); ctx.arc(x,y,0.7*r, 0, Math.PI, false); // eyes var reye = r/10; var f = 0.4; ctx.moveTo(x+f*r, y-f*r); ctx.arc(x+f*r-reye, y-f*r, reye, 0, 2*Math.PI); ctx.moveTo(x-f*r, y-f*r); ctx.arc(x-f*r+reye, y-f*r, reye, -Math.PI, Math.PI); // nose ctx.moveTo(x,y); ctx.lineTo(x, y-r/2); ctx.lineWidth = 1; ctx.stroke(); } function drawSmileyRed(x,y,r) { // outer border ctx.lineWidth = 3; ctx.beginPath(); ctx.arc(x,y,r, 0, 2*Math.PI); //red ctx.fillStyle="rgba(255,0,0, 0.5)"; //yellow ctx.fillStyle="rgba(255,255,0, 0.5)"; ctx.fill(); ctx.stroke(); // mouth ctx.beginPath(); ctx.moveTo(x+0.4*r, y+10); ctx.arc(x,y+10,0.4*r, 0, Math.PI, true); // eyes var reye = r/10; var f = 0.4; ctx.moveTo(x+f*r, y-f*r); ctx.arc(x+f*r-reye, y-f*r, reye, 0, 2*Math.PI); ctx.moveTo(x-f*r, y-f*r); ctx.arc(x-f*r+reye, y-f*r, reye, -Math.PI, Math.PI); // nose ctx.moveTo(x,y); ctx.lineTo(x, y-r/2); ctx.lineWidth = 1; ctx.stroke(); } // --- Animation of smiley moving with constant speed and bounce back at edges of canvas --- var tprev = 0; // this is used to calculate the time step between two successive calls of run function run(t) { requestAnimationFrame(run); if (t === undefined) { t=0; } var h = t - tprev; // time step tprev = t; SmileyApp.xpos += SmileyApp.xspeed * h/1000; // update position according to constant speed SmileyApp.ypos += SmileyApp.yspeed * h/1000; // update position according to constant speed for (var i=0; i<SmileyReds.length; i++){ SmileyReds[i].xpos += SmileyReds[i].xspeed * h/1000; // update position according to constant speed SmileyReds[i].ypos += SmileyReds[i].yspeed * h/1000; // update position according to constant speed } // change speed direction if smiley hits canvas edges if (lineDistance(SmileyApp.xpos, SmileyApp.ypos) + SmileyApp.radius > 300) { alert("Game Over"); } // redraw smiley at new position ctx.clearRect(0,0,canvas.height, canvas.width); drawBigCircle(); drawSmiley(SmileyApp.xpos, SmileyApp.ypos, SmileyApp.radius); for (var i=0; i<SmileyReds.length; i++){ drawSmileyRed(SmileyReds[i].xpos, SmileyReds[i].ypos, SmileyReds[i].radius); } } // uncomment these two lines to get every going // SmileyApp.speed = 100; run(); // --- Control smiley motion with left/right arrow keys function arrowkeyCB(event) { event.preventDefault(); if (event.keyCode === 37) { // left arrow SmileyApp.xspeed = -100; SmileyApp.yspeed = 0; } else if (event.keyCode === 39) { // right arrow SmileyApp.xspeed = 100; SmileyApp.yspeed = 0; } else if (event.keyCode === 38) { // up arrow SmileyApp.yspeed = -100; SmileyApp.xspeed = 0; } else if (event.keyCode === 40) { // right arrow SmileyApp.yspeed = 100; SmileyApp.xspeed = 0; } } document.addEventListener('keydown', arrowkeyCB, true); JSFiddle : http://jsfiddle.net/gj4Q7/

    Read the article

facebook twitter digg google delicious technorati stumbleupon myspace wordpress linkedin gmail igoogle windows live tumblr viadeo yahoo buzz yahoo mail yahoo bookmarks favorites email print

< Previous Page | 99 100 101 102 103 104 105 106 107 108 109 110  | Next Page >