Pre-fetch question

Hi,
Q1>Whenever we execute some querries, there is a roundtrip time involved to get the desired data from server which is taking much of the execution time.
Therefore we had to investigate on the problem and found out that ORACLE OCI allows prefetch at the client side and it is possible to build a client side cache by setting the "prefetch" attribute for the allocated transaction statment.
I wanted to confirm as to whether implementing the same, can we have significant performance gains..
Q2> What is the difference between fetch and prefetch? If fetch retrieves 5000 rows the will prefetch also retrieve 5000??
Any help will be appreciated
Thanks

Reading the doc pointers should help, but in a few words, with or without prefetch you are still calling OCIStmtExecute and/or OCIStmtFetch to get the rows, and you get the same number of rows in either case. Assuming you retrieve all rows one at a time, out of 5,000 total rows, OCI will by default do 2,500 round-trips to the server, because by default one extra row is transfered from the server to the client, and the next OCIStmtFetch call will get that row locally from the prefetch-cache.
Pre-fetching more rows further reduces round-trips at the expense of using more memory client-side. On win32 XP, I noticed that pre-fetching more than 32KB worth of row data (the actual row count depends on the row size) doesn't provide any improvement.
Using array defines is an alternative to pre-fetching which is a little faster. --DD                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               

Similar Messages

  • CryptoAPI pre-fetching crl details

    My question is that I can’t see crl pre-fetching working the way it is described as working, or I am misunderstanding the description.
    If I force a crl retrieval via certuil –verify (I have also tried CertGetCertificateChain), I can see new cache files generated at the location specified for the disk cache on Windows 7: c:\users\<username>\AppData\LocalLow\Microsoft\CryptnetUrlCache.
    I can see the required crl in there, which has a 4 hour lifetime and nextPublish about 1 hour before certificate expiry. So far, so good.
    The thing is, I never see an attempt to retrieve that crl again if I take no more action and just watch the cache. From reading the documentation my understanding is that pre-fetch should retrieve the crl before it is needed. Then, if the crl isn’t used after
    the initial pre-fetch cycle, it won’t be pre-fetched again. Also from what I could make out, I thought some sort of background retrieval mechanism periodically iterates over all items in the Cryptnet cache to see if pre-fetching needs to be performed. 
    BTW I’m not clear what actually would do this retrieval, the docs for CRYPTNET_URL_CACHE_PRE_FETCH_INFO structure docs hint that it might be “the Cryptnet URL Cache (CUC) service”, whatever that is (or was)- can't find any info on it or how often this might
    happen.
    I have to specifically attempt to make another CertGetCertificateChain/certutil –verify type call before I can see a crl retrieval attempted (http request and local cache update). With CertGetCertificateChain, I get a dwErrorStatus of 0, and dwInfoStatus of
    0x00000100/cert_trust_has_preferred_issuer, so I don’t see any problems with my verification. My crl has no AIA, only CDP listing a http location (no delta crls).
    As mentioned, first retrieval I see the cache files and http request, but no updates in the local store or attempts over the network after that. I was expecting  an automagic background HTTP request with conditional GET to occur sometime after nextPublish
    time - am I misunderstanding the concept of pre-fetch?
    Thanks!
    J

    Hi J. Crown,
    For this question, it is recommended to post in MSDN forum for more help, they focus on Windows Desktop development issue:
    http://social.msdn.microsoft.com/Forums/windowsdesktop/en-US/home?forum=windowsgeneraldevelopmentissues
    Thanks for understanding.
    Alex Zhao
    TechNet Community Support

  • Can we send pre-test score (Pre-test question slide) and post test score (Question slide) both in lms from captivate?

    can we send pre-test score (Pre-test question slide) and post test score (Question slide) both in lms from captivate?

    Pretest score normally is not sent to the LMS. However it is stored in a system variable, you could use JS to send it.

  • J2EE optimisation by pre-fetching pages

    Hi,
    I´m working on a J2EE application/site that generates pages
    dynamically. Although the pages vary significantly, the sequence of
    pages for a particular user is generally predictable. The client is
    worried about performance and wants the design to predict the next
    page, generate and cache it, so that it is ready for the user´s
    request.
    The pages are, at base, JSP pages (whose content depends largely on
    information in the user´s session that, over a single session, remains
    constant). Is there any way to do what the client wants (with a
    reasonable amount of effort)? I can't see how to work around the
    demand-driven nature of a J2EE server (in this case Weblogic's).
    Presumably, if there is a solution, it involves sending all requests
    to a Servlet, rather than the JSP page directly, and somehow the
    Servlet handles caching and prefetching. But that means the Servlet
    has to fake a SessionContext and redirect it to the appropriate JSP
    page (I think). Has anyone done this?
    Thanks,
    Andrew

    Also, take a look at Tangosol's Coherence caching product:
    http://www.tangosol.com/products-clustering.jsp
    Peace,
    Cameron Purdy
    Tangosol, Inc.
    Clustering Weblogic? You're either using Coherence, or you should be!
    Download a Tangosol Coherence eval today at http://www.tangosol.com/
    "Andrew Cooke" <[email protected]> wrote in message
    news:[email protected]...
    Actually, this reply is a bit too self-critical. It also makes sense for
    this data to be inside the JSP page for other, more sensible, reasons too.
    The data from page to page are very variable, requiring individual
    formatting.
    Andrew
    "Andrew Cooke" <[email protected]> escribió en el mensaje
    news:[email protected]...
    Hi,
    Thanks for the reply. Unfortunately it's difficult to know which data
    to
    cache until the page is generated, because the information that drivesthe
    page generation is inside the JSP page, not the database. In otherwords,
    when the page is generated, data within the page pulls the informationfrom
    the database.
    A cleaner design, obviously, would be to have everything in the
    database,
    but for practical reasons (limited resources - I am working for a small
    company in S. America, with a tight deadline and no other Java/JSP
    knowledge - that mean the closer this is to a "traditional" page-by-page
    website, the more likely it is to work) this is not possible. Or, atleast,
    it needs a very strong argument to change the approach (and I'm not sure
    this is it).
    Does that make sense? It means, as far as I can see, that only way to
    prepopulate the beans is by pre-generating the page...
    Thanks,
    Andrew
    <[email protected]> escribió en el mensaje
    news:[email protected]...
    Why don't you simply cache the data needed to render these
    pages? Unless your rendering mechanism is slow by itself, for a
    regular JSP which renders some pre-populated beans, pre-fetching
    shouldn't make much (if any) difference.
    Andrew Cooke <[email protected]> wrote:
    Hi,
    I´m working on a J2EE application/site that generates pages
    dynamically. Although the pages vary significantly, the sequence of
    pages for a particular user is generally predictable. The client is
    worried about performance and wants the design to predict the next
    page, generate and cache it, so that it is ready for the user´s
    request.
    The pages are, at base, JSP pages (whose content depends largely on
    information in the user´s session that, over a single session,
    remains
    constant). Is there any way to do what the client wants (with a
    reasonable amount of effort)? I can't see how to work around the
    demand-driven nature of a J2EE server (in this case Weblogic's).
    Presumably, if there is a solution, it involves sending all requests
    to a Servlet, rather than the JSP page directly, and somehow the
    Servlet handles caching and prefetching. But that means the Servlet
    has to fake a SessionContext and redirect it to the appropriate JSP
    page (I think). Has anyone done this?
    Thanks,
    Andrew--
    Dimitri

  • Pre-fetching of website stopped working, yet I have it selected as enabled

    Hi,
    Until today, every time I used firefox, once I'd start writing the address of the website, the address would come up entirely and I'd just have to select it. I think it's called pre-fetching of website?
    Today, it just stopped working but I have not changed any of my settings.
    Best,
    Sandrine

    Start Firefox in <u>[[Safe Mode]]</u> to check if one of the extensions is causing the problem (switch to the DEFAULT theme: Firefox (Tools) > Add-ons > Appearance/Themes).
    * Don't make any changes on the Safe mode start window.
    * https://support.mozilla.com/kb/Safe+Mode
    * https://support.mozilla.com/kb/Troubleshooting+extensions+and+themes

  • Pre Sales Questions CS4

    I want to ask a pre-sales question on Premier CS4. How do I do that. I cant
    see anything on the web, technical support only allows registered software.
    Also, does the student version of the above include watermarks on the image output? I am a teacher and have as student son at unversity. What is the cheapest option for Premier CS4 WITHOUT watermarks
    Thanks

    The
    education version does not watermark output.
    Cheers
    Eddie
    PremiereProPedia   (
    RSS feed)
    - Over 300 frequently answered questions
    - Over 250 free tutorials
    - Maintained by editors like
    you
    Forum FAQ

  • Pre-Sales Questions: RV220W & SRP 547W

    Hi,
    Couple of pre-sales questions if you don't mind:
    1. Does the RV220W Wireless-N Network Security Firewall allow RADIUS authentication of wireless clients on the WLAN?
    2.  Does the Cisco SRP 547W Wireless Router allow wireless clients to tunnel into the WLAN using VPN?
    Thanks very much.

    Yes it supports all the functionality. For more detail you can see the below link
    http://www.cisco.com/c/en/us/products/collateral/routers/rv220w-wireless-network-security-firewall/data_sheet_c78-630461.html

  • Pre-sales Questions

    What is the best way to contact Apple and get answers to pre-sale questions? Sometimes these forums work well, other times there is no response. When there is a response, is there a way one can determine it the responder is a representative from Apple?
    Is there an 800 number that one can call for pre-sales questions? Unfortunately the closest AppleStore is 60 miles from my home.

    1159/3605
    Hi jbmth,
    Responses you get here are from other users that volunteer to share their Apple knowledge.
    The quality is probably the best possible, since every time something goes wrong in one's advice, someone else would quickly jump in and offer corrections/additions.
    Determine representatives from Apple:
    They have a white apple in front of their names. You won't see an answer from an Apple employee more than once or twice a year here really.
    This is most a user-to-user forum.
    What kind of info do you need?
    Generally speaking, speculation about new products is quite very unwelcome by Apple standards.
    Here in the Discussions, it is simply forbidden by the Terms of Use.
    I hope this helps.
    Axl

  • Pre-test Questions used as Knowledge Builders

    I need questions at the beginning of a course that will not be included when the learner clicks on the Retake and Review Quiz buttons. Pre-test questions work but they don't allow the use of text captions for incorrect and correct feedback. They use smart shapes. Are there any other limitations of Pre-test Questions?

    There are a lot of limitations. You are expected to have pretest question in sequence at the start of the course. All navigation will be cancelled: by playbar and by TOC.
    But what you say about the feedback being only smart shapes as text containers is not true at all. They follow the same styling as normal question slides. If you turned on Use Shapes for SFH in the Preferences, Defaults, it will be shapes. You can always change this even after having inserted the Quiz/Preset slides by going to the corresponding master slide, right-click on the shape and choose 'Revert to caption'. Although to me it is a mystery why you should prefer captions, just my personal opinion because shapes are so much easier to format.
    If you use normal question slides, and do not add its score to the Total score, most system quizzing variables will not include those 'Knowledge' slides. But indeed, as you point out, Review and Retake will go to the first Question slide, which may be a knowledge slide. With a shared/advanced action it is pretty easy to circumvent this. I almost never use Pretest slides for its limitations. Whenever possible I use custom question slides instead of the default ones for Knowledge slides.

  • Pre-Order Question

    So I pre-ordered Titanfall for Xbox1, but I won't be in town when it comes out and one of my friends really wants to pick it up on the 11th at the midnight release. Is there any way to change the name on the pre-order, or would I have to cancel the one I have and create a new pre-order?

    Hi cgarvey4,
    I hate to be the bearer of bad news, but we don't currently offer a way to update the name associated with pre-orders. To have a friend pick up the game for you, you would need to cancel your existing pre-order and place a new one in their name. Also, remember that pre-orders must be picked up within 5 days of the title's release! If you aren't able to pick up your copy of Titanfall within that timeframe, then it will be returned to store inventory and made available for purchase to other customers.
    Let me know if you have any other questions!
    Aaron|Social Media Specialist | Best Buy® Corporate
     Private Message

  • Pre-order question for iPhone5/Verizon

    I have a quick question...
    My wife and son currently have the iPhone4 with AT&T and they are now out of contract.  I'm currently on Verizon.  I want to change their carrier and pre-order the iPhone5 for both of them from Verizon.  I want to add those lines to my existing Verizon account and transfer their numbers as well.  Would I be able to do this through this through the Apple Store's web-pre-order system? (or possibly Verizon's?) or will I have to speak with someone about doing this?
    Thank you so much in advance

    Here's what AT&T had to say about new customers when the 3GS was launched. I'd guess the same applies to iphone 4:
    http://www.att.com/wireless/iphone/en/iReady_checklist.pdf

  • IPhone pre-order question.

    I pre-ordered my iPhone 4 to be picked up at the Apple Store in my area. I chose the 32G like I did last year, but I've never used more than 12G at one time, so I think it's a waist to get the 32G, so I'm wanting to know when I pick up my iPhone 4 32G, can they change it to the 16G? I've tried calling the Apple Store and Apple's main line and apparently they don't staff enough people to take calls when it gets busy. Apple's main line won't even let you wait to speak to a representative. If anyone knows the answer to my question, please respond.

    Most likely they will not be able to answer your question now. I would think they will be willing to give you the 16 GB instead of the 32 GB you pre-ordered, but it's going to depend on the available stock in whatever store you're picking up at, at the time you get there. If they've sold out of the 16 GB model, then can't change your order.

  • A newbie pre-sales question

    I have hours of music on DAT tapes. Track numbers are already correct. Is it possible to record into Logic Express (or any program?) and have the tracks recognized? (My CD Recorder does that)
    Also, is it possible to record into Logic Ex from the direct digital input of the Mac without an audio interface box in between?
    Thank you

    Not sure about the 1st question but YES to the 2nd one - I record my electric guitar directly into LE using a standard guitar cable and a 1/4" -> 1/8" adapter - I use Amplitube and Guitar Rig plugins - sounds great. I also bought an M-Audio Audio Buddy (small 2 channel pre-amp) as it also has gain controls, etc and it seem to give me a fatter sound - it was only $59 at the Guitar Center - guitar plugs into it and the cable and adapter from it into my Imac...

  • Hit counter, hyperlink appearance, fetch questions

    Hi. New to Mac/Iweb. Hope you don't mind a compound question -
    I rebuilt my site recently using Iweb (www.joshbass.com), and for the most part, like it very much. However, there are a couple things I'm lost on. My settings tell me that on my hyperlinks, they're supposed to be white when "normal", and turn red on rollover. However, I have three browsers right now, IE, Firefox, and Safari, and only in Firefox do they behave correctly. In the other two browsers they simply stay white when rolled over.
    Secondly, I wanted to put a hit counter in (I believe there's no other way to measure traffic with this program, right?), but I'm NOT using .mac. So I found a random free counter, pasted the code into a web widget, and in Iweb it looked fine. Once published, not so much. Doesn't show up at all. Tried all the browsers, no difference.
    Third, I was wondering if there's a simple way to tie Fetch (for FTP) to Iweb, so that as I update the site in Iweb, I can publish only the changes, instead of FTPing the entire site's contents. There are indeed directions in the Iweb help files, but I'm afraid of doing something dumb and publishing garbage.
    Thanks.

    hi derek,
    1) the hit counter is a .mac only feature! you need to get an alternative counter such as statcounter:
    http://karreth.com/iweb/How%20To%20Add%20A%20StatCounter.html
    2)yes - check these instructions
    http://www.rowan-cottage.co.uk/Site/Autoplay_Music.html
    3)no sure what you mean with "automatically"
    4)my favorite is cyberduck. i also ahve some instructions for it:
    http://karreth.com/iweb/Host%20To%20Other%20Than%20.Mac.html
    max

  • CP5 pre-loader questions re: video

    Forgive the long post..
    I published a Cap 5 project in which the first 4 slides contain short videos.  The pre-loader is set to 70%.  When playing the course from a server or LMS the initial load ranges from 1 minute to 5 minutes depending on location and computer.   After the initial load reaches 70% the first video shows its own loading screen for a few additional seconds and begins to play.  Subsequent videos will also show a brief loader before they play.  The video and project sizes are as follows:
    - Course SWF is 12.5mb
    Slide one video is 827kb
    Slide two video is 1.2mb
    Slide three video is 1.9mb
    Slide four video is 950kb
    - All videos are encoded F4V
    - Project is published from CP5 on a Mac
    - Latest Flash player is being used for all user machines
    The Problems:
    - On high performance machines with fast internet connections, the content loads and plays as expected, however on slightly older machines, or slow connections the playback is jerky and unpredictable
    - Playback performance from a web server is generally better than from the SumTotal LMS regardless of the machine capabilities
    - On several occasions the audio for the next slide plays simultaneous with the video from the current slide.  Also bullets that are synched with the slide and video will fall out of synch.  After reloading or using the nav bar to back up, the slide generally plays as expected.
    My questions for the community:
    1 - Will the pre-loader load F4V videos before playback?  Can someone explain what the pre-loader will load what it will not?
    2 - If the preloader does not load video, is there a downside to setting the preloader to a minimum number (say 10%) to at least reduce the initial load cycle?
    3 -  How do I ensure that audio from the next slide will not play on the current slide or video synchronized with timeline elements do not fall out of synch?
    Thanks
    Jeff

    Hi there,
    Please find the answers for your questions.
    1 - Will the pre-loader load F4V videos before playback?  Can someone explain what the pre-loader will load what it will not?
    <ashwin> No, preloader only loads the main Movie. Once the slide with video is encountered the Video load is initiated and this loads the video. FLV video and Animations (if they are externalised) are not loaded while preloading the movie.
    </ashwin>
    2  - If the preloader does not load video, is there a downside to setting  the preloader to a minimum number (say 10%) to at least reduce the  initial load cycle?
    <ashwin> Yes you can do this but i am not sure if this reduces the project loading time. I will try to explain you the way preloader works.  A 10% preloader percentage indicates to load 10% of the SWF size, For instance if the SWF size is 12 MB then the project starts to play once 1.2 MB of the project is loaded. So if the project has rich content in the first few slides then loading may take time as it has it has to load the first slide fully to show the first slide. So it all depends on how the project is created and how the content is distributed in the project.
    </ashwin>
    3 -  How do I ensure that audio from  the next slide will not play on the current slide or video synchronized  with timeline elements do not fall out of synch?
    <ashwin>Do you have the audio from first slide to n slides continuously, if thats the case then in the published output a single audio file will be created with the published SWF and is stored in the SWF (In this case we say audio is sitched). So if a preloader is set to 10%, even though only 10% of the movie will be preloaded the entire audio should be loaded which might create sync issues as the movie will have started to play while the audio is being loaded.
    To overcome this you may choose to increase the slide time of the slides containing audio by around 0.1 second. This activity will split the audio in the published output and hence will have small audio fragments to load while the movie is loading.If around 20 slides have audio continuously then you can increase the slide time for around 6-7 slides and not all.
    </ashwin>
    Try them and let me know if this information helped.
    Thanks
    Ashwin Bharghav B
    Adobe Captivate team

Maybe you are looking for