Concept behind, strengths and weaknesses of NetWeaver?

Hi,
what is the concept behind NetWeaver?
What kind of strengths and weaknesses did NetWeaver have?
And which position did NetWeaver have in the context of ESOA?
Dokuments are also from interesting!
THX
Chris

Hi,
You're not asking for facts but for opinions.
I think you have to make your own opinions....
And everybody will have different ones.
From <b>my personnal point of view, </b>
The strength of Netweaver is the 15 years experience of robustness and reliability which began in 1992 with SAP R/3.
The weakness of Netweaver is the dual stack concept ( abap and java in the same product) which pretends that the Netweaver Application server exists, also there is an ABAP application server and a JAVA application server which are (in my opinion) badly glued together.
Regards,
Olivier

Similar Messages

  • What are strengths and weaknesses of the sap.m.Shell and sap.m.App as container for an application wrapper?

    My question today: What are strengths and weaknesses of the
    sap.m.Shell
    and
    sap.m.App
    as container for an application wrapper?
    I currently use non of them, but I see the advantages.
    SAP best practise is to use one of them or nest them as shell > app. What are strengths and weaknesses of the approaches?

    We prefer to say, the tender mercies of the Fetch block.
    You'll notice the "wfm info" cluster as an output of the Fetch
    VI.  This cluster has gain and offset members- use these to scale
    your binary data to floating point voltage representation.
    As you point out, the numbers won't be exactly what you expect given
    the vertical range and the device resolution.  There are a few
    reasons for this.  First, we don't use the absolute full scale of
    the ADC for the maximum input voltage.  There are a couple of
    codes of headroom at the top and bottom.  This allows your input
    signal to slightly exceed the specified vertical range without clipping
    (you'll get a warning from the driver).  Second, these gain and
    offset numbers take into account calibration information for your
    device.  Using these numbers will correct for small deviations
    from the ideal offset and gain.
    So if you use the gain and offset numbers from the driver, your
    measurements will be more accurate than if you assume 0 offset and
    perfect gain.  They will be most accurate if you make sure to
    perform a self-calibration once the device is in your system and at the
    temperature that your tests will be run.

  • Considering FileVault usage on an iMac with OSX 10.6.7. Wanting to know more about strengths and weaknesses of Filevault with Time Machine, Parallels and Carbon Copy Cloner. Any negatives to consider before flipping the switch?

    I'm considering FileVault usage on an iMac with OSX 10.6.7. Wanting to know more about strengths and weaknesses of Filevault with Time Machine, Parallels and Carbon Copy Cloner. Any negatives to consider before flipping the switch? Any information ?
    Would specifically like to know:
    Filevault impact on performance and application usability.
    TIme machine impacts, and whether TIme Machine volume will also be secure?
    Will a Cloned copy made with CCC be  bootable, or usable?
    Any other negatives to consider before turning this thing on?
    Thanks,

    Filevault impact on performance and application usability.
    FileVault encrypts the user's home directory only - it is basically an encrypted disk image that is automatically opened when you log in and closed when you log out. It does cause disk accesses to be a bit slower so it isn't a good idea to use an encrypted account with sound or video or large graphics files. I have no experience with it with Parallels Desktop but suspect that you'd notice a bit of a slowdown. Note: in the User's directory you'll see a Shared Folder. Since FV only encrypts the user's directory you can put data you don't need encrypted in this shared folder which won't be encrypted.
    TIme machine impacts, and whether TIme Machine volume will also be secure?
    There are issues with TM and FV but how extensive they are with Snow Leopard I can't say, once I realized that TM and FV hadn't substantially changed between 10.5 and 10.6 I stopped dealing with it. When FV and TM were first introduced my testing led me to recommend using CCC or SuperDuper! for backing up.
    Will a Cloned copy made with CCC be  bootable, or usable?
    Yes
    Any other negatives to consider before turning this thing on?
    Since FV is nothing more than a disk image - and disk images can fail - there are several problems with FV in my opinion. If an encrypted disk image fails it isn't possible to access any of that data. If you don't have a recent (undamaged) backup of the disk image and/or recent backup of the data within the disk image you are in a world of hurt. This isn't just a warning of possibility - this is a real danger that I've seen all too often at the shop where I have worked part time for a number of years.
    At this time I strongly recommend that people who have sensitive data keep that sensitive data segregated from their general data and encrypt only the sensitive data or use other methods of locking it down such as saving on an external drive that is kept in a safe. Call me paranoid but I don't even keep all my sensitive data in one encrypted disk image. I use multiple disk images and keep multiple sequential backups. That way I might someday find that my up-to-date investments disk image won't mount but (a) I can go back a week to a backup and (b) that won't impact my up-to-date business records because those are in a different disk image

  • Strengths and Weaknesses of BPEL PM Engine..

    We are planning to implement to Web Services and use BPEL PM Engine to Orchestrate the Business Processes. Before we go on a full scale Development, trying to understand the plus/minus of this tool.
    Will really appreciate feedback and pointers on the Strength and Weak areas of this Engine. Any major bottleneck faced during Production/implementation.. Any area performed well above your expectations..
    Thanks.

    Hi user -
    Can you please share the resutls of your analysis?

  • Strengths and weaknesses of Oracle developer/2000

    I wanted to know what the drawbacks and limitations of dev/2000(forms and reports) are. Would appreciate a discussion from various experienced developers on the limitations they faced and even the solutions they have provided/been provided.

    If you're planning on developing a system with good user interface design I wouldn't recommend using Forms 2000.
    We tried to do a design which mimics the Start menu of Windows on Forms 6 and my fellow programmer in charge of user interface had a hard time even with Oracle Technical Support.
    I'm just not sure if later releases of Forms 2000 can address such issues (moving windows, hiding of the title bar, following of the mouse, etc.).

  • Concept behind getNamedDispatcher in interface ServletContext ?

    I'm thinking about the concept behind getNamedDispatcher and getRequestDispatcher.
    In the deployment descriptor you align a display name to a servlet, which you then map on an URL-pattern or since Servlet API 2.5 also on multiple ...
    1. Involved in my problem are include vs. forwad server intern.
    2. Redirects to Servlets
    For problem one I could use getNamedDispatcher with the display-name, but for solving problem two I cannot; therefore I need to know the path within the context. So I decided to only use getRequestDispatcher and forget about the display-name.
    Actually since Servlet API 2.5 as there are multiple URL-patterns allowed, I think even if there is a solution to get an URL by display-name of the servlet, it would disambigous (because multiple are possible; how to determine which to choose?).
    Did I miss somehting? - How do you solve both matters and why did you decide in favour of your solution?

    Perhaps this page can help you:
    http://www.javaworld.com/javaworld/jw-01-2006/jw-0102-servlet-p3.html

  • Concept of util and helper classes

    whats the concept behind helper and util classes, how to divide code according to both perspective would any one state clear separation line b/w them.

    They're just names, dude

  • Srengths and weaknesses of Crystal Enterprise or Arcplan dynaSight

    Hi!
    I want to implement web reporting in my company. My favourite is Bex Web Application Designer. But my boss means I should look for others tools, too.
    Now, can someone give me information on the strengths and weaknesses of Crystal Enterprise and Arcplan dynaSight?
    Thanks a lot.

    Moved to BOE Admin forum

  • What is the concept behind using table PA0002 and structure p0002.

    Hi,
    What is the concept behind using table PA0002 and structure p0002.
    Many times, I have seen Looping at structure e.g. p0002, p0006 etc. and data is processed and also seen Looping at table PA0002, PA0006 etc. with further appropriate subtypes if any to retrieve a value and process the same.
    In which context tables like PA,HRP,IT etc. are used and structures p0001,p0002 etc. are used.
    Regards,
    Ameet

    HI,
    As mentioned  that that Structure Pnnnn is user as a internal tablw when a LDB is used.
    Ex.
    TABLES: PERNR.
    INFOTYPES: 0001.
      GET PERNR.
        PROVIDE * FROM P0001 BETWEEN PN-BEGDA AND PN-ENDDA.
        WRITE:  / P0001-PERNR,
                  P0001-STELL,
                  P0001-BEGDA,
                  P0001-ENDDA.
        ENDPROVIDE.
    Here it is important to declare the infotypes you want to read  and so the system will create internal tables  ex. P0001.
    Tables PA0001 are database table with following fields
    MANDT
    .INCLUDE  PAKEY
    .INCLUDE  PSHD1
    .INCLUDE   PS nnnn
    Thanks,
    Poonam.

  • Points and discussion about the concepts behind making a 2D game

    Hi all. I'm currently trying my hand at a 2D RPG style game. I have a few questions open for discussion, however, regarding the general points of implementation behind it.
    1. Tiles vs static background argument aside, am I right in the thinking the best way of implementing movement across a big background map is actually to move the background and other sprites in relation to the player, giving the illusion of player movement although he in fact never moves from the center of the screen? Is there a way of actually creating a large map, populating it with sprites, and then moving the player across it and panning the screen after him?
    2. Collision detection. If we go with the idea of moving everything in relation to the player rather than the player himself, woudln't this mean a large amount of calculations when collision detection occurs. Example, u want the player to stop moving when he hits a brick wall so you have to tell every entity sprite to stop moving in response to key presses (because remember, pressing up moves everything else up, not the player character).
    3. Isometric. Is the easiest way of implementing an isometric style game just to develop isometric graphics and sprites?
    4. Tiles. I've been doing a lot of reading into tiles but I'm not sure what's a better implementation: a tile system which consists of the entire background and things like houses split into tiles or a simple big background which can be scrolled around and is populated by house sprites which are all moved in relation to the player and their collision detections done individually. Tiles certainly seem like a nice innovation but I feel like the graphics suffer as a consequence and u end up with a 8bit NES looking game!
    I'm just getting to grips with the concepts behind a lot of game design and all of this are some interesting points which I thought would be insightful to discuss. Opinions please!
    Cheers.

    Everything I'm about to say is based off of a tile-based Maze game I made.
    1. Using a big background map is a bit lazy if you ask me, but graphically it would be easier to create a nice-looking game. You also have to think about memory. Let's say you manage to get an 800x600 background image at the size of 200kb with JPEG compression. That 200kb jpg image you use as a background is not 200kb when loaded into memory. I dont remember exactly how much bigger it gets in memory, so this next statement is probably wrong. But I heard that: 4bytes per pixel for the ARGB data, and there are 800x600 (480,000) pixels. That's 480000*4 (1920000) bytes if that statement is right. But it's probably wrong, so hopefully Abuse or someone will correct it :)
    2. Move EVERYTHING up? I think you're a little mixed up. When the background (and all objects with it) is scrolled, their actual X,Y data is not changed. It is simply drawn according to the scroll offset. The only thing that is being calclulated for movement is the character/scroll offset. Think of it as a camera. The entire world is not moving, just the location of where you're viewing it. Quick example:
    You have a Tree who's location is (10,10). If the scroll offset is (0,0), then that Tree will be drawn at (10,10). But let's say your character moves up 5 pixels. You say you want him centered, so the scroll offset also moves up 5 pixels, making the offset (0,-5). When it's time to draw again, that Tree can't be at the same place it was, because the character moved. If the character moved up, the Tree would appear as if it moved DOWN. So you'd draw it like:
    g.drawImage(tree.getImage(),tree.x-scrollOffsetX,tree.y-scrollOffsetY,null);With that math, the Tree is actually drawn at (10,15), which is lower than before the character moved. You don't actually loop through all of your objects and change each and every X and Y value, you just change the global scroll offset
    Speaking of collision detection, this is where TileMaps have a large advantage. In a TileMap, you don't have to check through each and every object in the world for collision. You can just grab the 4 surrounding tiles (north,west,east,south) from the Tile array. This means there are only 4 collision detections no matter where the character is in any situation on the map.
    If you used an image for background instead of tiles, you'd have to check each object in the world for collision because it's harder to tell how close it is to the character. One thing you could do is just grab all of the object currently visible on screen and check their collision, but, depending on where you are on the map, this could easily cause an inconsistent frame rate (you could be in a field of grass that has no collision or a forest of 50 trees each screen- LOTS of collision!).
    3. I dont got much to say about isometric :P
    4. an 8bit NES game, eh? That is the common misconception that you have to use images that are the same size of the tiles. Have you seen the game "The Legend of Zelda: Four Swords"? That's a tile-based game, but on several occasions you see trees that are 10 times bigger than the 20x20 tiles. They mixed it up, they used a TileMap for the basic ground, and put good looking large images on top. The large images (such as a large tree) uses 4 or 5 "base" tiles that are used for collision. That way the character will only collide with the bottom-half of the tree, giving the appearance of going "behind" the tree when the Y value is above the middle line of the tree. This is far away from looking like an 8bit NES game :)
    I hope that was more helpful than it was confusing, but I tend to babble, so.. anyway. Hope it helps :o

  • Concept of hide and clear

    Problem understanding hide and clear concept...
    for example, in the code below, it works fine with the write statement and have no problem with the output value when use the back button.
    however, when i use it to call a transaction (eg. mm03) and pass the value over, it seems to be a value behind (eg. when i click on row1, it gives a previously stored value. Press back button and when i click on row2, it gives me value of 1. Press back button and when i click on row5, it gives me value of 2. so the clear doesn't seem to "clear")
    what could be the problem? hope to understand this concept of hide and clear better (eg. when to clear?)...
    Cheers,
    Charles
    ABAP newbie
    DATA number(18) TYPE c.
    START-OF-SELECTION.
      CLEAR number.
      DO 9 TIMES.
        WRITE: / 'Row', (2) sy-index.
        number = sy-index.
        HIDE number.
      ENDDO.
    AT PF8.
      CHECK NOT number IS INITIAL.
      CALL TRANSACTION 'MM03'.
      SET PARAMETER ID 'MAT' FIELD number.
      CLEAR number.
    *  WRITE: / 'Cursor was in row', (2) NUMBER.

    hi,
    chk this.
    HIDE
    Basic form
    HIDE f.
    In an ABAP Objects context, a more severe syntax check is performed that in other ABAP areas.
    See Constants not allowed in HIDE area.
    Effect
    Retains the contents of f related to the current output line.
    When the user selects the line from the list f is automatically filled with the retained value.
    The selection can occur in:
    AT LINE-SELECTION
    AT PFx
    AT USER-COMMAND
    READ LINE
    The contents of the field do not have to have been displayed using WRITE in order for you to retain them.
    The HIDE statement does not support deep structures (structures that contain internal tables).
    Useful system fields for interactive reporting are listed in the System Fields for Lists documentation.
    Note
    Lines or components of lines of an internal table that you address using a field symbol (see ASSIGNING addition to the READ and LOOP statements), cannot be retained using HIDE. You can store them using a global variable instead.
    Note
    Runtime errors:
    HIDE_FIELD_TOO_LARGE: The field is too long for HIDE.
    HIDE_ON_EMPTY_PAGE: HIDE not possible on an empty page.
    HIDE_NO_LOCAL: HIDE not possible for a local field.
    HIDE_ILLEGAL_ITAB_SYMBOL: HIDE not possible for a table line or component of a table line.
    and also a sample program.
    report zxy_0003.
    data: begin of itab occurs 0,
    field type c,
    end of itab.
    itab-field = 'A'. append itab.
    itab-field = 'B'. append itab.
    itab-field = 'C'. append itab.
    itab-field = 'D'. append itab.
    itab-field = 'E'. append itab.
    loop at itab.
    format hotspot on.
    write:/ itab-field.
    hide itab-field.
    format hotspot off.
    endloop.
    at line-selection.
    write:/ 'You clicked', itab-field.
    It kind of "remembers" what is written, so that when you click on it, it knows what the value is.
    Regards,
    anver

  • What is the concept behind MASTER_IDOC_DISTRIBUTE

    What is the concept behind MASTER_IDOC_DISTRIBUTE ?

    Hi Manjunath,
    Please check the Documentation of the Function Module in SE37
    This function module is the interface from the application to the ALE layer on the outbound side. The application can pass an IDoc, the so-called master IDoc, as an internal table using the parameters MASTER_IDOC_CONTROL and MASTER_IDOC_DATA.
    This IDoc is then converted into one or communcation IDocs and stored in the ALE layer. IDocs for which no errors occurred are passed to dispatch control.
    In the table parameter  COMMUNICATION_IDOC_CONTROL the header records for the communication IDocs created are retturned. You can tell whether processing was successful from the field STATUS.
    A COMMIT WORK must be dispatched in the calling program, otherwise the IDocs may not be dispatched.
    Best regards,
    raam

  • Concept behind User Administration

    Good Morning all,
    Please explain the concept behind user administration.  My question is,
    1). How to assign the user authorization?  Do we need to maintain the user details under HR module?
    2). I've created a user and assigned full authorization, but the user is not able to access the entire menu bar, i.e. the menu bar does not contain the marketing documents and some of the reports.  If I add another user, the recently added user does not even have the access as like the previously created user.
    Can anyone please look into this and provide a solution.
    Thanks,
    Balaji

    Hello Balaji,
    1). How to assign the user authorization? Do we need to maintain the user details under HR module?
    The HR module does not have anything to do with the User Authorization.  The Human Resources module is just a place to store employee information and other pertinent details on their employment.
    2). I've created a user and assigned full authorization, but the user is not able to access the entire menu bar, i.e. the menu bar does not contain the marketing documents and some of the reports. If I add another user, the recently added user does not even have the access as like the previously created user.
    The authorization manages what a user can access and what he/she cannot.   If you are not able to see some of the Menu items it could be possible that they are diabled in the Form Settings.  After login as the user Click on Form Settings (Tools > Form Settings) and make sure that the menu items the user needs are checked visible.
    Make sure the the Visible check box is checked.
    Suda

  • Concept behind OCA

    Hi,
    I really don’t understand the concept behind OCA, if I was to follow the track line
    Introduction to SQL and Fundamentals of Database Administration (Part I) which includes concepts of instance, etc and also creating databases, tablespaces, etc. What sort of job can an OCA DBA apply to? Since Fundamentals of Database Administration (II) contains the backup and recover, networking concepts and issues while Performance tuning contains the performance related concepts, issues etc. As a DBA backup and recovery as well performance tuning are among the DBA tasks.
    Please clarify.
    Thank you,
    Adith

    OCA is only a track to satisfiying your OCP. But still you can get a junior DBA's position if you are clear in your concepts.

  • Pending Licenses : What is the concept behind pending licenses ?

    hi,
    I am seeing pending licenses in device license in call manager 8.5.1
    a) How can i filter devices whose licenses are pending.
    b) What is the concept behind pending licenses.

    Thanx [email protected]
       By gettign total no. of pending DLU's = some units apart form total alloted DLU's , consumed units and free units.
       What i understand is
        total alloted DLUS's =  total consumed DLU  + total free DLU + Pending DLU's.
       thus one pending licenseis as good as consuming a  license ?
       M i thinking in correct direction,if yes how can i free these pending DLU's .

Maybe you are looking for