Implementation of a 3 tier model architecture

Hi,
I've just gotten a project to complete where users at 20 customer sites need to gain access to a database in our company and download/upload lots of tables/data from a web page (thru a firewall etc)
The way I initally intended to go was using an applet,RMI and JDBC in a 3 tier approach (mainly cos I've seen this approach used before)
Is this the best approach to take. What other alternatives could I use??
Could I use a more simple CGI type approach??
I am pretty new to java so appologies if this question sounds stupid
Willie

One alternative is HTML/Servlet/JDBC.

Similar Messages

  • 2 tier model or 3 tier model

    Which one is better for BC4J JClient application ? using 2 tier model or 3 tier model ?
    Thx,
    Ricky H.P.

    Ricky,
    it depends on your business case
    If you need to share business logic with other web applications or if the client PC is not as powerfuly as it should be, then using a three tier architecture is what I would recommend.
    If you are free to choose with dependecies then a two tier architecture is what I would prefer. However, if you need your application client to be able to run offline /stand alone) against a local database, then two tier is what you should do.
    My personal preference is to position JClient in a two tier architecture running on Java Web Start.
    Frank

  • Advice on Security Model Architecture..

    Hi all,
    Just looking for the advice of the experts :)
    I am working on the security model architecture for multi-tiered java application. The application architecture breaks down roughly as follows:
    Presentation Layer (JSP/Java)
    Business Layer (Java)
    Persistence Layer (JDBC/Oracle DB)
    Now, in the DB we will preserve information about various users, as well as the user's application permissions. My question pertains to authentication/authorization. Where is it most appropriate or efficient to verify a user's access to a functionality? Assume that the user and permission information is retrieved upon login and is made available to all levels.
    The options, as I see them, include the following:
    Presention layer - UI exposes only functionality applicable to the user.
    Business layer - Encode the logic in this facade for the backend.
    Persistence layer - Encode the logic in the data access objects.
    Any thoughts?

    Well, the layered approach is one way in which java applications are constructed.. the user interface is the top layer, which is composed of jsp files and other java files, and the objects that talk to the database are the bottom layer. Maybe an example would help..
    You're looking at a page on the Java Discussion Forums. It's a jsp page. You click on the 'Watches' link (upper right). The link points to a servlet, which calls a method in an object that is in what I call the "business" or middle layer/tier. An object in this layer has methods that correspond to any request that needs to be made of the db.
    This method in turn calls method/methods in the backend, or data layer, which queries the database and returns the watches for this particular user...
    So, if you have a request/response transaction (click on a link or button, processing, and new page is loaded), it would make a round trip through the layers:
    Presentation -> Business -> Data -> DB -> Data -> Business -> Presentation

  • Best way to implement service in 3-tier webarchitecture

    Hi,
    What would be the best way to implement a service in the following 3-tier TopLink architecture: no ejb and a webclient (jsp/servlets)?
    Currently I have a server session type and a service pojo using a singleton SessionFactory. The service pojo is used by the jsp/servlets.
    Should the service pojo be a singleton? Should a SessionManager be used?
    In the examples on OTN a clientsession is directly acquired in the jsp/servlets.
    Thanks,
    Ronald

    Ronald,
    There are numerous ways to do this. I would recommend using the SessionFactory. The SessionFactory makes use of the SessionManager, which holds onto the singleton Server session. You can then acquire client sessions using the oracle.toplink.sessions.Session interface from the SessionFactory. There is no need to hold the SessionFactory in a singleton yourself and you do not need to reference the ClientSession directly either.
    We have introduced this latest approach to simplify coding. It will work in both the web architecture with JSP/Servlets accessing the sessions as well as within an EJB tier (Session/Message Driven beans).
    Doug

  • 3-Tier model of linking DB

    Dear all,
    I have some questions about linking DB. As you know, I directly copy the data into the local DB. It is not a good choice! So I would like to implement 3-tier model to access DB so that it much more efficient choice as we can access DB anywhere.
    Thus, I have look into it. The method sounds like this:
    1. Get the data from Applet and copy them to Serializable Object or XML file and pass it to a JSP Server (Client).
    2. The JSP Server listen the request and obtain the Serializable Object or XML file so the data can be retrieved. Then call the Bean and pass these parameters to it (Middle-Tier).
    3. The Bean receive the request and access the DB (Server).
    Do I interpret the wrong idea? Actually I don't too understand too much about this model. Moreover, I don't know how to send the Serializable Object or XML file to JSP Server so the JSP script can do the task.
    Hope that you may give me some idea and suggestion for that! Thanks a lot!
    Best regards,
    Peter Ip

    Sounds like your on the right track. Keep in mind that the reason serializable objects are used is to keep client pages decoupled from the middle tier, along with providing a clean and consitant way of passing data to the middle tier.
    "I don't know how to send the Serializable Object or XML file to JSP Server so the JSP script can do the task."
    Here's how I do this. Client pages send form data to a servlet (via HTML action=""). The servlets job is to parse/approve the data, and create a ValueObject (VO) represeneting the data in a db table like format. The VO is object is Serializable, and moreover, it has no functionality, just variables with getter/setter methods. Its only purpose is to transport data between teirs. This concept is important, as it keeps your client(s) pages loosly coupled, and alows reuse.
    So, client sends in data, servlet parses data and creates a VO, then the servlet invokes the middle teir by sending in the VO. At that point, the middle tier does its thing from the VO. The middle tier does not have to worry about parsing or validating the data, and it's client independent, as the middle tier should be (e.g. no http request usage, session usage, etc.)
    For more information, look into the Data Access patterns and/or visit the J2EE paterns form. You'll find information about Data Access Object (DAOs) and ValueObjects (VOs
    Hope this helps without adding confusion. Let me know if I, or anyone, can elaborate on the concepts.
    bRi

  • OC4J 10.1.2 - No XPathFactory implementation found for the object model

    Hello,
    We are doing some maintenance on an old 10.1.2.3.0 container. The project has to be extended with a library built with xml-apis-1.3.03.jar. I know you will suggest a upgrade, but that is a no-go - sorry, since other projects also are running on this server. So I hope some of you remember or know how to avoid or replace the oracle xml parser.
    It seems like it gets the wrong XPathFactory
    XPathFactory#newInstance() failed to create an XPathFactory for the default object model: http://java.sun.com/jaxp/xpath/dom
    with the XPathFactoryConfigurationException: javax.xml.xpath.XPathFactoryConfigurationException: No XPathFctory implementation found for the object model:
    I have searched the documentation for 10.1.2 and everything relevant on the Internet for a solution, but i seems to non-trivial when dealing with oracle xml parser. We tried to put the jar-files (xml-apis-1.3.03.jar, xercesImpl-2.8.0.jar, xalan-2.6.0.jar) in the endorsed library under the jdk, and we have already added <web-app-class-loader search-local-classes-first="true" include-war-manifest-class-path="true"/> to our orion-web.xml.
    I also know that the <shared-library> tag got introduced in 10.1.3 where you easily can replace the parser.
    Any ideas on how to deal with this exception or comments to the interpretation on the exception are appreciated.
    -MK

    Hello,
    You are correct, in OracleAS 10g (10.1.2), Oracle does not expose JMX MBean server.
    OracleAS 10g R3 (10.1.3) is a J2EE 1.4 container will full support for JMX, and all the JSR-77 MBeans exposed. You can also create and register your own MBeans.
    Could you give us the different requirements you have around JMX for your application?
    Regards
    Tugdual Grall

  • Three-tier model howto

    Hi... i�m newbie in Java.
    I am supossed to create an application using three-tier model, i have made it using two-tier model.
    Can anybody give me a hand with the code,
    I need an example to start i can not fiugre this out.
    The application needs to interact with a database.
    I�ve heard that in the three-tier model you do not have to make the whole statement in one method, instead you have to create some utilitary classes in which each method returns data corresponding to a part of the query. How true is this?
    Please i need some examples!
    i am desperate.

    To get you started have a look at the RMI and JDBC API's. These are basicly all you need for a 3 tier client server archtecture. There are plenty of resources/examples at www.java.sun.com.

  • Performance issues with LOV bindings in 3-tier BC4J architecture

    We are running BC4J and JClient (Jdeveloper 9.0.3.4/9iAS 9.0.2) in a 3-tier architecture, and have problems with the performance.
    One of our problems are comboboxes with LOV bindings. The view objects that provides data for the LOV bindings contains simple queries from tables with only 4-10 rows, and there are no view links or entity objects to these views.
    To create the LOV binding and to set the model for the combobox takes about 1 second for each combobox.
    We have tried most of tips in http://otn.oracle.com/products/jdev/tips/muench/jclientperf/index.html, but they do not seem to help on our problem.
    The performance is OK (if not great) when the same code is running as 2-tier.
    Does anyone have any good suggestions?

    I can recommend that you look at the following two bugs in Metalink: Bug 2640945 and Bug 3621502
    They are related to the disabling of the TCP socket-level acknowledgement which slows down remote communications for EJB components using ORMI (the protocol used by Oracle OC4J) to communicate between remote EJB client and server.
    A BC4J Application Module deployed as an EJB suffers this same network latency penalty due to the TCP acknowledgement.
    A customer sent me information (that you'll see there as a part of Bug# 3621502) like this on a related issue:
    We found our application runs very slow in 3-Tier mode (JClient, BC4J deployed
    as EJB Session Bean on 9iAS server 9.0.2 enterprise edition). We spent a lot
    of time to tune up our codes but that helped very little. Eventually, we found
    the problem seemed to happen on TCP level. There is a 200ms delay in TCP
    level. After we read some documents about Nagle Algorithm,  we disabled a
    registry key (TcpDelAckTicks) in windows2000  on both client and server. This
    makes our program a lot faster.
    Anyway, we think we should provide our clients a better solution other than
    changing windows registry for them, for example, there may be a way to disable
    that Nagle's algorithm through java.net.Socket.setTcpNoDelay(true), in BC4J,
    or anywhere in our codes. We have not figured out yet.
    Bug 2640945 was fixed in Oracle Application Server 10g (v9.0.4) and it now disables this TCP Acknowledgement on the server side in that release. In the BugDB, I see backport patches available for earlier 9.0.3 and 9.0.2 releases of IAS as well.
    Bug 3621502 is requesting that that same disabling also be performed on the client side by the ORMI code. I have received a test patch from development to try out, but haven't had the chance yet.
    The customer's workaround in the interim was to disable this TCP Acknowledgement at the OS level by modifying a Windows registry setting as noted above.
    See Also http://support.microsoft.com/default.aspx?kbid=328890
    "New registry entry for controlling the TCP Acknowledgment (ACK) behavior in Windows XP and in Windows Server 2003" which documents that the registry entry to change disable this acknowledgement has a different name in Windows XP and Windows 2003.
    Hope this info helps. It would be useful to hear back from you on whether this helps your performance issue.

  • How to implement more than one physical models from the same model

    Hello,
    We have designed a relational model for our application and now, we would like to implement it for our 4 environments: Development, Test, User-Test and Real. Each one has the same logical schema but, different physical clauses (i.e.: in development environment we will not use partitioning clauses).
    Is this possible?
    Thank you,
    Bernat Fabregat

    Hello Bernat,
    yes it's possible:
    1) open your design and go to Tools>"RDBMS site administration"
    2) add your sites there (Current design) - Development, Test, User-Test and Real; press ok when finish
    3) go to "Physical models" node in browser and use "new" from pop-up menu - your sites will appear in list
    4) after you set physical objects in one model (need to save it as well)l, you can clone those objects in other models - you can clone from lower or the same database version
    You can consider usage of "name substitution" functionality during DDL generation if objects should have different name (prefix) in different environment - in this there will be no need to change the name of objects in models.
    Philip

  • 3-tier vs. 2-tier system architecture (DEV- QAS- PRD)

    Dear Xperts,
    I would like to know your opinion about a 3-tier (DEV->QAS->PRD) vs 2-tier (DEV/QAS->PRD) BW system architecture.
    Why should you choose for 3 instead of 2, or the other way around. Thus pro's & cons.
    Thx.
    regards,
    JvB
    ps xcuse me 4 the cross-post in the 2004s part. It should be posted in this forum.

    Hello
    2 Tier architecture will have Presentation Layer and DataBase Layer. So each and every time we have to access the Database directly and this may affect the DB Performance and if the number of concurrent users are more then it will be worser.
    in 3 Tier case another layer "Application layer" has been incorporated as an intermediate between the above said 2 layers and this prevents the direct access to the DB. The performance is not highly affected. can Manage multiple concurrent users.
    Heard about 4 tier came and installation process going in in Gulf. (not sure)
    Sankar.

  • The Commissionaire Model (aka the 3-tier  Model)

    Hi Guys !
    Commissionaire model is commonly followed by companies in Europe to get tax advantage.  In this model there is one company code(the Principal) which owns all inventory and records all sales and revenues whereas the Commissionaire company codes in different European countries face the customer and act on behalf of the Principal.  The customer only sees and deals with the Commissionaire(the principal is invisible to the customer).  At the end of the month, the Principal company code pays out a commission to the commissionaire company codes based on the sales made in their country.
    If anybody has worked with this model, I would like to talk to you ! I'm looking for the R/3 org structure in such a model and the logistic and financial document flow.
    I'm also looking for some reference SAP customers who have implemented this model.
    You can call me at 610-653-6627.  It is urgent !!
    Thanks for your help in advance
    Mohammed Jamaluddin
    COE Financials
    Active Global Support
    NSQ.

    Hi Mohammed
    We have such a scenario and below are some guidelines
    - use of plants aborad functionality
    - one face to customer functionality
    - single point of invoice model
    These are the functionalities to be used and you need to have many sales order types for this
    Thanks
    Ashok
    Assign points for useful answer

  • EJB business model architecture + packaging

    Hello,
    During a development of large Web applications (ERP for example), how to organize the business model (EJB) tier?
    I thought of centralizing the business model of the entire application within only one (ejb) module whose each Web module would depend there. What do you think ?
    Then, how to package it ? All the modules (Web and business) in the same ear? It would be simplest but not the ideal solution. Because for each update I have to redeploy the entire application.
    Thank you.

    Thanks for your help mr_jaber.
    I would like just to know what is the best way to package and deploy large application to the application server.
    Is better to use a single ejb-jar module who contain all buisness logic or split the buisness tier in multiple ejb-jar ?
    I know there is the admin_client.jar utility who allows incremental or partial redeployment of EJB modules within an application running in an OC4J instance. This is fine..
    So.. I have to put all my modules in a ear archive or is better to deploy Web and EJB modules in standalone mode ?

  • Difference between MVC and 3-Tier model?

    What is the difference between a Model View Controller design and a 3-tier design? I have read that the 3 tiers are independant, but that is also the case in the MVC design so are the two kind of designs not rather similar??

    When I think of the 3-Tier system, doesn't necessarily mean hardware, but a separation of laters of logic, which may or may not be physically independent.
    The 3 tiers are presentation, domain logic, and data source. When mapping these three layers to Model View Controller, the layers are not 1:1.
    Presentation maps pretty close to the View, but Model encompases both the Data Source and the most of the Domain Logic. It includes how you get the data and what you do with it. The Controller is that part of the domain logic that determines a request-response control (progression from one view to the next, calling the model logic necessary along the way).

  • Weblogic 8.1 - 2 Tier Proxy Architecture

    Hi,
    I want to know more about the round-robin behavior of the HTTP proxy when I set my infrastructure up like the 2 Tier Proxy illustrated at:
    http://e-docs.bea.com/wls/docs81/cluster/planning.html#1094289
    If you have more than one HTTP Proxy instance, does the Proxies communicate with each other such that it can coordinate the round-robin behavior?
    Next question, goes for both HW and SW load balancing. If I configure and start a server instance that joins the cluster while the PRoxy or HW load balancer is running, is it possible to update the Proxy or HW load balancer's server list such that the load balancer will know to spread the load to the newly configured and ready server?
    Thanks,

    The exception is thrown whenever I hit either the Commit or the Rollback button.

  • Implementation Options: Server Side Object Model & Client Side Object Model

    Hi All, I'm design a custom web application and it will be deploy on SharePoint 2010/2013 (to be confirmed) server, and I found some aritical as follows:
    http://consultingblogs.emc.com/sanjaypatel/archive/2011/09/17/comparison-between-server-side-object-model-client-object-model-rest-api-s-and-decision-matrix-on-what-to-use-when.aspx
    May I ask ask your advice for following items?
    1) Is the above URL still valid? e.g. Vice Versa for SharePoint 2013?
    2) Microsoft has a phase out plan for server side object model or not? (e.g. MOSS 202X)
    3) For programmatcally update of SharePoint permission: should I call client-side OM instead of server-side OM even I will deploy the web application to SharePoint server?
    Thank you very much.
    .NET Beginner 3.5

    In sharepoint 2013 I don't think this link is
    http://consultingblogs.emc.com/sanjaypatel/archive/2011/09/17/comparison-between-server-side-object-model-client-object-model-rest-api-s-and-decision-matrix-on-what-to-use-when.aspxis reliable at all.
    what is your say?
    ====================================================
     Hi All, I'm design a custom web application and it will be deploy on SharePoint 2010/2013 (to be confirmed) server, and I found some aritical as follows:
    http://consultingblogs.emc.com/sanjaypatel/archive/2011/09/17/comparison-between-server-side-object-model-client-object-model-rest-api-s-and-decision-matrix-on-what-to-use-when.aspx
    May I ask ask your advice for following items?
    1) Is the above URL still valid? e.g. Vice Versa for SharePoint 2013?
    2) Microsoft has a phase out plan for server side object model or not? (e.g. MOSS 202X)
    3) For programmatcally update of SharePoint permission: should I call client-side OM instead of server-side OM even I will deploy the web application to SharePoint server?
    Thank you very much.
    .NET Beginner 3.5

Maybe you are looking for

  • How to find the root document in my application, if application is deployed as  EAR file ??

              Hi,           I want the real path, i mean root document.           I my EAR file, i have only one WAR file. In my WAR file the following are the           folders:-           enterprise/..           properties/sql/..           locale/..   

  • Thumbnails in CAD desktop

    Hi, we have upgraded our 4.7 system to extension set 2.0 patch level 09. I have made the required configuration for Thumbnails in the "Applications-> parameters- > check in original for thumbnail" and also defined the GIF workstation application, but

  • What are the components & support package levels required in 4.7 to extract

    Hi All, In my landscape we have ECC 6.0 and 4.7EE, as of now we are extracting data from ECC 6.0 to BI 7.0.Now we have requirement to extract data from 4.7EE. Could you please suggest me what are the components & support package levels required in 4.

  • Mac Dictate - delete command?

    Using mac dictate on OSX, does anyone know the voice commands for delete or voice commands to 'teach' the app new words? It's not on the Mac basics page. Thx

  • Fine-tuning using a glossary

    Hello,  Please clarify if a relatively small glossary can be used to "hard-wire" or "lock in" some proper names and fixed constructions in the trained engine. As I understand, this can be done after cloning and launching a training with the glossary