Need design pattern advice

HI all,
Im trying to create a single router application that routes transactions to a number of Processing applications. Both applications reside on different machines. Upon system start up of the Processing application, the router will be notified and will register this application to its local registry.
When a transaction comes in, the router will redirect transactions to the processing applications depending on the processing applications load factor.
Im using jgroups and ha-jdbc to help me with group notification, how ever, i would like to know the best pattern to be used on Router applicaiton and Processing application.
Thanks in advance.

Thanks for the reply, but routers are not capable of
detecting the load of an application as well for the
heart beat of the communication.
Any thoughts?an F5 will do exactly that:
http://www.f5.com/
better than anything you'll ever write.
%

Similar Messages

Design Pattern help needed for MDI application

Hi,
I am writing an mdi application which contains a JSplitPane. The left pane contains a JTree and the right contains a JDesktopPane which houses n number of JInternalFrames. The JInternal Frames need access to the JTree. Also both panes (objects w/in) need access to elements of the JFrame. I am currently implemeting access by making most of the JFrame objects static and creating getter/setter methods for access. Does anyone know any desing patterns I might use to solve this dilemma?
Thanks,
John

The observer design pattern causes the child objects to chage state when the parent changes state. The type of pattern I need would allow the child objects of the parent JFrame to be available globally without having to make them static. I am wrong in this? Would the observer pattern solve this issue.

Need a Effective Design Pattern for Different Validation methods ???

Hi All,
I am having different validation methods with separate functions. I like to know the Best and Effective Design Pattern to do this.
for example
validateUserId()
validatePassword()
validateAuthentication()
validateSession()
validateConnection()
.... etc......
Can anybody help me to solve this?
Thanks,
J.Kathir

Hi there - is this the kind of thing you wanted?.
This structure has served me well. There is quite a lot to it and lots you need to infer but you should be able to work it out!
It works for EJB & swing etc (I believe)..
It doesn't use exceptions to return validation messages.
It doesn't use lots of little objects. The code is very explicit & pretty simple.
public class PersonValidator
Person myData;
public PersonValidator() {
//personId will be 0 and we are in nsert mode
myData = new Person();
public PersonValidator(long personId) {
//personId will have a value and we are in update mode
myData = PersonDb.getPerson(personId);
// =================================Single field validation
// =================================For swing apps to call - not over network.
// =================================Use validateData method
public String validateDateOfBirth(String dataOfBirthStr) {
String msg = "";
msg = validateDateStr(dataOfBirthStr);
if (!msg.equals("")) return msg;
public String validateSurname(String surname) {
String msg = "";
if (surname.equals("")) return "surname must be entered";
//=============================================Cross Validation
public List crossValidateAndSave(boolean saveData) {
String msg = "";
List valerrs = new ArrayList();
if (surname.equals("Collins") && dateOfBirth.compareTo("???")) {
List.add("Candidate is lying about their age. Please enter a realistic Date of birth");
if (valerrs.size() == 0) {
if (saveData()) {
personDb.saveData(myData);
// =========================To validate across the netwrok this lets you
// pass all the data in 1 go & receive multiple messages in reply.
// AN EJB can delegate to this code or this code could be in the EJB.
// @param stopAtFirstMessage flag allow efficiency if app ccan't handle multiple validation
// erros at a time.
// @param saveData flag allows validation without saving - allows for confirm message.
// @param EditPersonGuiData - What the user types in. Numeric data & dates can be
// in Strings. Using struts this could be the formBean.
public List validateData(EditPersonGuiData inputData, boolean stopAtFirstMessage, boolean saveData) {
String msg = "";
List valerrs = new ArrayList();
msg = validateDateOfBirth(inputData.dateOfBirth);
if (!msg.equals("")) valerrs.add(msg);
if (stopAtFirstMessage) return valerrs;
valerrs = crossValidateAndSave()
return valerrs;
}

Does any body have any design patterns of JSF Web Application Developping?

Can any one answer me some questions?
#1.I am an amatuar of people who develop Web Application.For some reason,We choose the JSF to develop our item.through some introduction,I know the UI component of JSF is resided in Server side,is it right?but I am a little confused that:if there are many users who are exploring our jsf website.(to simplify my question,image I had just one web page and just one button)How many UI components(buttons) will be there?How it(they?) works?
#2.Does any body have the success experience (for example design pattern)to develop web applications?if We just concern about the Add,Delete,Modify,Query operations of some data.
I just do my job according to my feeling.
I will give every page a pagebean(backing bean),and I am not sure how to combine the business data with the pagebean.some one suggested that I should use delegate pattern to separate my business log and page logic.But I am still confused by following things:
#2.1 does JSF have the same ability to help us construct the model dialog just like swing to
help us control the operation flow?
#2.2 If there is not,Does the following way work?I put every tabledata's property as corresponed component.if user choosed the row in the table,My Listener will syncronize the row data to the components.But
#2.2.1 if JSF has the components according to the web users' number,how can My Listener tell which component should be update?Should I maintain the map?
#2.2.2 If the problem I imaged above is false,Does any body can tell me how to custom the ListDataModel,so I can use it like Swing?because now I can just use some view data to insert into ListDataModel,but after some selection operation,my business object must be find according to the selected data,it is not an interesting job!
I am waiting for your advice!

Ok I'll try to explain Step by step please correct me if I make any mistake because I have not played much with shared variables.
To create a shared variable to an RT target go to the target if tou have already otherwise add an RT target by right clicking the Project>>Add targets and Devices
Then in the target Right clikc and select the variable as shown below.
Then once the Shared variable settings window opens Enter a variable name and then Select the type "Network Published"
In the right side you can select the data type for the shared variable and even you can choose your custome controls.
After selecting the data type go for the Network and select buffering if required else leave it if you are planning to use the variable just for display purpose.
Then you can enable the RT FIFO if required (Not able to explain how it works and why it is used).
Then after completing the Shared variable setup you can access the variable in the VI in both the Host and the RT.
You can bind the variable to a control so that if any data from the RT is coming you can read the data from that control.
Once you have placed your shared variable in the BD you can change the access typr to read or write depending on your need.
This might not explain the complete shared variable concept but I believe that this would defenelty give you a kick off to start using the shared variable. Please correct or add more comments if anybody know better.
Good luck.
The best solution is the one you find it by yourself

Design pattern -- using the correct one...?

I'm in need of some advice regarding design patterns in general.
I have a prototype whereby Objects B, C and D observe Object A's attribute (let's call it theAttribute -- imaginative, eh? )
In my system, every time theAttribute is changed, I've implemented the Gang's Observer pattern so that the observer objects are notified (...and do something with the information that Object A has changed in some way... they basically update a theTimestamp attribute when the subject's updated... but that really is beside the point right now).
However, only ONE observer will ever be created for each observable subject.
To my dismay, I've just been reading on the IBM Research site that the memory implications of this are horrendous (that for potentially a great many subjects only having a few observers each is bad practice) -- d'oh.
I've looked at other ways I could achieve what I need to without the above practice, but the only things I can come up with are the Mediator pattern or the State pattern, neither of which I'm sure are ideal (the former because of I'm worried that it'd be total overkill, the latter because I don't really need dynamic reclassification if the truth's told, I think).
Anyone got any thoughts on this? (I've considered referencing observers with their observables using a hash table but don't really wish to go down that route unless absolutely neccesary... it seems a bit like putting electrical tape over a sparking wire :) I'd rather just do it properly).
Sarah.

However, only ONE observer will ever be created for each observable subject.If you can absolutely, positively, definitely, mathematically, logically, legally
and morally prove that fact, indeed, you're wasting some resources here,
a couple of hundred bytes mayhap ...
In the mean time, keep your Observer/Observable pattern/implementation
or have a look at the PropertyChangeSupport class ...
kind regards,
Jos

Program for designing patterns for fabric

I need advice on which program would be best for designing patterns to then use for fabric printing, wall paper etc.
I am a complete adobe novice.
Thanks,
Kim

Hi kimtreasure,
For designing you can use Adobe Illustrator and Adobe Photoshop.
For more info:
Photoshop : Adobe Photoshop CC new features | Buy Photoshop CC
Illustrator : Adobe Illustrator CC new features | Buy Illustrator CC
Download trial version.
Creative Cloud Help | Install, update, or uninstall apps
Thanks,
Atul Saini

Cor design pattern

I use cor design pattern little different than what I see in google.
here is m y cor
     public static abstract class CORHandler    {
          CORHandler nextHandler;
          public void process(){
               if(canHandleRequest()){
                    handleRequest();
               }else {
                    if(nextHandler!=null) nextHandler.process();
          protected abstract boolean canHandleRequest();
          protected abstract void handleRequest();
          public void setNextHandler(CORHandler corHandler){
               nextHandler=corHandler;
     }this serves most of my need and also I can override process method for change behaviour .
this code helps me save lots of if/else conditions , I am not passing any object to the method here reason is I use this always as inner class inside a method so I can access all final fields
advice me is this right when applying cor pattern?

Making it a base class increases connectivity to no good purpose.
It violates the concept of inheritance by making the objects COR classes which is not what they are.
It violates the principal of COR in that the intent is to handle a responsibility not all responsibilities. This same problem obscures the pattern as well by eliminating the objects to be acted upon.

How can I develop a web application using EJB design pattern?

I have searched over the web and found quite a lot of tutorials on how to use the EJB design pattern.
I know that there will be a home interface, EJB object interface and a SessionBean.
But the tutorials often only cover a single class, this made me unable to get a complete picture of how EJB design pattern can be implemented into a whole system.
I am now required to devleop an online shopping web application using EJB and JSP page.
I think I will need to create a lot of classes: Member, ShoppingCart, Product...etc.
What I want to ask is that, do I need to create a home interface, EJB object interface and a SessionBean for each of these classes?
I really need some ideas on how to develop this system using EJB + JSP pages.
Many thanks to you all.

For every EJB that you want to create, you will need to code a home and remote interface and a bean class.
You could start getting your ideas here
http://www.theserverside.com/books/wiley/masteringEJB/
http://www.coreservlets.com

SERVICE LOCATOR ?? Is it really an interesting Design pattern??

Hi everybody,
i've a problem with the J2EE Design Pattern "Services locator" (it's a singleton).
It is said that by making use of a Service Locator we can :
- hide to the client the complexities of initial context creation, EJB home object lookup,and EJB objectre-creation.
- multiple clients can reuse the Service Locator object to reduce code complexity, provide a single point of control, and improve performance by providing a caching facility.
But i would like to understand at which side should that service locator object reside??!!??
If it is at server side then the clients need well an initial context in order to make a lookup on that object.
Conclusion :
the service locator doesn't hide the complexities of initial context!!
Furthermore the client has to perform a look-up on that service locator object!! The only advantage left is caching facility.
If it is at client side, each client needs his own services locator object
Conclusion :
multiple client don't reuse the same service locator. What's the advantage to be a singleton ???
There is certainly something that i don't understand so help me please!! Thanks.

Hi Yves,
But i would like to understand at which side should
that service locator object reside??!!??
If it is at client side, each client needs his own
services locator object
Conclusion :
multiple client don't reuse the same service locator.
What's the advantage to be a singleton ???The service locator resides on the client side and is implemented as
a singleton. Since it is possible that there could be multiple
class loaders/JVMs on the client side, and therefore, multiple
instances of the "singleton" service locator. This is typical
in a distributed environment (e.g. servlets/JSPs in a web-tier
cluster using service locator). Thus service locator is not
a truly "distributed singleton" object. But, the empahsis
is to design the service locator such that it does not hold
any state that needs to be replicated across multiple
instances across different JVMs as mentioned. Thus, there
is no need for multiple clients to use the "same" service locator,
but still the benefits of implementing this pattern is realized.
By making it a singleton, and keeping it from holding state
that needs to be replicated, we realize the benefits of this pattern.
You may also want to visit the J2EE Pattern interest list
and see these relevant discussions :
Topic: Service Locator and passivation
http://archives.java.sun.com/cgi-bin/wa?A2=ind0106&L=j2eepatterns-interest&F=&S=&P=1026
Topic: Caching EJBHome interfaces
http://archives.java.sun.com/cgi-bin/wa?A2=ind0106&L=j2eepatterns-interest&F=&S=&P=9226
Topic: Using Service Locator for Data Source caching
http://archives.java.sun.com/cgi-bin/wa?A1=ind0106&L=j2eepatterns-interest#31
hope this helps,
thanks,
-deepak

Business delegate and Session facade design patterns

Does any one tell me, what is the difference between business delegate and session facade design patterns.

1. Session Facade decouples client code from Entity beans introducing session bean as a middle layer while Business Delegate decouples client code from EJB layer ( Session beans).
2. SF reduces network overhead while BD reduces maintenance overhead.
3. In SF any change in Session bean would make client code change.
While in DB client is totally separate from Session bean because BD layer insulate client from Session beans(EJB layer).
3. In only SF scenario, Client coder has to know about EJB programming but BD pattern no EJB specialization needed.
4.SF emphasizes on separation of Verb, Noun scenario while BD emphasizes on separation of client(presentable) and EJB layer.
Anybody pls suggest more differences ?

Producer/Consumer Design Pattern with Classes

I'm starting a new project which involves acquiring data from various pieces of equipment using a GPIB port. I thought this would be a good time to start using Classes. I created a GPIB class which contains member data of: Address, Open State, Error; with member vis such as Set Address, Get Address, Open, Close...general actions that all GPIB devices need to do. I then created a child class for a specific instrument (Agilent N1912 Power Meter for this example) which inherits from the GPIB class but also adds member data such as Channel A power and Channel B power and the associated Member Functions to obtain the data from the hardware. This went fine and I created a Test vi for verfication utilizing a typical Event Structure architecture.
However, in other applications (without classes) I typically use the Producer/Consumer Design Pattern with Event Structure so that the main loop is not delayed by any hardware interaction. My queue data is a cluster of an "action" enum and a variant to pass data. Is it OK to use this pattern with classes? I created a vi and it works fine and attached is a png (of 1 case) of it.
Are there any problems doing it this way?
Jason

JTerosky wrote:
I'm starting a new project which involves acquiring data from various pieces of equipment using a GPIB port. I thought this would be a good time to start using Classes. I created a GPIB class which contains member data of: Address, Open State, Error; with member vis such as Set Address, Get Address, Open, Close...general actions that all GPIB devices need to do. I then created a child class for a specific instrument (Agilent N1912 Power Meter for this example) which inherits from the GPIB class but also adds member data such as Channel A power and Channel B power and the associated Member Functions to obtain the data from the hardware. This went fine and I created a Test vi for verfication utilizing a typical Event Structure architecture.
However, in other applications (without classes) I typically use the Producer/Consumer Design Pattern with Event Structure so that the main loop is not delayed by any hardware interaction. My queue data is a cluster of an "action" enum and a variant to pass data. Is it OK to use this pattern with classes? I created a vi and it works fine and attached is a png (of 1 case) of it.
Are there any problems doing it this way?
Including the error cluster as part of the private data is something I have never seen done and ... well I'll have to think about that one.
Ben
Ben Rayner
I am currently active on.. MainStream Preppers
Rayner's Ridge is under construction

Producer/Consumer Design Pattern + DAQmx + Event Structure

Dear community users,
at the moment I try to implement the Producer/Consumer Design Pattern into my current project. For this purpose I am trying to make a "minimal-example" in which I use the Producer/Consumer Design Pattern for the data acquisition and analysis (display in GraphXY, calculation etc..) process. In the Producer Loop I perform the data readings via DAQmxRead.VI and in the Consumer Loop, in this example only, I display the data in a GraphXY. At the same time I want to control the data acquisition via an Event Structure, for example Start Measurement/Stop Measurement Buttons. To configure the measurement time, for example if the user only wants to measure 150 sec, I use the Time Out Event of the Event Structure. This is a technique which is often used, when data acquisition and Event Structures are used at the same time (Excuse me if this is wrong!). Due to understand the Producer/Consumer Design Pattern I looked up in the Template\Framework of LabVIEW.
In the attachment you will find my "minimal-example", which -sadly-, not working. What should I do ?
Thank you.
Best regards,
tugrul öztürk
Attachments:
daqEngine_PCP.vi ‏35 KB

Your VI will never stop due to the Ok Button that stops the producer loop is read OUTSIDE of the loop. Add an event case for it.
Move the Start and the Ok Button inside of their respective event cases. This will allow the latch to reset when they are read.
Change the Ok Button to a stop button.
Use a Greater Or Equal instead of Equal for the measurement time comparison. What if somebody randomly changes it to 0 on you? Again, you will never stop.
Is the DAQ only supposed to read for "Measurement Time". Right now, it constantly reads. So if you don't use the DAQmx Read, you will get a buffer overflow error. You need to put the DAQmx Start Task inside of the Start event case and then the Stop DAQmx Task inside of the Timeout event case when the measurement time is exceeded. You should also perform the read regardless inside of the Timeout event case.
I HATE telling the consumer loop to stop by destroying the queue. You can look data that way. Instead, send something an empty array. In the consumer, check for an empty array. If it is empty, stop. Destroy the queue only AFTER BOTH LOOPS HAVE COMPLETED.
You should also report your errors in some way. Simple Error Handler at the minimum.
Since you are using Continous Samples, don't limit the buffer. Leave the Samples per Channel unwired on the DAQmx Timing VI.
There are only two ways to tell somebody thanks: Kudos and Marked Solutions
Unofficial Forum Rules and Guidelines

Choice of design pattern for data acquisition system

Hello all
I have a trouble about selecting the suitable design pattern / architecture for a data acquisition system.
Here is the details of the desired system:
There is data acquisition hardware and I need to use it by observing parameters on User interface.
the data acquisiton period, channel list to scan should be chosen on User interface. Besides, there are many user interface interactions. e.g. if user selects a channel to add scanlist, then I need to enable and make visible some other parts on user interface.
When user completes the channel selection, then he will press the button to start data acquisition. Then I also need to show the scanned values on a graph in real time and log them in txt file.
I know that I cannot use producer consumer pattern here. because the data acquisition loop should wait for parameters to scan channels. and it works in a given period by user. so the user interface loop performs higher rate then consumer loop (data acquisition loop). it means queue will be bigger bigger. if I use notifier it will loss some data come from user interface.
is there any idea about that ? is there any suitable design pattern for this case ?
Thanks in advance
best regards
Veli BAYAR
Embedded Systems Software and Hardware Engineer
"You live in a graphical world. Why not program in one?"
Solved!
Go to Solution.

johnsold wrote:
Veli,
I recommend the Producer/Consumer model with some modifications.
You might need three loops. I cannot tell for sure from your brief description.
The User Interface loop responds to the user inputs for configuration and start/stop of acquisition. The parameters and commands are passed to the Data Acquisition loop via a queue. In this loop is a state machine which has Idle, Configuration, Acquisition, and Shutdown states (and perhaps others). The data is sent to the Processing loop via a different queue. The Processing loop performs any data processing, displays the data to the user, and saves it to file. A notifier can be used to send the Stop or shutdown command from the User Interface loop to the other loops. If the amount of processing is minimal and the file write times are not too long, the Processing loop functions might be able to occur in the Timeout case of the UI loop Event structure. This simplifies things somewhat but is not as flexible when changes need to be made.
I am not sure that a Design Pattern for this exact setup exists but it is basically a combination of the Producer/Consumer (Events) and Producer/Consumer (Data) Design Patterns.
Lynn
Check out this thread: http://forums.ni.com/t5/LabVIEW/Multiple-poll-case-structures-to-event-help/td-p/2551309
There are discussions there about a 3-loop architecture that may help you.
Jeff
Jeffrey Zola

SAP Fiori - Layout Design patterns

Dear All Gurus,
This may be a layman question but expecting your expert advice and perspective on this.
I am creting a fully customized fiori application -" Purchase Order Creation " and User wants to have the UI design very simillar to SAP screens(ME21n) adding only required fields into that.
I know that fiori has a design pattern(Ex : Main items in left and details in right side , 'back' button etc..) which gives me a good impression. But if I go ahead with user's requirement will there be an impact ? Ex : performance / run on mobile.
Thanking you all in advance..
Tags edited by: Michael Appleby

Hi Masayuki,
Thank you for the reply..
+Target device: windows desktop(mainly) and mobile
+Who will use: Employees (mainly)/MRP specialist
+Number of users: 500
+Is there Purchase Requisition process before Purchase Order : NO . Direct PO creation. But PO Approval is there using srtandard Fiori App.
+PO will be created entering everything manually
+Where is the data source for creating PO? : Original data is on paper

Design pattern / data loading solution

Hello all!
I have been working on a few projects which involve loading data, sometimes remotely, sometimes local, sometimes JSON, sometimes XML. The problem I am having is that due to the speed of development and the changing minds of various clients I am finding my designs are too rigid and I would like them to be more flexable. I have been trying to think of a reusable solution to data loading, and would like some advice as I imagine many of you out there have had the same problem.
What I would like to do is to create a generic LoadingOperation abstract class, which has member variables of type Parser and Loader, which have parse() and loadData() methods respectivly. The Parser and Loader classed are interfaces and classes that implement these could be XMLParser and JSONParser, LocalLoader and RemoteLoader etc. With something like this i would like to have a new class which extends LoadingOperation for each thing to be loaded, weather thats a local XML file, or remote JSON, or whatever.
The problem is is that specific Parser implementation cannot return custom data types without breaking the polymorphic behavior of the LoadingOperation class. I have been messing around with generics and declaring subclasses of LoadingOperation like
class SpecificLoader extends LoadingOperation<CustomDataType>
and doing similar things with Parser classes, but this seems a bit weird.
Does anyone have any suggestions on what im doing wrong / could be doing better. I want to be able to react quickly to changing specifications (ignoring the fact that they shouldnt be changing that much!) and have a logical seperation of code etc...
thanks for any help!
psi have also asked this question here [http://stackoverflow.com/questions/4329087/design-pattern-data-loading-solution]

rackham wrote:
Hello all!
I have been working on a few projects which involve loading data, sometimes remotely, sometimes local, sometimes JSON, sometimes XML. The problem I am having is that due to the speed of development and the changing minds of various clients I am finding my designs are too rigid and I would like them to be more flexable. I have been trying to think of a reusable solution to data loading, and would like some advice as I imagine many of you out there have had the same problem.
What I would like to do is to create a generic LoadingOperation abstract class, which has member variables of type Parser and Loader, which have parse() and loadData() methods respectivly. The Parser and Loader classed are interfaces and classes that implement these could be XMLParser and JSONParser, LocalLoader and RemoteLoader etc. With something like this i would like to have a new class which extends LoadingOperation for each thing to be loaded, weather thats a local XML file, or remote JSON, or whatever.
The problem is is that specific Parser implementation cannot return custom data types without breaking the polymorphic behavior of the LoadingOperation class. I have been messing around with generics and declaring subclasses of LoadingOperation like
class SpecificLoader extends LoadingOperation<CustomDataType>
and doing similar things with Parser classes, but this seems a bit weird.
Does anyone have any suggestions on what im doing wrong / could be doing better. I want to be able to react quickly to changing specifications (ignoring the fact that they shouldnt be changing that much!) and have a logical seperation of code etc...That depends on the specifics.
The fact that it seems like processes are similar doesn't mean that they are in fact the same. My code editor and Word both seem to be basically the same but I am rather sure that generalizing between the two would be a big mistake.
And I speak from experience (parsing customer data and attempting to generalize the process.)
The problem with attempting to generalize is if you generalize functionality that is not in fact the same. And then you end up with conditional logic all over the place to deal with differences dependent on the users. Rather than saving time that actually costs time because the code becomes more fragile.
Doesn't mean it isn't possible but just rather that you should insure that it is in fact common behavior before implementing anything.

Need design pattern advice

Similar Messages

Maybe you are looking for