Best Practice On Taking Count in Bean

Hi folks,
Here i'm using Jdev 11.1.1.7.0. where I have a case take count on inserted rows in bean.
here we go..
Bean code
developer way 1:
button_click(){
int count = voins.getRowcount();
developer way 2:
button_click(){
int count = iteratorbind.getEstimatedRowcount();
developer way 1: fetching view object in bean for purpose of taking row count.
---developer 1 says that best practice on taking count is using view object 'getRowcount()'.
Note: developer 1 comment that developer 2 way is worst practice. it's a time taken process
developer way 2: using iterator getting row count
---developer 2 says that for taking count we can use 'estimatedRowcount'.
Note: developer 2 comment that developer 1 way is worst practice taking view object on bean.
i.e developer 1 try catch the model layer directly from the code. we can use iterator to do all the stuff.
which is best practice please suggest me... is there any other practice taking row count in bean.
thanks.

Dev way 1)
A look into the java doc reveals
Note that this method retrieves all rows from the database then returns the number of rows in the Row Set collection.
So if your table contains a lot of rows calling getRowCount takes a lot of time and memory.
Dev way 2)
Calling getEstimatedRowCount executes a select count(*) from (your query) to get the count. This is fast but the task count may differ if many changes are done on the table. I personally had never a problem with this, but I can't speak for you.
Timo

Similar Messages

Best practice for taking Site collection Backup with more than 100GB

Hi,
I have site collection data is more than 100 GB. Can anyone please suggest me the best practice to take backup?
Thanks in advance....
Regards,
Saya

Hi
i think Using powershell script we can do..
Add this command in powershell
Add-PSSnapin Microsoft.SharePoint.PowerShell
Web application backup & restore
Backup-SPFarm -Directory \\WebAppBackup\Development -BackupMethod Full -Item "Web application name"
Site Collection backup & restore
Backup-SPSite http://1632/sites/TestSite -Path C:\Backup\TestSite1.bak
Restore-SPSite http://1632/sites/TestSite2 -Path C:\Backup\TestSite1.bak -Force
Regards
manikandan

Bean load best practice

I am not new to Java, but up until now have been a programmer. I am now getting more into design and architecture and have a question about best practice. This question arises from a UML class I was taking. But in the class we stayed within the UML and did not get into implementation.
My Question
When creating classes and designing how they interact, what is the best practice for implementing associative relationships. For example, if I were modeling a Barn that contained Animals, I would create a Barn bean and an Animal bean. Since the Barn contained Animals I could create the code like this:
public class Barn {
String color;
Collection animals;
void setColor(String newcolor);
String getColor( );
void setAnimals(Collection newanimals);
Collection getAnimals( );
public class Animals{
String name;
void setName(String newname);
String getName( );
}The Collection within the Barn bean would be made up of Animal beans.
This seems fairly straight forward. However, what if I loaded the bean from a database? When building the bean, do I also find all animals and build the Animal beans and create the Collection to store within the Barn object?
Or
Do I omit the animal Collection from my Barn bean and only populate the Collection at runtime, when someone calls the getAnimals method?
I am confident that the latter is the better design for performance and synchonization issues. But I wanted to get other opinions.
Do I need to read up more on object design?
Thanks,
Lonnie

And lazy initialization. Basically, unless the data is needed. Don't load it.

JSF - Best Practice For Using Managed Bean

I want to discuss what is the best practice for managed bean usage, especially using session scope or request scope to build database driven pages
---- Session Bean ----
- In the book Core Java Server Faces, the author mentioned that most of the cases session bean should be used, unless the processing is passed on to other handler. Since JSF can store the state on client side, i think storing everything in session is not a big memory concern. (can some expert confirm this is true?) Session objects are easy to manage and states can be shared across the pages. It can make programming easy.
In the case of a page binded to a resultset, the bean usually helds a java.util.List object for the result, which is intialized in the constructor by query the database first. However, this approach has a problem: when user navigates to other page and comes back, the data is not refreshed. You can of course solve the problem by issuing query everytime in your getXXX method. But you need to be very careful that you don't bind this XXX property too many times. In the case of querying in getXXX, setXXX is also tricky as you don't have a member to set. You usually don't want to persist the resultset changes in the setXXX as the changes may not be final, in stead, you want to handle in the actionlistener (like a save(actionevent)).
I would glad to see your thought on this.
--- Request Bean ---
request bean is initialized everytime a reuqest is made. It sometimes drove me nuts because JSF seems not to be every consistent in updating model values. Suppose you have a page showing parent-children a list of records from database, and you also allow user to change directly on the children. if I hbind the parent to a bean called #{Parent} and you bind the children to ADF table (value="#{Parent.children}" var="rowValue". If I set Parent as a request scope, the setChildren method is never called when I submit the form. Not sure if this is just for ADF or it is JSF problem. But if you change the bean to session scope, everything works fine.
I believe JSF doesn't update the bindings for all component attributes. It only update the input component value binding. Some one please verify this is true.
In many cases, i found request bean is very hard to work with if there are lots of updates. (I have lots of trouble with update the binding value for rendered attributes).
However, request bean is working fine for read only pages and simple binded forms. It definitely frees up memory quicker than session bean.
----- any comments or opinions are welcome!!! ------

I think it should be either Option 2 or Option 3.
Option 2 would be necessary if the bean data depends on some request parameters.
(Example: Getting customer bean for a particular customer id)
Otherwise Option 3 seems the reasonable approach.
But, I am also pondering on this issue. The above are just my initial thoughts.

Best practice for retail store cycle count

Hi experts
Would you have any experience/best practice in designing for cycle count process for your clients?
By cycle count, I mean for frequent counting (daily) for specific item category / exceptional count request.
My client want to implement a cycle count process to minimize the effort for full stock take which happen around 3 times per year.
They are using SAP as the core system to keep track of inventory, and using external store system, so some considerations would be on, for example:
- whether the count variance or counted quantity shall be interfaced back to SAP
- whether SAP should be the source to initiate a stock count process or POS initiate
It would be very nice if you can share some best practice as in the process flow, system information flow, etc.
Best regards
Dominic

Hello
Both can be done but as you say SAP is the core system at least for tracking inventories, better create and integrate inventories is SAP aswell.
The main advice I would give is to avoid specific tables for inventory integration because standard tables and specific won't be updated well sometimes and support would be needed, stick on standard transactions.
If the inventory is centralised in SAP, you can for example count them with PDA, send to SAP connectors your counting which will be integrated from a xml file transformed in a Idoc for SAP.
One parameter is important to consider as the freeze for inventory. You have to decide if you take into account any stock movement that could occur after your inventory creation or not. If you do, the variance after the counting will depend on the initial stock value and also the movement.
If you don't, the variance will only take into account your initial stock value withotu considering the stock movement.
Hope it helps you,
Génia.

Best practice for backing bean population? (also, ActionListener RANT)

Hello,
I am about 3/4 of the way through development of a small to medium size JSF application. Sometimes I really like JSF, but much of the time I am left puzzled or frustrated for hours trying to find workarounds to JSF's bugs/glitches and design flaws.
For example, early on, I was impressed with how easily it was to invoke a method from a page using an actionlistener. Now that I'm actually building things with JSF, the actionlistener funtionality still seems cool, but incredibly half baked. I find myself using request parameters LIKE CRAZY to work around the fact that JSF doesnt support passing parameters directly to backing bean methods. This feels awkward and wrong considering the fact that JSF is intended to abstract the HTTP underpinnings. To add insult to injury, I often have to iterate through ALL of the request parameters looking for one that has an id with an ending matching my desired property name (since JSF appends it's own crap to the beginning). I don't like doing things in a hacky way. This seems very hacky, and I feel dirty doing it.
So, my first question is, what is the best practice for populating backing beans??? How do others accomplish this. I can think of several other approaches, but none feel less hacky.
Second, are there plans in the next spec (please say there are) to allow parameters to be passed to backing bean methods? If not, WHY THE HECK NOT?
Even though JSF expert group people have been conspicuously absent from this forum of late, I'd really appreciate responses from you as well.
Thank you for your thoughts.

Hi BrownBear,
I've been using JSF for about 6 months now and I'd be glade to help as much as I can.
Concerning parameters, I'm not sure what your issue is but I use the f:param tag to pass them. If you could post an example of what you are trying to do, I could see exactly what your issue is. Maybe the f:param can't help you.
As for best practice for populating backing beans, I personaly try to let JSF do as much as possible. For example, if I have a backing bean with five properties, I make sure that they all are on the JSP page the bean serves. If one of the property is just there as an Id like, lets say, a Person ID (DB row key), then I put it on my JSP page as a hidden input field. I do the same with the properties that only for display, if I want them to be back in my bean when request comesback.
Hope this help some how. Please, feel free to ask specific questions related to your specific problem and I monitor this post and trnasfer to you the ;little JSF experience I have.
I'm pretty happy with JSF as it is but it sure needs improvements. :) What the heck, it's version 1.01 after all, and the next release should be a great one with the integration of JSTL.
Cheers

Looking for best practice on application scope beans

Hey â€“ a portal newbie here. Iâ€™ve got some application scope beans that need to be initialized on startup. First thought was to create a servlet that would set the bean. Then I saw the GlobalApp setting, but I think that looks like it is more session scope than applicationâ€¦ Looking to be corrected here if I am wrong.
Is there a place where these type of things traditionally happen? Read only, so no cluster worries (I think) Using WLP 8.1 SP4 and looking for best practices. Thanks for the help!

To support "code sharing" you need an integrated source code control system. Several options are out there but CVS (https://www.cvshome.org/) is a nice choice, and it's completely free and it runs on Windows, Linux, and most UNIX variants.
Your next decision is on IDE and application server. These are usually from a single "source". For instance, you can choose Oracle's JDeveloper and Deploy to Oracle Application Server; or go with free NetBeans IDE and Jakarta Tomcat; or IBM's WebSphere and their application server. Selection of IDE and AppServer will likely result in heated debates.

Best practice for initializing objects in a JSF backing bean?

Hi,
What is the best practice for initializing some objects in the JSF to-page backing bean before the to-page is displayed for the first time? The initialization would vary and depend upon a command link in the from-page.
Regards,
Al Malin

f:view has two new attributes in 1.2: beforePhase and afterPhase
which allows you to specify a phase listener method
which will be called before and after the view is processed.

JSF best practice for managed bean population

hi;
consider a simple session scope managed bean with 1 attribute which is calculated dynamically.
there seem to be 3 ways of populating that attribute (that i could think of):
1. have the bean's get<attribute> get the data and populate it
2. have the constructor of the bean get the data and populate it
3. have an action get the data, and set it from the action code (maybe use a valueBinding to get the bean if the bean which needs population is not the same bean that holds the action) .
option 1 seems ok for situations where i need my data to be calculated for every valueRef for that attribute but is an overhead for data that need not be calculated for every valueRef.
option 2 will ensure that te data is calculated only once per session. but it still looks kinda strange to do it
with option 3, it seems you should populate the dynamic content of the next page and its intuitive but some places in the page might be common content with dynamic data and this has nothing to do with navigation and the next page to be displayed.
is there a best practice of how to populate a managed bean data?
are different cases fit different ways?
any thoughts are appriciated.

I think it should be either Option 2 or Option 3.
Option 2 would be necessary if the bean data depends on some request parameters.
(Example: Getting customer bean for a particular customer id)
Otherwise Option 3 seems the reasonable approach.
But, I am also pondering on this issue. The above are just my initial thoughts.

Bean best practice question

Simple questions (hopefully), I know how to code this but just want some advice on the best way to do the following:
1. User enters data into HTML form and submits
2. Some Java at the backend grabs these details and emails them off somewhere
I am thinking of doing the following, but whats the best way?
1. HTML form submits and data is sent directly to a JavaBean (FormBean.java)
2. FormBean.java contains standard getters/setters but also contains a method called sendMail(), is this bad practice? Do I need a second Bean sendMail.java? Or is this completely the wrong way to do things i.e. should I do this entirely in a servlet with only 1 bean to grab the data (FormBean) and then access from servlet?
Just a bit confused on whats best practice for this stuff?
Thanks!

2. FormBean.java contains standard getters/setters but also contains a method called sendMail(), is this bad practice? Do I need a second Bean sendMail.java? A better approach is this way
a) Have all the form data in the form bean
b) Write sendMail in a all together different class, as action.
c) Send the form bean as a parameter to sendMail for processing and sending an email
This way your sendMail() will become a kind of a service. Tomorrow you might have some other data, which you will have to send it in an email. In that case, you just reuse sendMail() method. Otherwise, if you have sendMail() in form bean itself, then if there are many form beans, then you would have to write sendMail() in every form bean, which is a bad practice. One principle of OOAD is to separate the functionality, which is redundant in your classes and make it as a separate module. If there are changes to the sendMail() functionality then, by having it in one module, you only have to change it at one place.
Or is this completely the wrong way to do things i.e. should I do this entirely in a servlet with only 1 bean to grab the data (FormBean) and then access from servlet?You can have a servlet which acts like a controller, which receives the request parameters, constructs the form bean and invokes appropriate Action (in your case sendMail()). This is same as an MVC framework. Instead of you re-inventing the wheel to create a servlet controller, form bean, action etc. You could use one of the several MVC frameworks available in the market, such as Struts or Spring MVC.

Managed Bean Best Practices

Hi
Are there any best practices for using Managed Beans?
We plan to use our own custom-built JSF components. Need to understand how to design backing beans for performance/effort optimization.
For example :
1. How to make managed beans thread-safe for concurrent requests, without compromising on efficiency/speed?
2. How to enforce the J2EE security with managed-beans?
3. How to decide the scope of these beans to ensure minimal data-storage in session?
4. How to decide the granularity at which a managed-bean should be used for example :
4.1 One bean-per-component
Advantages :
a) if complex components require special data-holding/processing/event-handling capabilities from bean
(e.g. datagrid model,tree model,menu model)
Problems :
- with multiple components in a page/form
a) it becomes tedious to debug/change/track which bean serves which component
b) if session scope is required, too many beans will be cached in session
c) unnecessarily too many beans will be created on server (= n pages * m components-per-page)
d) unnecessarily increases the length of faces-config.xml
4.2 One bean-per-form
Advantages :
a) in multi-form web pages, to ensure the functional behaviour of each form is separate/modular in its own bean.
b) each managed-bean would map to specific/meaningful functionality/user-interaction in use-case.
Problems :
- if form includes complex components (datagrid/tree/menu) requiring a specialised bean class, then
a) either one of the specialized bean has to be augmented with additional logic to handle data/events for all other components within the form
(Not good, as it mixes-up the responsibilities of component-specific-beans, and the bean may no more be reusable in another form)
b) or without using component-specific beans, only single form bean should handle the data/events for all components in the form?
(Neither good, since if a complex tree-compoent is reused in multiple forms, then the logic to handle data/events for such a component will be repeated in those many form-specific managed beans)
4.3 One bean-per-page
Advantages :
a) seems more modular/meaningful way - since a page would map to some feature within the use-case
b) bean will contain only behaviour which is relevent for the associated page/function within use-case
Problems :
a) in multi-form pages, can single bean handle data/events for multiple forms?
b) if page uses complex component (e.g datagrid, tree) that needs its own bean - how does page-bean exchange data with component-bean?
Thanks,
Arti

Are there any best practices for using Managed
Beans?There are no best practices for using Managed Beans in terms of Sun, or other vendors recommendations. But there are some patterns that can be applied to the managed beans (The managed bean is already a pattern). Also common sense is allways a good practice.
For example, the managed bean should not have business logic code, only presentation logic code, etc.
1. How to make managed beans thread-safe for
concurrent requests, without compromising on
efficiency/speed?The beans can be created by request, so concurrency is not an issue. If they are session scope, also is not an issue because, a user can only have one thread running. Only in application scope you must have carefull.
>
2. How to enforce the J2EE security with
managed-beans?see this:
http://jsf-security.sourceforge.net/
About jsf-security
>
3. How to decide the scope of these beans to ensure
minimal data-storage in session?If you wnat minimal data-storage in session the answer is request or none.
In question 4, you make the question and give the answer ;-)

JSF CDI : Conversation scope bean[s] best practice

*JSF CDI : Conversation scope bean[s] best practice*
Hello, im currently learning about JSF 2.0 and im so glad for the existence of this conversation scope feature, which is very helpful in opening a new tab or a new window on the same page and having separate resources, not overriding one another.
But im curious on how to implement this in a good way, about when to start the conversation and when to close it.
In my case, i have each CDI bean for each JSF page. And let's say that i have a menu, and when it's clicked, this will lead to page A, and from A, could lead to B, B could lead to C, C could lead to D, all these 4 pages are connected in one chain.
Accessing A's bean properties from B or C or D beans is possible, accessing B's properties is also possible from C or D beans and so forth.
Now im quite confused about :
* whether all these A B C D should be in conversation scope or not, or perhaps just A ? Because i think sometimes from another page that is outside the ABCD chain, like a page F, it could navigate to page B, although i dont know how to supply the data to the bean B yet.
* whether all these A B C D should be combined into one bean
* where and when to start the conversation, im thinking about the constructor, but i dont think it's a good idea, because i prefer starting the conversation when first accessing the page, not the bean
* where and when to stop the conversation, so that there wont be unused resources hanging around
Regards,
Albert Kam

There are also some commandButtons on the view page, which performs actions on Product object, fetched by passed ID.

[Beginner] Best practice Backing Bean, DataBinding

Hi everybody,
I followed an oracle course in November 'Oracle Fusion Middleware 11g: Build Applications with ADF I'
Great, now I know what are the different parts of the framework.
But now I'd like to go a bit in depth.
Here is a simple example
Login page
Backing Bean
Session Bean
Read only View Object (select from view)
Application module
We have a username and password in a table but the password is encrypted in the column, in fact this is the checksum of the password.
Here is what should be done.
Login page (username, password) --> proceed button --> get data from VO --> compare usernames --> transform password coming from the login page (MD5) --> compare it to the database password -->
setting params to session bean --> redirecting to login2 page.
Here is what I have actually
I have a tuned AM with a java class having a method doLogin(String username, String password) with a String return value.
This method is exposed to the UI via a client interface.
Here are my questions.
Where do I have to check and transform all those params. In the Read Only VO via a tuned Java class extending ViewRowImpl or in the tuned AM java class ??
And now for the session bean where do I have to instanciate it. In the Backing Bean I suppose.
Wouldn't it be better to call the client interface from the backing bean, then instanciate the session bean, with params coming from the AM, in that BackBean.
I have so many question that I don't know where to begin. :-(
Kind Regards for your help.
Stessy Delcroix

Hi Frank,
thanks for your response.
Now where do I have to implement the logic behind the login.
I explain a bit the application.
We have coordinator and partner who can connect to the website.
If a coordinator didn't change the passwords for both coordinator and partner, the partner cannot login.
And a message will be displayed on the login page.
If the username or password is not correct, another message will be displayed.
Now how to return the params of the query ??
Do I have to implement the logic at the AM level or at the VO level ??
I mean calling findViewObject in the AM, play with the result of the query, implement the logic and return the params that will have to be set in a session bean
Here is the query we have
SELECT A.PASSWORD, A.TYPE, A.PROPOSAL_ID, R.CALL_ID, R.INSTRUMENT_ID, R.ONLINE_PREPARATION, A.STATUS, I.ANNEXES_UPLOAD, C.SINGLE_PARTNER, I.REFEREE_FUNCTIONALITY, A.REFEREE_ID, I.INSTRUMENT_CODE, C.TWO_PARTB_FUNCTIONALITY, C.STAGE_TYPE, EIL.SETUP_TYPE, C.CALL_CODE, I.MANDATORY_ANNEXES " +
" FROM EPSS_CREDENTIALS_VIEW A, EPSS_REGISTRATION R, EPSS_INSTRUMENT I, EPSS_CALL C, EPSS_INSTRUMENT_LIST EIL" +
" WHERE (USERNAME = ? AND A. PROPOSAL_ID = R. PROPOSAL_ID AND R.INSTRUMENT_ID = I.INSTRUMENT_ID AND C.CALL_ID = R.CALL_ID AND I.INSTRUMENT_CODE = EIL.INSTRUMENT)
All the values except Password will be set in a session bean.
I know that question is a bit annoying, but I don't want to start implementing things in the wrong way. And having dependencies between layers.
That's strange that no best practices examples have been published yet.
Thanks a lot for your precious help.
Regards,
Stessy
Edited by: Stessy.Delcroix on Mar 10, 2011 3:51 PM

Best practice for dealing with Recordsets

Hi all,
I'm wondering what is best practice for dealing with data retrieved via JDBC as Recordsets without involving third part products such as Hibernate etc. I've been told to NOT use RecordSets throughout in my applications since they are taking up resources and are expensive. I'm wondering which collection type is best to convert RecordSets into. The apps I'm building are webbased using JSPs as presentation layer, beans and servlets.
Many thanks
Erik

There is no requirement that DAO's have a direct mapping to Database Tables. One of the advantages of the DAO pattern is that the business layer isn't directly aware of the persistence layer. If the joined data is used in the business code as if it were an unnormalized table, then you might want to provide a DAO for the joined data. If the joined data provides a subsiduray object within some particular object, you might add the access method to the DAO for the outer object.
eg:
In a user permissioning system where:
1 user has many userRoles
1 role has many userRoles
1 role has many rolePermissions
1 permission has many rolePermissions
ie. there is a many to many relationship between users and roles, and between roles and permissions.
The administrator needs to be able to add and delete permissions for roles and roles for users, so the crud for the rolePermissions table is probably most useful in the RoleDAO, and the crud for the userRoles table in the UserDAO. DOA's also can call each other.
During operation the system needs to be able to get all permissions for a user at login, so the UserDAO should provide a readPermissions method that does a rather complex join across the user, userRole, rolePermission and permission tables..
Note that f the system I just described were done with LDAP, a Hierarchical database or an Object database, the userRoles and rolePermissions tables wouldn't even exist, these are RDBMS artifacts since relational databases don't understand many to many relationships. This is good reason to avoid providing DAO's that give access to those tables.

Slow starup of Java application - best practices for fine tuning JVM?

We are having problems with a java application, which takes a long time to startup.
In order to understand our question we better start with some background info. You will find the question(s) after that.
Background:
The setup is as follows:
In a client-server solution we have a win xp, fat client running java 1.6.0.18.
(Sun JRE). The fat client containt a lot of GUI, and connects to a server for DB access. Client machines are typical 1 to 3 years old (there are problems even on brand new machines). They have the client version of JRE - standard edition installed (Java SE 6 update 10 or better) Pretty much usual stuff so far.
We have done a lot of profiling on the client code, and yes we have found parts of our own Java code that needs improving. we are all over this. Server side seems ok with good response times. So far, we havent found anything about shaky net connections or endless loops in the java client code or similiar.
Still, things are not good. Starting the application takes a long time. too long.
There are many complicating factors, but here is what we think we have observed:
There is a problem with cold vs. varm starts of the application. Apparently, after a reboot of the client PC - things are really, really bad - and it takes (sometimes) up to 30-40 secs to start the application (until we arrive at the start GUI in our app).
If we run our application, close it down, and then restart
without rebooting, things are a lot better. It then usually takes
something like 15 - 20 sec. which is "acceptable". Not good, but acceptable,
Any ideas why?
I have googled it, and some links seems to suggest that the reason could be disk cache. Where vital jar are already in disk cache on th warm start? Does that make any sense? Virus scanners presumable runs in both cases.
People still think that 15 - 20 sec in start up on the warm start is an awful long time, even though there is a lot, a lot, of functionality in the application.
We got a suggestion to use IBMs JRE - as it can do some tricks (not sure what) our SUN JRE cant do concerning the warm and cold start problem. But thats is not an option for us. And noone has come up with any really good suggestions with the SUN JRE so far?
On the Java Quick Starter (JQS) -
improves initial startup time for most java applets and applications.
Which might be helpful? People on the internet seem more interested
in uninstalling the thing than actually installing it though?
And it seems very proprietary, where we cant give our Jar files to it?
We could obviously try to "hide" the problem in some way and make it "seem" quicker. Where perceived performance can be just as good as actual performance. But it does seem a bad solution. So for the cold start we will probably try reading the jar files and thereby have them in disk cache before startup of our application. And see if that helps us.
Still, ok the cold start is the real killer, but warm start isn't exactly wonderfull either.
People have suggested that we read more on the JVM and performance.
java.sun.com.javase/technologies/performance.jsp
java.sun.com.docs/hotspot/gc5.0/gc_tuning_5.html
the use of JVM flags "-Xms" "-Xmx" etc etc.
And here comes the question .. da da ...
Concerning various suggested reading material.
it is very much appreciated - but we will like to ask people here - if it is possibe to get more specific pointers. to where the gold might be buried.
I.e. in a an ideal world we would have time to read and understand all of these documents in depth. However, in this less than ideal world we are also doing a lot of very timeconsuming profiling in our own java code.
E.g. java garbage collection is is a huge subject - and JVm settings also. Sure, in the end we will probably have to do this all very thoroughly. But for now we are hoping for some heuristics on what other people are doing when facing a problem like ours..?
Young generation, large memory pages, garbage collection threads ect. all sounds interesting - but what would you start with?
If you don't have info to decide - what kind of profiling would you be running and then adjust what JVM setting in your trials?
In this pressed for time scenario. Ignorance is not bliss. But makes it hard to pinpoint the or those JVM parameters to adjust. So some good pointers from experienced JVM "configurators" will be much appreciated!
Actually, If we can establish that finetuning of these parameters is a good idea, it will certainly also be much easier to allocate the time for doing so. - reading, experimenting etc. in our project.
So, All in all , what kinds of performance improvements can we hope for? 5 out of 20 secs on the warm start? Or is it 10 % nitpicking? Whats the ball park figure for what we can hope to achieve here given our setup? What do you think based on above?
Maybe someone out there have done some finetuning of JVM parameters in a similiar PC environments like, with similiar fat clients...? Finetuning so and so - gave 5 secs. So start your work with these one-two parameters?
Something like that - some best practices? Thats what we are hoping for.
best wishes
-Simon

Thanks for helpful answer from both you and kajbj.
The app doesn't use shared network drives.
What are you doing between main starts to get executed and the UI is
displayed?
Basicly, Calculating what to show in the UI. Accessing server - not so much, there are some reads from a cache, but the profiling doesnt indicate that it should be a problem. Sure, I could shift the startup time to some other slot, but sofar I havent found a place where the end-user wouldnt be annoyed.> Caching of something would seem most obvious. Normal VM stuff >seems unlikely. With profiling i basicly find that ''everything'' takes a lot longer in the cold start scenario. Some of our local java methods are going to be rewritten following our review. But what else can be tuned?You guys dont think the Java Quick Start approach, with more jars in disk cache will give something? And how should that be done/ what does people do?I.e. For the class loader I read something about
1.Bootstrap class loader
2.Extensions class loader
3.System class loader
and is wondering if this has something to do with the cold start problem?
The extensions class loader loads the code in the extensions directories (<JAVA_HOME>/lib/ext
So, we should move app classes to ext? Put them in one jar file? (We have many). Best practice about that?
Otherwise it seems to me that it must be about finetuning the JVM?
I imagine that it is a question about:
1. the right heap size
2. the right garbage collection scheme
Googling heap size for XP
CHE22 writes:
You are right; -Xms1600M works well, but -Xms1700M bombs
Thats one best practice or what?
On garbage collection, there are numerous posts, and much "masters of Java black art" IMHO, And according to profiling GC is not really that much of a problem anyway? Still,
Based on my description I was hoping for a short reply like "try setting these two parameters on your xp box, it worked for me" ...or something like that. With no takers on that one, I fear people are saying that there is nothing to be gained there?
we read:
[ -Xmx3800m -Xms3800m
Configures a large Java heap to take advantage of the large memory system.
-Xmn2g
Configures a large heap for the young generation (which can be collected in parallel), again taking advantage of the large memory system. It helps prevent short lived objects from being prematurely promoted to the old generation, where garbage collection is more expensive.
Unless you have problems with pauses, try granting as much memory as possible to the virtual machine. The default size (64MB) is often too small.
Setting -Xms and -Xmx to the same value increases predictability by removing the most important sizing decision from the virtual machine. On the other hand, the virtual machine can't compensate if you make a poor choice.
The -XX:+AggressiveHeap+ option inspects the machine resources (size of memory and number of processors) and attempts to set various parameters to be optimal for long-running, memory allocation-intensive jobs]
So is Setting -Xms and -Xmx and -XX:AggressiveHeap
best practice? What kind of performance improvement should we expect?
Concerning JIT:
I read this one
[the impact of the JIT compiler is obvious on the graph: at startup the time taken is around 500us for the first few values, then quickly drops to 130us, before falling again to 70us, where it stays for 30 minutes,
for this specific issue, I greatly improved my performances by configuring another VM argument: I set -XX:CompileThreshold=50]
The size of the cache can be changed with
-Xmaxjitcodesize
This sounds like you should do something with JIT args, but reading
// We disable the JIT during toolkit initialization. This
// tends to touch lots of classes that aren't needed again
// later and therefore JITing is counter-productiive.
java.lang.Compiler.disable();
However, finding
the sweet spots for compilation thresholds has been tricky, so we're
still experimenting with the recompilation policy. Work on it
continues.
sounds like there is no such straigth forward path, it all depends...
Ok, its good, when
[Small methods that can be more easily analyzed, optimized, and inlined where necessary (and not inlined where not necessary). Clearly delineated uses of data so that usage patterns and lifetimes are apparent. ]
but when I read this:
[The virtual machine is responsible for byte code execution, storage allocation, thread synchronization, etc. Running with the virtual machine are native code libraries that handle input and output through the operating system, especially graphics operations through the window system. Programs that spend significant portions of their time in those native code libraries will not see their performance on HotSpot improved as much as programs that spend most of their time executing byte codes.]
I have the feeling that we might not able to improve performance that way?
Any comments?
otherwise i was wondering about
-XX:CompileThreshold=50 -Xmaxjitcodesize (large, how large?)
Somehow, we still feel that someone out there should have experienced similiar problems? But obviously there is no guarantee that the someone should surf by here!
In c++ we used to just write everything ourselves. Here it does seem to be a question about the right use of other peoples stuff?
Where you are kind of hoping for a shortcut, so you dont have to read endless number of documents, but can find a short document that actually addresses your problem ... well.
-Simon
Edited by: simoncpm on Mar 15, 2010 3:43 PM
Edited by: simoncpm on Mar 15, 2010 3:53 PM

Best Practice On Taking Count in Bean

Similar Messages

Maybe you are looking for