Pros/Cons of separating servlet/EJB clusters

          What and the pros and cons of establishing a cluster for the servlet layer and
          a separate cluster for the EJB layer?
          The rational behind this decision would be to take advantage of a shared infrastructure
          where multiple EJB components could be deployed in a cluster where they are accessed
          from multiple servlet/clusters.
          The obvious con is the fact that all calls between the servlet layer and the EJB
          layer will be remote calls however, there should be a greater level of availability
          and scalability.
          What are the other issues behind this deployment strategy?

I don't recall any other problems with this topology other than the one
          that you already mentioned. Infact i have seen several customer using
          this topology. i.e. having WebTier Cluste and EJB tier cluster.
          Kumar
          JD wrote:
          > What and the pros and cons of establishing a cluster for the servlet layer and
          > a separate cluster for the EJB layer?
          >
          > The rational behind this decision would be to take advantage of a shared infrastructure
          > where multiple EJB components could be deployed in a cluster where they are accessed
          > from multiple servlet/clusters.
          >
          > The obvious con is the fact that all calls between the servlet layer and the EJB
          > layer will be remote calls however, there should be a greater level of availability
          > and scalability.
          >
          > What are the other issues behind this deployment strategy?
          >

Similar Messages

Using Business area for External Reporting-Pros & Cons?

Dear Freinds,
We are a Retail Business Group half way through SAP IS Retail implementation. We have 8 legal entities, currently their complete Annual Financials are being made & audited separately. In SAP the structure is that there is Single Co. Code & transactions will be seggregated by Business Area (made manadatory for data entry) and Annual Accounts will be drawn up using Business Area.
This to me does not seem to be the right structure. I have joined the group recently, I will be much thankful to you all if you please help me list the Pros & Cons of using Business Area for external reporting.
Rajesh Agarwal
[email protected]
Edited by: Rajesh Agarwal on Jan 27, 2008 6:54 PM

Dan: For management reporting you may not want to post eliminations between companies where the business areas are the same.
If the Business wants to see the eliminations between companies, within the same business area, as part of BEx reports, what should I do? They do not want to execute consolidation functions in UCMON but just want to see how the eliminations have taken place between companies with in the same business area.
Dan: As long as the partner company and partner business area are consistently included for the intercompany transactions, the eliminations between Legal and Management will agree at the top-level of each hierarchy.
Have I to introduce one more InfoObject Partner Business Area in the data basis and assign the role Sub-assignment once again?
Thanks in advance.

Pros & Cons of FileVault

I'm wondering what the group thinks about FileVault? I haven't used it. I have two files on my computer that, if it were ever lost or stolen, would put information out there that I would rather keep to myself.
An overall Pro/Con discussion would be great.
Thanks.
Chip

Hi, Chip.
You're wise to ask.
I recommend that one only implement FileVault if you really need that level of security:• You have a laptop and travel with it.
• You share a Mac with others or your Mac is in an office easily accessible by others and you have personal, confidential information on such that must be protected across your Home folder.In lieu of FileVault, you can always protect data selectively using an encrypted disk image.
The key risk with FileVault is that you put "all your eggs" in one encrypted disk image. If a bad sector or block develops on the hard drive where the encrypted Home folder resides, your data is toast. Having a comprehensive Backup and Recovery solution, and using it regularly, is critical for all users, and especially important if you use FileVault. Most backup and recovery utilities require that you backup a FileVault-protected Home folder separately from the rest of your hard drive. If you use FileVault, I recommend you backup your encrypted Home folder daily.
Some additional advice should you elect to use FileVault:• Before implementing FileVault:• Check your disk thoroughly. At a minimum, run Steps 1-3 of the procedure in my Resolving Disk, Permission, and Cache Corruption FAQ.
• You must have sufficient free space on your hard disk in order to enable FileVault. You need free disk space at least a bit larger than your Home folder in order for the encryption process to take place. That process copies your current Home folder into an encrypted, sparse disk image file (,sparseimage) — hence the need for the free space — and then securely erases your original, unencrypted Home folder when the copy has completed successfully.• Do not lose the Master Password for your Mac. This must be set in order to enable FileVault and is essential to recovering any encrypted Home folders on your Mac should a user forget their account password. The Master password should again be a "good" password: hard to guess or discern from publicly-available information about you.
• To avoid performance problems when using iMovie and either of Final Cut Pro or Express, save project-related files, scratch disks, and other linked media outside your Home folder. See:• “iMovie: Using FileVault Can Affect Performance.”
•“Final Cut Pro and Final Cut Express: About Using FileVault.”
• Repairing your encrypted Home folder requires special instructions.
• To disable FileVault, you again need free disk space greater than or equal to the size of your encrypted Home folder. In disabling FileVault, your Mac first copies the contents of your Home folder from the encrypted Sparse image to a new Home folder — hence why so much free space is required — and then securely erases the encrypted Home folder.Note that some of the information above is from the "Security" chapter of my book, Troubleshooting Mac® OS X, Tiger Edition.
Dr. Smoke
Author: Troubleshooting Mac® OS X
Note: The information provided in the link(s) above is freely available. However, because I own The X Lab™, a commercial Web site to which some of these links point, the Apple Discussions Terms of Use require I include the following disclosure statement with this post:
I may receive some form of compensation, financial or otherwise, from my recommendation or link.

Trying to build a web application with stateful ejb clustering

Hi to everyone,
I'd like to know if someone gets to work a stateful ejb clustering. The stateful ejbs are called from a servlet so i'm not using RMI.
The ejb clustering sample works! but It is using RMI and this is not use for me.
My application already works but not in a cluster. Is this possible? I supose yes AND HOW???
Thanks in advanced.

Might look at: https://jsfportletbridge.dev.java.net to get started.

Hola tengo una mac pro con pantalla de retina de 13 pulgadas y le quise conectar un proyector hacer serie c120 pero dise que no lo puede leer y el proyector lo provee con otras laos y si funciona me podrían ayudar porfavor'

hola tengo una mac pro con pantalla de retina de 13 pulgadas y le quise conectar un proyector hacer serie c120 pero dise que no lo puede leer y el proyector lo provee con otras laos y si funciona me podrían ayudar porfavor?

Yes in an STL file the subs have a comma after each In and Out Timecode (as a separator) and also they don't normally have line spaces between each entry. If you want the subs to be on two lines and "force" what goes on to the second line (rather than SP deciding for you) then where you want the return to be you need to add "bar" symbol (the one above the backslash on a Mac keyboard)
Try it on a couple of lines of your subtitles to see if it works before committing. I usually create my subs as STL files but it should work
Steve

Pro/cons to have indexes on its own tablspace

hi ,
i have a new db running on oracle 10G R2 , various table spaces have been created that can be extended from 8M , 2M , 128K
my dba recommends that large insertion/updates tables goes to the 8M tablespace , but i notice that the indexes of those tables are in the same tablespace
ques :
1. can index exists on its own tablespace
2 . are there pro/cons of having indexes in its own tablspace ?
tks & rgds

Not really the proper forum for this question. You should have rather posted the question to the [url http://forums.oracle.com/forums/forum.jspa?forumID=61]Database Forum
> 1. can index exists on its own tablespace
Yes.
> 2 . are there pro/cons of having indexes in its own tablspace ?
Maybe.. it depends entirely on what your requirements are, and the nature of the data and hardware.
One reason to place indexes in a separate tablespace is simply for space management. E.g. this prevents some idiot developer of eating up all the space for tables, by indexing each and every column in a table and swamping the tablespaces with indexes.
Another reason is block sizes. Tablespaces can each have a different block sizes. You may want to use large 128KB blocks for data and smaller 32KB blocks for indexes.
That all said - personally I tend these days to prefer a single tablespace per logical database/application. In the past (especially back in Oracle 7) I always separated indexes and tables. But Oracle space management has improved so much, and we're dealing with such large data volumes these days, I do not think it is warranted without some sound technical reason to split data and indexes.

Pros & Cons of Using SAP PI Interfaces for Report Generation

Hi Guru's
I have a Scenario's like
I have to generate a customized report in SAP with the main data's available in SAP ECC and some required data available in the Legacy System.
I want to know the Pros & cons of using SAP PI RFC/Proxy adapter interface to get the data from the legacy system for each time the user execute the report in SAP ECC.
Thanks in Advance

There are couple of "dimensions" to consider in your PI interface design. For exemple when you are running the sizing exercise (Since we are considering adding a net new interface), you will need to capture specific information about new interface.i.e. S/A, adapters, frequency, avg payload size. etc..Note that the last two attributes will be hard to size properly in this case since you can't predict how frequent the end user will run the report which will impact the latency time required to pull the data. Latency will affect the user experience as visible side affect and definitely the SLA for other interfaces running at the same time.
On the other hand, the data you are trying to retrieve from the legacy won't be used for transactional purpose but for the end user to pull KPIs from the system which can affect ECC as well. You may end up doing lot hot fixes for your report (Assuming that the report is a medium complexity code)
There are other factors to consider but let's consider these are the major one.
Cheers,
F

How can I use firefox with pro con latte extention on a linux OS that runs via a USB?

What linux distro to use?
What version of firefox to use? (Since it has to be compatible with pro con latte)

Firefox tarballs from Mozilla Org use prefs within Firefox for updates. The "toggle" for not automatically updating - '''app.update.enabled''' and where to look for updates - '''app.update.url'''. The first needs to be '''''false''''' and the second could be a '''''null''''' string, do an update can't be found.
''I use both prefs to keep Firefox from updating by itself. I install each new version of Firefox as it is released, and keep the older versions around for reference purposes. Those prefs are the same since Firefox 1.5 came out.''
The versions of Firefox that are part of an operating system installation usually are that OS's own build of Firefox, and don't use the same update setup as Mozilla uses, in most cases. I am not familiar with Cent OS, but Ubuntu and Debian look for Firefox updates and from their own program repository servers outside of Firefox - as with the tarballs from Mozilla.

How can you get the filter off if you can't remember the password? It's"pro con" and it wont even let me check out the rates for aflight.

How can you get the filter off if you can't remember the password? It's"pro con" and it wont even let me check out the rates for aflight.

You can check the file prefs.js in the [http://kb.mozillazine.org/Profile_folder_-_Firefox Firefox Profile Folder] and remove the line(s) related to that extension (procon.password).

Is there anything available that covers the pros/cons of a simple and network PCD migration?

Is there anything available that covers the pros/cons of a simple and network PCD migration?

Hi,
as such there is no DOC but I can share my views:
Under Pros:
a. PCD is an application use to migrate older versions[6.1.5, 7.x , 8.x] running on MCS to VM servers.
b.It can be used to change the IP address or hostname on new servers without affecting existing version .
c. Many users using MCS Hardware are not willing to migrate to latest version because of existing HW not supporting interim releases which they have to hop through to get on to the target release.
d. Further, this requires license re-hosting as well for interim releases and major downtimes as well which can be avoided by using PCD without disturbing existing setup.
e. Also, DRAM/HDD sizing needs to be considered while doing migration on MCS.PCD addresses these issues for migration.
Under CONS:
a. It needs to be ordered through Product Upgrade Tool and bootable is not downloadable from Cisco site.
b. You could face some issues while doing migration which could be bugs but can be resolved.
c.could face issues while discovering cluster/doing migration but can be resolved.
regds,
aman

Servlets/JDBC vs. servlets/EJB performance comparison/benchmark

I have a PHB who believes that EJB has no ___performance___ benefit
against straightforward servlets/JSP/JDBC. Personally, I believe that
using EJB is more scalable instead of using servlets/JDBC with
connection pooling.
However, I am at a lost on how to prove it. There is all the theory, but
I would appreciate it if anyone has benchmarks or comparison of
servlets/JSP/JDBC and servlets/JSP/EJB performance, assuming that they
were tasked to do the same thing ( e.g. performance the same SQL
statement, on the same set of tables, etc. ).
Or some guide on how to setup such a benchmark and prove it internally.
In other words, the PHB needs numbers, showing performance and
scalability. In particular, I would like this to be in WLS 6.0.
Any help appreciated.

First off, whether you use servlets + JDBC or servlets + EJB, you'll
most likely spend much of your time in the database.
I would strongly suggest that you avoid the servlets + JDBC
architecture. If you want to do straight JDBC code, then it's
preferable to use a stateless session EJB between the presentation layer
and the persistence layer.
So, you should definitely consider an architecture where you have:
servlets/jsp --> stateless session ejb --> JDBC code
Your servlet / jsp layer handles presentation.
The stateless session EJB layer abstracts the persistence layer and
handles transaction demarcation.
Modularity is important here. There's no reason that your presentation
layer should be concerned with your persistence logic. Your application
might be re-used or later enhanced with an Entity EJB, or JCA Connector,
or a JMS queue providing the persistence layer.
Also, you will usually have web or graphic designers who are modifying
the web pages. Generally, they should not be exposed to transactions
and jdbc code.
We optimize the RMI calls so they are just local method calls. The
stateless session ejb instances are pooled. You won't see much if any
performance overhead.
-- Rob
jms wrote:
>
I have a PHB who believes that EJB has no ___performance___ benefit
against straightforward servlets/JSP/JDBC. Personally, I believe that
using EJB is more scalable instead of using servlets/JDBC with
connection pooling.
However, I am at a lost on how to prove it. There is all the theory, but
I would appreciate it if anyone has benchmarks or comparison of
servlets/JSP/JDBC and servlets/JSP/EJB performance, assuming that they
were tasked to do the same thing ( e.g. performance the same SQL
statement, on the same set of tables, etc. ).
Or some guide on how to setup such a benchmark and prove it internally.
In other words, the PHB needs numbers, showing performance and
scalability. In particular, I would like this to be in WLS 6.0.
Any help appreciated.--
Coming Soon: Building J2EE Applications & BEA WebLogic Server
by Michael Girdley, Rob Woollen, and Sandra Emerson
http://learnweblogic.com

What are pros/cons of using xmarks vs. firefox sync?

Used Firefox for a long while. Bookmarks morphed to Xmarks, which in turn has been acquired twice. I continue to use the newer Xmarks. In process of setting up a new computer, found that Firefox has it's own bookmarks sync.
Like to keep things as simple as possible. Does Firefox sync do all that Xmarks does? What are pros/cons of using xmarks vs. firefox sync?

I am not in a business environment, just my home. So, I don't know your specific requirements. I operate a dual G5 xserve, a dual quad-core mac pro, 6+ Apple MacOSX computers of varying types and several other linux servers plus a couple of Windows machines. If it were me, I would get two mac pro's running MacOSX Server and use one for backup of the other. The xserve RAID is too expensive from my point of view. I copy data from the xserve to several MacOSX computers with firewire drives for my backup and I have every computer on a ups. (I am assuming that they will not all fail at once!
I run Tiger servers and would suggest that you significantly test Leopard before committing to it.
Hope that helps.

Best practice for Servlet EJB integration

 I'm wondering what the best practice is for Servlet EJB integration in terms of
 caching the home and remote objects. My understanding is that the Home object
 is threadsafe and could therefore be cached as an attribute of the Servlet. This
 would remove the need for a JNDI lookup for each request. Similarly caching the
 ProxyObject would yield further savings. However, I have noticed that most examples
 don't use either of these practices. Why not?
 Thanks in advance,
 Geordie


This has been answered repeatedly. WL allows you to cache JNDI context
 objects, ejb homes and remotes without any problems. (EJB remote interfaces
 must only be used by one thread at a time, but that requirement is provided
 by the EJB spec itself.)
 Peace,
 Cameron Purdy
 Tangosol, Inc.
 http://www.tangosol.com
 +1.617.623.5782
 WebLogic Consulting Available
 "Geordie" <[email protected]> wrote in message
 news:3af9579f$[email protected]..
 >
 > I'm wondering what the best practice is for Servlet EJB integration in
 terms of
 > caching the home and remote objects. My understanding is that the Home
 object
 > is threadsafe and could therefore be cached as an attribute of the
 Servlet. This
 > would remove the need for a JNDI lookup for each request. Similarly
 caching the
 > ProxyObject would yield further savings. However, I have noticed that
 most examples
 > don't use either of these practices. Why not?
 >
 > Thanks in advance,
 > Geordie

Pros / Cons of Running APEX on separate server

Seeking feedback/thoughts on this general question.
I am maintaining an APEX application that utilizes multiple schemas all residing in the same database as APEX. I am wondering what might be the pros / cons of putting APEX in a separate database and server.
Would this necessarily yield security or performance benefits (or costs)?
How many changes would need to be made in the APEX configuration and code? I have read here in the forum that at a minimum database links would need to be created to the utilized schemas on the other server, with appropriate grants. What else?
Thanks.

Please help us by putting your first name into your profile. Thanks.
So what you want to do is have your application (and Application Express) access tables in a remote database. Yes, you can do that and database links are required.
This almost always has negative performance impact and sometimes results in things not working due to database version differences, especially with down-level Oracle versions.
It also makes application development more difficult if the Application Builder doesn't have local access to application parsing schemas. You can work around most of these problems by creating local views of remote tables, but I don't think 100% of the wizards and some SQL Workshop functionality are amenable to remote schema access. So I don't recommend it. What is your motivation to do this?
By the way, "servers" don't matter in this discussion, it's all about databases.
Scott

Pros & Cons for consuming web services in ABAP using ABAP PROXY

Hi,
Other then performence is there any other disadvantages like security,etc for consuming web services in ABAP using ABAP proxy?
I really appreciate if some one provide the more details(Pros & Cons ) regarding cosuming web services and I also want to know is there any other way to consume web services in ABAP.
Thanks.

 is there any other way to consume web services in ABAP
you can use cl_http_client class to make your program to act as http client and post the soap message too webservice. This way you dont need to generate proxy, but you should know the soap message format.
Regards
Raja

Pros/Cons of separating servlet/EJB clusters

Similar Messages

Maybe you are looking for