Advantages of using a decentralized adapter

What are the advantages of using a decentralized adapter engine

Hi Narendra,
Every XI installation will install a default adapter engine and these are called as the Central / Local Adapter Engines. Most of the times this one Adapter Engine is more than sufficient.
At times, when the load on your server is going to be high , you can install a Decentral Adapter Engines and these are nothing but seprate instances of your Adapter Engine that can pick data and send it to your Integration Engine.
The need for a decentral Adapter Engine also arises when you have security restrictions, DMZ , etc in the landscape where one adapter engie is used for one particluar DMZ and another acts as the local adapter engine etc.
The main advantage is load balancing and reducing risk of "single point of failure" so if something is wrong with central system or network then atleast your message will persist for later processing.
Another practical advantage is related to maintenance, every decentral AE is independent of other so maintenance will affect small set of interfaces not others.
See this link
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/30f8bd90-0201-0010-dd9a-c8a7f52c47aa
*Reward points if this helps
Regards
Vani.

Similar Messages

Retrive SID of XI server from EJB used in decentral adapter engine

Hi Experts,
We have a requirement where we need to call an RFC from a EJB bean using load balancing. To do the JCO call, we need the SID of the XI system as an input parameter for the createConnection function.
The SID can be retrieved using System.getProperty("SAPSYSTEMNAME"), however, we need to know if this is called in a decentral adapter engine, whether the getProperty() function will return the SID of the XI server, when called from a bean used in the sender adapter of a decentral adapter engine.
Regards

> however, we need to know if this is called in a decentral adapter engine, whether the getProperty() function will return the SID of the XI server, when called from a bean used in the sender adapter of a decentral adapter engine.
The decentral adapter engine has its own SID. So you need another solution for retrieving the SID of XI server. Maybe a property file?
Regards
Stefan

Can someone clarify what are the advantages of using a EBS Adapter?

Hi,
I'm a little confused on what the EBS Adapter is all about. We have Integrated SOA Gateway(ISG) and SOA Suie 11g installed. I have a few questions:
Why would I need to use a EBS Adapter in my composite?
Why can't I use a web service adapter to call a service from ISG instead of using a EBS adapter?
Why can't or use a Database adpater(which I can run PL/SQL) instead of using a EBS adapater?
What is the security setup like with the EBS adapter?
Thanks! I can't really seem to find any good documentation on this anywhere.

You can find lots of information here.
http://docs.oracle.com/cd/E15523_01/integration.1111/e10231/intro.htm#BHCECCFE
Regarding the security, as this falls under JCA adapter, it will carry all the security that JCA can support.
With the help of EBusiness adapter you talk to concurrent program, business events and much more. Database adapter wont let you to talk with businessevent and concurrent programs.
It is considered good etiquette to reward answerers with points (as "helpful" - 5 pts - or "correct" - 10pts).
Thanks,
Vijay

Decentralized Adapter Engine

Hello,
We have installed an XI landscape (XID, XIQ, XIP) on Unix along with a single decentralized Adapter Engine on Windows (XD1) which is currently configured for use with our XID system. We would like to also use this same decentralized Adapter Engine with our XIQ system. However, when we export a communication channel from XID that is configured to use the decentralized adapter engine (af.XID.calxid1) and subsequently import the communication channel into XIQ the decentralized Adapter
engine that was configured in XID (af.XD1.calxid1) is
replaced with "Integration Server" (Default) in the Communication Channel configuration in XIQ. The pull down to choose an adapter engine shows only the (Default) Integration server as an available choice. Currently XID and XIQ are using the same SLD and the decentralized adapter engine (af.XD1.calxid1) is identified there. I have been unable to determine how to use the decentralized adapter engine with our XIQ system. Does anyone know whether it is possible to have multiple XI systems using the same de-centralized Adapter Engine? And if it is possible how does one configure the second XI system (XIQ) to use the decentralized adapter engine?
Thanks.

Hello Dan,
a decentralized AE can only be attached to a single Integration Server.
Regards,
Thilo

Soap LookupService with Decentralized Adapter Engine

Hi Experts,
I would like to use a Decentralized adapter engine with the Lookup Service on PI 7.1 , but when try to use that, it doesn't work, because Lookup Service is default configure to use Central adapter only. I would like how to set a decentralize adapter.
How to do a Soap Lookup using a Decentral adapter?
Now, I use the following:
Channel channel = LookupService.getChannel("BC_SF_***","CC**");
SystemAccessor accessor = null;
accessor = LookupService.getSystemAccessor(channel);
Thanks in advance.
Fernando Pena.

You define in the communication channel, which adapter engine should be used, not in the mapping.

Decentralized Adapter Engine or Local Adapter Engine

Hi Experts,
I hv a doubt like what exactly we are achieving by installing a local Adapter engine? Is this something to handle the firewall protocols? In one of the SAP material, it has been mentioned that aditional adapter engines may need to be installed "Close" (in a network sense) to the Business System? What does it mean? What are the advantages over a centralized adapter engine?
regards

Vijaya Lakshmi,
Every XI installation will install a default adapter engine and these are called as the Central / Local Adapter Engines. Most of the times this one Adapter Engine is more than sufficient.
At times, when the load on your server is going to be high , you can install a Decentral Adapter Engines and these are nothing but seprate instances of your Adapter Engine that can pick data and send it to your Integration Engine.
The need for a decentral Adapter Engine also arises when you have security restrictions, DMZ , etc in the landscape where one adapter engie is used for one particluar DMZ and another acts as the local adapter engine etc.
Regards
Bhavesh

Sender File Adapter Not Showing in Decentral Adapter Engine

Hi,
We have recently implemented a Decentral Adapter Engine on a windows box. The adapter engine seems to be working fine because my adapters show and seem to work fine in it.
However, when I define a Sender File Adapter that uses File Content Conversion message protocol instead of just File messaeg protocol, the adapter does not show up in the decentral adapter engine.
I made sure that Adapter Engine parameter in the Sender File Adapter shows the new adapter engine but that does not seem to be working. If change the Adapter Enginer parameter to show the Central Adapter Engine, then I can see the Sender File Adapter on the Central Adapter engine and it works fine. (Note: The Central Adapter Engine is installed on a Unix box)
Here are the paramters I defined for the Sender File Adapter:
- Transport Protocol: File System (NFS)
- Message Protocol: File Content Conversion
- Adapter Engine: (new Adapter Name)
Source Directory: (
server/drive/path)
File Name: purchaseOrder.txt
Quality of Service: Exactly Once
Poll Interval (secs): 60000
Poll Interval (msecs):
Retry Interval (secs):
Processing Mode: Archive
Add Time Stamp (checked)
Archive Direcotry: (
server/drive/path/archive)
Process Read-Only files: (not checked)
Processing Sequence: By Name
Fiel Type: Text
File Encoding:
Operating System Command:
Additional Files (not checked)
Content Conversion Parameters
Document Name: MT_Request_File_Data
Document Namespace: http://nxy.tots/fg_admin
Document Offset:
RecordSet Name: Purchasing
RecordSet Namespace:
RecordSet Structure: Purchase_Record,*
RecordSet Sequence: Ascending
RecordSets per Message: 1
Key Field Name:
Key Field Type: String(Case-Insensitive)
Purchase_Record.fieldFixedLengths: 8,4,2,10,10,12,18,12
Purchase_Record.fieldFixedType: char
Purchase_Record.fieldNames: Request_Action,Request_PO_Num,Request_PO_LN,Request_PO_Date,Request_Purchasing_Org,Request_Purchasing_GName,Request_Created_By,Request_Vendor_Name
Purchase_Record.fileContentFormatting: trim
Adapter Status: Active
Advanced Mode: (not checked)
Your help is much appreciated.

Hi Alaa,
look into this document:
http://service.sap.com/nw-howtoguides -> Exchange Infrastructure -> Troubleshooting: Adapter Engine 3.0 (Expert Call slides)
If the File Adapter works without conversion, then you should count the number of fields in a row. If it differs from the configuration, then it runs on error.
Another idea is to check the content type of the file. When you do not enter a File Encoding, then the file adapter assumes utf-8. If the file is for example latin-1, then the number of characters in a row are not counted correctly.
Regards
Stefan

Advantages of using labview in embedded application when compared to c ?

Hello all,
I am looking to develop embedded application with help of labview programming
Started with Lm3s8962 evaluation board.
So i would like to know
1 Advantages of using labview in embedded application when compared to c ?
2 Can we deploy the code in any kind of controller by writting drivers for it ?

hello, the LM3S8962 microcontroller is a good sispositivos for developing applications of medium complexity, according to your questions:
1. if you work applications with embedded systems will save labview time in prototyping, we know that if you work with a company specific microcontroller, you must learn the language with which you are working Asm, Basic, C, then if you change hardware again start adapting your code in the libraries of the new compiler. Labview does not happen.
2. Labview only works with some microcontroller manufacturers and specific models.
Atom
Certified LabVIEW Associate Developer

Advantages of using PI 7.1 over PI 7.0

Hi All,
I need to suggest to my client, the advantages of using PI 7.1 over PI 7.0, so that we can use the same for our development.
Please dont send me the links from SDN, just suggest the advantages.
Useful answers will be rewarded.
Regards
Ranjit

Hi,
PI 7.1 will be in rampup then - not global availibility
when it will be available you will get an upgreade guide as always
Currently PI 7.0 is only available for real time implementation. You could have the version of PI 7.1 from SDN, but that not for the real time implementation. Its limited only for individual usage.
There is huge difference between PI 7.0 and PI 7.1.
1. PI 7.0 is the part of Netweaver 2004s and PI 7.1 is key entity in eSOA.
2. PI 7.1 have various mapping Enhancements as well as reusability of UDFs across the mappings etc. which is not available in PI 7.0
/people/william.li/blog/2008/01/02/sap-pi-71-mapping-enhancements-series-share-user-defined-functions
3. PI 7.1 comes with the concept of folders for more flexibale and Organized way of developements.
/people/william.li/blog/2007/08/07/using-folders-in-pi-71 -- Folders in PI 7.1
4. Advanced adapter engine is used in PI 7.1, which have overcome the various communication related limitations of previous versions
Refer
Upgrade to SAP NetWeaver Process Integration 7.1
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/8085e299-718c-2a10-de94-928f62b763ce
Features of PI 7.1
/people/udo.paltzer/blog/2007/04/26/new-sap-netweaver-process-integration-release-planned-for-2007
High Volume support in PI 7.1
/people/holger.faulhaber/blog/2007/12/12/high-volume-support-in-pi-71
Usability Features in SAP NetWeaver PI 7.1 Development and Configuration Times
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/a0e7734f-e969-2a10-24b6-df58a710941c
SAP NetWeaver Process Integration 7.1 - Overview of New Capabilities
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/706005a3-3bd6-2910-91ae-a2016239bdcf
Usability Features in SAP NetWeaver PI 7.1 Development and Configuration Times
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/a0e7734f-e969-2a10-24b6-df58a710941c
SAP Network Blog: Share User-Defined Functions in Message Mappings of PI 7.1
/people/william.li/blog/2008/01/02/sap-pi-71-mapping-enhancements-series-share-user-defined-functions
Preview on New Features of the Integration Directory in SAP NetWeaver Process Integration 7.1
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/10c0de4b-7876-2a10-e286-8412668643a8
SAP Network Blog: Mapping Enhancements in SAP NetWeaver Process Integration (PI) 7.1
/people/jin.shin/blog/2008/01/11/sap-pi-71-mapping-enhancements-series-mapping-enhancements-demo
New Business Process Engine Features in SAP NetWeaver Process Integration
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/e0068bc1-6f8c-2a10-52bb-c6ee3562feb2
Thanks
Swarup

HTTP UPLOAD using HTTP receiver adapter

Hello Experts,
I am facing a problem where i have to upload a file to webserver using POST , I am using a file adapter to pick the file and http or soap adapter to post the file to webserver. Here the request structure expected by the webserver is a file.
below is the form based web request:
<form enctype="multipart/form-data" name="Upload_File" method="POST" action="https://-------/servlet/zxcdsf">
Choose File: <input type=file name="Upload_File" value="fileUpload" ><br><br>
<input type=submit value="upload">
</form>
Please guide me how to achieve this using SOAP or HTTP.

Hi Hemant
Yes you can try with proglog and epilog
Develop the Java mapping to convert the paylaod to HTML tags and replace the tags with corresponding (&, +, =, and %) . This may be able to post the data.
Try to create generic mapping to convert the payload to HTML this will help if you have more than one interface for the same requirement.
Java proxy code is having an advantage that you can always have that as a receiver for any of these request instead of dealing with individual Java mappings.
Try with epilog and prolog and as you said next step can be Java proxy.
Thanks
Gaurav

Decentral Adapter not showing up in RWB

I have just installed a J2ee Decentral Adapter instance, configured the exchange profile to connect back to the main PI instance and connected it to the SLD (both the J2ee Decentral Adapter and main PI instance use the same SLD).
One of the last steps is to check see if the Decentral Adapter shows up in the PI Instances Runtime Work Bench (RWB), but it does not.
I've noticed that on the SLD, the J2ee Decentral Adapter shows up as a Technical System, but when going to the SLD > Administration > Content Maintenance > All Instances > Class: XI Adapter Framework, that only my main PI instances show up, not the recently installed J2ee Decentral Adapter.
Anybody have any ideas of what else to check ?
Thanks..SB

Aaron,
1.1 SAP XI AF CPA Cache are set at the defaults of the install. The SLDAccess is set to true which according to the following SAP Help doc says that the entris are ignored: http://help.sap.com/saphelp_nw04s/helpdata/en/29/22ee41c334c717e10000000a155106/frameset.htm
1.2 SLD DATA Supplier is set correctly. I can do a CIM Test and trigger the data upload successfully. Still no entry under the Content Maintenance though.
2. SXI_CACHE > Adapter Engine still does not show the Decentral Adapter. I've even bounced both systems and the SLD.
per Chirag I've also double checked blog "Basic Steps In Decentralized Adapter Engine
This Decentral adapter is in DMZ #1, so I must use the IP of my SLD for the SLD access and my Main PI is in DMZ #2. I've checked to make sure ports 50000 and 3300 from Decentral to my PI system work - are there any other ports ?
Thanks...Scott B.

Decentralized Adapter Engine working

Hi guys,
I have installed a decentralized adapter engine for my XI Server. I have a scenario which i run with default Adapter Engine(Integration Server), runs fine.
However, when i run it with choosing Decentralized AE it gives such errors in
Message Monitoring:
Exception caught by adapter framework: null
Delivery of the message to the application using connection AFW failed, due to: RecoverableException.
Also i am testing this for two simple file2file scenarios. It's acting differently; in a scenario sender commn channel picks file, in other not.
Is there any configuration settings that's remaining? Or something else?
Regards
Arpil

Hi,
For the picked message, i see this in moni....
<Trace level="1" type="T">--start sender interface action determination</Trace>
<Trace level="1" type="T">select interface MI_XmlFile_out*</Trace>
<Trace level="1" type="T">select interface namespace urn:Plain2XML</Trace>
<Trace level="1" type="T">no interface found</Trace>
<Trace level="1" type="T">--start receiver interface action determination</Trace>
<Trace level="1" type="T">Loop 0000000001</Trace>
<Trace level="1" type="T">select interface *</Trace>
<Trace level="1" type="T">select interface namespace</Trace>
<Trace level="1" type="T">no interface found</Trace>
<Trace level="1" type="T">--no sender or receiver interface definition found</Trace>
<Trace level="1" type="T">Hence set action to DEL</Trace>
- <Trace level="1" type="B" name="CL_XMS_MAIN-PERSIST_READ_MESSAGE">
Any clue?
Regards
Arpil

To drive a monitor from a Mini, is there an advantage to using the Thunderbolt port versus the HDMI port?

Is there an advantage to using the Thunderbolt port versus the HDMI port to drive a monitor?
Can i just plug in my 18" LCD TV to the HDMI port and use it as a monitor?
Thanks, friends ~

Thanks, lllaass,
I have a new Mini arriving tomorrow. It will be replacing my 9 year old iMac.
I have at my disposal an older VGA monitor and a regular LCD TV with
an HDMI input. I'm trying to decide which one will work the best for me. If I use the TV, I won't have to buy
a Thunderbolt to VGA adapter.

Decentral Adapter Engine

Currently documentation only indicates the ability to create a single decentralized adapter engine. (1) Is this in addition to the existing central adapter engine or do I have to choose which one I will use? (2) Does the decentralized adapter engine have to run on its own WAS? (3) If I am planning for a high volume of messages will the decentralized adapter engine perform load balancing for my integration engine? (4) Can I have more than one decentralized adapter engine? (5) If so, how do I configure these decentralized adapter engines to communicate to my integration engine? (6) In addition, how do I control message flow from the integration engine to 1 to many decentralized adapter engines? Feedback as well as any reference to additional information is much appreciated.

Hi Kirk,
(1): in addition
(2): no, jou need to install J2SE adapter engine
(3): yes
(4): as much as you like
(5): you configure like in XI 2.0 with configuration files. That is described at the XI documentation. For messages going to decentral adapter use channel XI with specified URL, for incoming messages you don't need a channel.
(6):1 to many: Just put your receivers to receiver determination.
Regard,
Udo

Decentral adapter engine sp14

Hi All,
Do we compulsory needed De central adapter engine and how to do this.
According my understanding If I use in my scenarios rfc/Idoc adpter i can use proxy server and central adapter.
Pls anyone clarify my assumption.
Venu.

Hi Venu,
no, you do not need a decentral adapter engine. Not for RFC and IDoc nor for any other technical adapter. The central adapter engine which will be installed with the integration server can hold all technical adapters (File, JDBC etc). Decentral adapter engines are only used, if there are special requirements such as network separation etc.
best regards
Christine

Advantages of using a decentralized adapter

Similar Messages

Maybe you are looking for