Monday, April 23, 2012

Increasing the performance of EM console in Oracle SOA Suite 11g - Part1

Increasing the performance of EM console in Oracle SOA Suite 11g - Part1

EM console is very slow when accessing to view the composite/component details or the instance details.
We can try to improve the performance by little by following the below steps.

1. Login to the Oracle SOA Suite 11g EM Console
2. Right Click on SOA-INFRA and go to SOA Administration->Common properties
3. Set the Data Display Options as shown below.
              Select the option “Disable fetching of instance and fault count metrics. Each metric can still be  retrieved on demand”.
Set the Duration Details to 1 hour (based on your requirement) as shown below.


This will enable to fetch the instance and fault count metrics based on the demand also the default search criteria will display the last one hour data, this will improve the performance of the em console.

Enable fmwctl discovery cache:

Logging into Enterprise Manager Fusion Middleware Control 11g (fmwctl) takes a long time.  Sometimes as long as 45-60 seconds.  This is often viewed as slow response time, performance or hanging issues, fmwctl discovery is always performed as part of login.  

For installations with thousands of targets, fmwctl discovery may take 45-60 seconds. This delay is expected because EM discovery mbeans need to be invoked for every target.
Solution is to cache the discovery results in the servlet context and use it for subsequent logins. This discovery result will be shared by all the fmwctl users. This will still require the entire discovery to be done at least once.

Follow the metalink note 1423893.1 to enable the discovery caching.

If the caching is enabled, fmwctl will use the cached discovery results and login to the em console will be fast.The default setting is "not use the cached results" 



Thursday, April 19, 2012

Audit Trail configurations in Oracle SOA Suite 11g


Audit Trail configurations in Oracle SOA Suite 11g:

Until 11.1.1.3,Oracle BPEL audit trails are saved to database in the same JTA transaction as the main transaction. The below are the some of the issues caused.
  • Latency of saving audit trail to database is included as part of the overall latency of the main business transaction.
  • When the main transaction is rolled back for whatever reason, the audit trail did not get saved, because audit trails are saved in the main transaction as well. Thus no trace of what had happened can be found on the BPEL Console (or EM Console in 11G).
Since Oracle SOA Suite 11.1.1.3, management of audit trail memory has been enhanced.

The properties to configure the audit trail:

auditStorePolicy

Use this flag to choose the audit store strategy
syncSingleWrite - would store audit data synchronously when it saves the cube instance, this is done as part of the cube instance transaction.
This is the default value. And the behavior is the same as in the 10.1.3.x version.
syncMultipleWrite - would store audit data synchronously using a separate local transaction.
By "synchronously" it means the BPEL service engine is saving the audit trail in the same thread as the main transaction. However, it is doing it in a "separate local transaction".

Because they are on the same thread, latency of saving audit trail is still included into overall latency of the main transaction.

However, because they are on separate transactions, you can configure BPEL engine (using AuditFlushByteThreshold and AuditFlushEventThreshold) to flush out the audit trail from memory to database periodically regardless how long the main transaction would take.
Moreover, having them on two separate transaction means the rollback of main transaction will NOT affect the audit trail transaction. That is, you will still see audit trail even if the main transaction rolls back.
async - would store the audit data asynchronously using an in-memory queue and pool of audit threads.
This is almost the same as "syncMultipleWrite", except that it is done not just in a separate transaction but also in a separate thread.

The pro is the audit trail latency is NOT directly included in the latency of main transaction (but because they still share the computing resources and database, the latency is still indirectly related).

The con is that because audit trails are being saved asynchronously, the audit trail may be out of sync from the main transaction (as the name 'async' implies).

AuditFlushByteThreshold and AuditFlushEventThreshold

When auditStorePolicy=syncMultipleWrite or auditStorePolicy=async, you use these two flags to control how often the engine should flush the audit events. These two properties do NOT apply to auditStorePolicy=syncSingleWrite.

auditFlushByteThreshold means after adding an event to the current batch, the engine would check if current batch byte size is greater than this value or not. if yes, then it would flush the current batch. The default value is 2048000 (byte), i.e. 2MB.

Similarly, auditFlushEventThreshold means this limit is reached; the engine would trigger the store call. The default value is 300 (event)

Both values need to be tuned based on the application and requirements.

AuditDetailThreshold

This is the maximum size (in bytes) an audit trail details string can be before it is stored separately from the audit trail. The default value is 50000 (byte). This is the same property as in 10.1.3.x.
AuditLevel.this is the same property in 10.1.3.x.

How to Configure

All the above properties can be configured via the SOA EM Console. The path is EM -> SOA ->SOA-INFR - > SOA Administration -> BPEL Properties -> More BPEL Configuration Properties


Tuesday, April 17, 2012

Persistence Properties in Oracle SOA Suite 11g

Persistence Properties in Oracle SOA Suite 11g:

Oracle BPEL Persistence properties are used to control, when and how a process need to be dehydrated.

inMemoryOptimization

This property indicates to Oracle BPEL Server that this process is a transient process and dehydration of the instance is not required. When set to true, Oracle BPEL Server keeps the instances of this process in memory only during the course of execution. This property can only be set to true for transient processes (process type does not incur any intermediate dehydration points during execution).
  • false (default): instances are persisted completely and recorded in the dehydration store database for a synchronous BPEL process.
  • true: Oracle BPEL Process Manager keeps instances in memory only.

completionPersistPolicy

This property controls if and when to persist instances. If an instance is not saved, it does not appear in Oracle BPEL Console. This property is applicable to transient BPEL processes (process type does not incur any intermediate dehydration points during execution).
This property is only used when inMemoryOptimization is set to true.
This parameter strongly impacts the amount of data stored in the database (in particular, the cube_instance, cube_scope, and work_item tables). It can also impact throughput.
  • on (default): The completed instance is saved normally.
  • deferred: The completed instance is saved, but with a different thread and in another transaction, If a server fails, some instances may not be saved.
  • faulted: Only the faulted instances are saved.
  • off: No instances of this process are saved.
<component name="mySampleBPELComponent">
...
<property name="bpel.config.completionPersistPolicy">faulted</property>
<property name="bpel.config.inMemoryOptimization">true</property>
...
</component>


oneWayDeliveryPolicy

The oneWayDeliveryPolicy is from the Oracle 10g configuration property deliveryPersistencePolicy.The new configuration property name is bpel.config.oneWayDeliveryPolicy.
This property controls database persistence of messages entering Oracle BPEL Server. Its used when we need to have a sync-type call based on a one way operation. This is mainly used when we need to make an adapter synchronous to the BPEL Process.
By default, incoming requests are saved in the following delivery service database tables: dlv_message
  • async.persist: Messages are persisted in the database.
  • sync.cache: Messages are stored in memory.
  • sync: Direct invocation occurs on the same thread.

<component name="mySampleBPELProcess">
...
<property name="bpel.config.transaction" >required</property>
<property name="bpel.config.oneWayDeliveryPolicy">sync</property>
...
</component>