Infosys’ blog on industry solutions, trends, business process transformation and global implementation in Oracle.

« March 2015 | Main | July 2015 »

April 23, 2015

Accounting and Finance Shared Services- Oracle Fusion Capabilities for Record-to-Report and Analytics


Accounting and Finance Shared Services have typically been formed in large multinational organizations with the objectives of cost efficiency, standardization and operational effectiveness towards the customers within the organization. Some of the business processes handled by F&A SS are:

-          Procure to Pay

-          Order to Cash

-          Record to Report

-          Financial Planning, Consolidation, Budgeting and Forecasting

-          Treasury and Cash Management

-          Audit and Control

-          Analytics on the financial data

-          Financial closing and Financial statement preparation

Over the periods; instead of merely being a cost center, shared services are emerging as strategic business partners- offering innovation and value addition to various business functions. The extended objectives of the F&A Shared services include extensive analytics and reporting.

My previous blog had stated the Oracle Fusion capability for Procure-to-Pay cycle. It can be accessed here.

The General Accounting- Record to Report Cycle, Consolidation, Reporting and Analytics- requirements of a Multinational, Multi-GAAP organization; using best-of-breed or legacy applications for sub-ledger transactions; can be catered by Oracle Fusion Accounting Hub (FAH) and Fusion General Ledger from Fusion Financials suite. The Financial Reporting Center of FAH involves embedded multi-dimensional Essbase cubes and offers state-of-the-art reporting and analytics which enables multi-dimensional analysis of various transactional and accounting data.

FAH can integrate and utilize the capabilities of number of reporting tools like Oracle BI Publisher, Oracle Hyperion Financial Reporting Studio, Fusion Edition and Oracle Business Transactional Business Intelligence. Drill down up to journal and transaction level is possible from any of the reports generated out of FAH which enables strong audit trail and control.

To record the transactions from Oracle/Non-Oracle applications and convert it into accounting entries, FAH offers a highly configurable rule transformation engine that generates the accounting entries, validates and transforms it into accounting journals. The rules can be configured based on the attributes of the transaction. For example Tax Accounts could be determined based on the country of transaction.

Different rules can be configured for different divisions complying with the local accounting regulations. FAH also captures additional attributes of the transaction for reference and reporting purposes.

FAH_Fin and Acc SS1.JPG

Some of the salient features of Fusion reporting are:

-          Fixed format reports and local statutory reports can be created using Oracle BI Publisher and Financial Reporting Studio/Workspace.

-          Interactive financial reporting and drill down can be achieved either by Oracle BI Publisher, Oracle transactional Business Intelligence, Account Monitor and Account inspector

-          Seeded BI reports are delivered in OTBI

-          Live and interactive reporting with multiple output options like HTML, Excel, PDF; available in Smart View

-          Drill down of the report information upto detailed journal and sub-ledger transaction level using Account Inspector

-          Bursting option available with BI Publisher to split the report and email or print them as per output option selected

-          Functionality to embed charts and graphs in the report

-          Drag and Drop report grid and dimensions

FAH also offers features for Budgeting and Budget Reporting, Consolidation and Inter-company processing. The advantage of FAH for an Account and Finance Shared Service specifically lies in its fitness to be integrated with diverse sub-ledger applications and providing robust accounting platform along with impressive reporting and analytics solution.

References: Oracle Fusion Accounting Hub training content

April 1, 2015

Need and Ways for Performance Optimization in OBIEE


In all my years of working with OBIEE, I have seen a special interest in performance testing and tuning in the recent years. I have been involved in recent past with one of the active engagements for an OBIA system for performance optimization. This lead to a lot of interesting reading and some on the job application, thought should share it with you all.

While Optimizing Performance the most challenging task is to know where to begin. Most of the time Performance issues are raised or known only when we are actually in the performance test phase of the project. Post Analysis we should take the right approach for optimizing the performance and not some shortcut ways. If the analysis suggests that the rpd metadata / dashboard design/ built is the suspect or the underlying data model is not modelled properly optimization techniques are not going to work and redevelopment should be considered as it is rightly said

If a system that has a fundamental design flaw cannot be optimized - it has to be redesigned.

By default whenever OBIEE reports take a long time to return results, end users tend to say "OBIEE does not perform well". There might be different reasons for it other than OBIEE some of them might be the underlying database, network or infrastructure. We will discuss more about what we can do within OBIEE for performance optimization considering we have all other levers correctly placed.

OBIEE Reports creates a SQL query with the help of BI server and that is purely driven by how the data modelling is done is executed directly against the underlying database. A quick look at the SQL query getting generated can give us an idea whether best practices of OBIEE design are followed or not. By following the best practices in OBIEE design the SQL query generated can be made free of unnecessary clauses and nesting statements.

OBIEE itself has many out of box functionality for performance optimization such as usage of aggregate tables, using cache, changing the log levels and so on but are not enough to handle the performance issues.

The components involved in Performance Tuning are

·         Weblogic Server

·         BI Server

·         Presentation Server

·         Data Warehouse

With the help of monitoring tools like Enterprise Manager Metric Palette, Performance monitor, Server related statistics are collected and should be acted accordingly.

Would list some of the performance optimization techniques less known yet effective

·         Try Tuning the Connection backlog buffering and Statement Cache on Oracle Weblogic server.  Try adjusting the accept backlog value which might be set too low  due to which connections are dropped or refused at the client.

·         Oracle BI Server performance can be improved by prohibiting RPD updates. By doing this Oracle BI server is not needed to handle lock control.

·         Various configuration options can be used to limit the maximum number of rows that can be downloaded, included in mail or processed.

·         Limit the maximum number of rows that can be processed to render a table. This significantly improves performance as it reduces the system resources that will be consumed by user session.

·         Providing Permission to objects by groups/roles.

·         Union requests, consumes more memory as they generate more temp file , hence should be avoided.

·         Load to BI resources can be reduced by avoiding usage of guided navigation

·         Session variables use should be minimized.

·         Dimensional Hierarchy should be modelled accurately so that Oracle BI optimizer can select the most economical source

·         Content tab of logical table sources should be used to limit the number of rows returned from the database

·         Exploiting aggregate tables, pre-calculated measures aggregated over a set of  dimensional attributes are stored in aggregated tables

·         Make NULL and NOT Null column in sync with RPD and Database to avoid full outer join with NVL

·         Bitmap Index for all Foreign Key columns (for Summary reports) and Bitmap Join Index (lowest granularity based query for detailed data)

·         Usage tracking should be switched off if not needed as it will avoid extra DB Operations that run against each query.

·         Remove Snow flaking and Create Foreign keys for each dimension into Fact

I would like to emphasize that no single solution or configuration can take care of all performance issues.  Other than listed above there are many more settings / ways that helps in performance optimization, my aim was to list some of them covering highlights. Performance optimization might target Database, Environment parameters, Data model, Repository Design or Report Design having the same objective. While performance cannot be optimized by applying certain set rules and standards, having faced different challenging environment and experience helps..

It's all about working around and coming up with the best solutions for the project understanding the project flows and business scenario's. To have an optimized performance application we need to follow the best practices and design standards.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter