Monthly Archives: April 2014

Roll out your roll forwards with HFM!

April 30, 2014

A roll forward is a reconciliation of the change in balance for an account over a period of time.  For accountants, roll forwards are essential monthly close tasks.  However, when these tasks are undertaken using antiquated spreadsheets the process can become incredibly tedious.  The manual process of inputting or validating the beginning and ending balances can be maddening.  Not to mention the added stress of source data load updates, broken formulas and human error which can all lead to unnoticed mistakes on roll forwards.  Not really the kind of process that generates a ton of faith in the numbers!

Enter Oracle’s consolidation tool, Hyperion Financial Management (HFM).  While not a built-in function of HFM, minor tweaks to metadata and rules, coupled with the creation of input forms, allow the application to automate much of the roll forward process.

For example, picture a typical fixed asset roll forward schedule displaying balances at the beginning of a period, additions, disposals, transfers, and balances at the end of the period.  HFM can be tailored to dynamically calculate the beginning and ending account balances from the source data.  Accountants would simply be left with having to justify the change in balance by inputting values into HFM for the additions, disposals and transfers adjustment members.  A validation account member would be set up to ensure that the beginning and ending balances do not have any unexplained variances.

RollForward_blog

The above table demonstrates a commonly constructed property, plant & equipment (PPE) roll forward in HFM..  These roll forwards can be broken out by cost and accumulated depreciation and amortization if/where necessary.  HFM can also capture the foreign exchange impact for reviewing roll forwards in translated currencies.  Roll forwards can also be configured to automatically push data into cash flow schedules.

From a system design perspective, the components to make these roll forwards functional would be built into one of the available custom dimensions of the application.  After the metadata has been set up, business rules would be constructed to pull beginning balances, ending balances, variances, and currency translation impacts.  Depending upon the complexity, rules can be reused for other roll forwards, which would minimize maintenance on the part of administrators.

Roll forward schedules would need to be created in HFM using web data entry forms, which enable accountants in the field to enter data for predefined views.  These data forms would allow a single platform for entering, reconciling and validating roll forward data for all entities in the organization.  Security settings should be set up to prohibit users from accessing data sets for which they are not provisioned.

When properly set up, HFM can do a lot of the heavy lifting associated with roll forward creation.  So what are you waiting for?  Roll out!  Roll out!  Roll out your roll forwards with HFM!

For more information on using Hyperion Financial Management, please email communications@performancearchitects.com.

Author: Joseph Francis, Performance Architects


© Performance Architects, Inc. and Performance Architects Blog, 2006 - present. Unauthorized use and/or duplication of this material without express and written permission from this blog's author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Performance Architects, Inc. and Performance Architects Blog with appropriate and specific direction to the original content.

Collaborate 2014 Conference Resources and Outcomes

April 23, 2014

The Performance Architects team recently attended the Collaborate 2014 Conference (COLLABORATE 14) in Las Vegas. COLLABORATE 14 is the largest meeting of Oracle user groups representing the full family of Oracle business applications and database software, and is co-hosted by Independent Oracle Users Group (IOUG), the Oracle Applications Users Group (OAUG), and Quest International Users Group (Quest).

Performance Architects made significant contributions to the COLLABORATE 14 conference by presenting nine sessions on a diverse set of business intelligence (BI) and enterprise performance management (EPM) topics; by participating in several Special Interest Groups (SIG) and conference sessions; and by sponsoring a booth.

Presentations

Our presentations included:

  • Monday April 7th, 2:10 PM – 3:10 PM PDT: Endeca SIG Meeting (Session ID#: 15332)
  • Tuesday April 8th, 11:30 AM – 12:30 PM PDT: It’s a Whole New World …Rethinking Business Intelligence (Session ID #: 14685)
  • Tuesday April 8th, 3:00 PM – 4:00 PM PDT: What Is FDMEE? And Why Should I Care? (Session ID #: 14801)
  • Tuesday April 8th, 3:00 PM – 4:00 PM PDT: Integrated Strategic and Operational Planning at The University of Chicago (Session ID #: 14757)
  • Wednesday April 9th, 8:30 AM – 9:30 AM PDT: Oracle EPM (Hyperion): Sorting Out the Acronyms (Session ID #: 14698)
  • Wednesday April 9th, 2:00 PM – 3:00 PM PDT: The Three Bears: BI, EPM, and Endeca (Session ID #: 14649)
  • Friday April 11th, 8:30 AM – 9:30 AM PDT: Building a Financial Pulse Oximeter: Using Hyperion Strategic Finance to Evaluate Strategic Choices (Session ID #: 14763)
  • Friday April 11th, 9:45 AM – 10:45 AM PDT: Implementing Hyperion Planning on Exalytics: Harvard University Case Study (Session ID #: 14739)
  • Friday April 11th, 11:00 AM – 12:00 AM PDT: It’s a Whole New World …Rethinking Business Intelligence (Encore Session)

If you missed any of these presentations and want to learn more, these presentation decks are posted on the Performance Architects Learning Center. Our Learning Center is a free community and forum that provides access to functional, technical and industry-specific sessions, webinars, and white papers developed during our many years of experience working with organizations with similar interests and needs; sign up here to access this information.

Community Sessions

The Performance Architects team had a chance to lead and/or participate in on a handful of the OAUG Special Interest Group (SIG) sessions, including the Oracle Business Intelligence – Big Data Special Interest Group (OBI-BD SIG), Hyperion and Endeca sessions. SIG sessions provide an educational and networking forum for people interested in specific product areas and an input mechanism for customers to help shape future enhancements to these products.

The highlights of the Oracle Business Intelligence – Big Data Special Interest Group (OBI-BD SIG) were the Oracle roadmap and the customer panel, comprised of representatives from Ameren, Dell and Humana.  Panelists shared their opportunities and challenges with their Oracle Business Intelligence Enterprise Edition (OBIEE) and Oracle Business Intelligence Applications (OBIA) environments.

The Hyperion SIG focuses on the Oracle EPM (Hyperion) product suite; key discussion topics from the Hyperion SIG session included:

Performance Architects’ Kirby Lunger is co-chair of the Endeca SIG. The Endeca SIG was formed to allow users to share information and best practices on use of Oracle Endeca products, including Oracle Commerce, Oracle Endeca Information Discovery and Oracle E-Business Suite Extensions for Oracle Endeca. Oracle product or SIG team members provided updates on each of these product roadmaps during the session.

Want to participate in any of the SIGs?  Sign up here: http://oaug.org/communities.

Authors: Sree Kumar, Performance Architects


© Performance Architects, Inc. and Performance Architects Blog, 2006 - present. Unauthorized use and/or duplication of this material without express and written permission from this blog's author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Performance Architects, Inc. and Performance Architects Blog with appropriate and specific direction to the original content.

Large JVMs or Many Small JVMs for Oracle Hyperion Planning

April 16, 2014

With the cost of server memory (RAM) falling, and availability of highly capable hardware (with high Mean-Time-to-Failure) now common, it is not abnormal to deploy multiple instances of the Oracle Hyperion Planning service on a single server. This improves server hardware utilization, and also provides some degree of high-availability. This is especially important as it is not uncommon to have a user-base count in the many hundreds for a Hyperion Planning environment. This is a result of the enterprise-wide consolidation of budgeting and forecasting solutions to drive both efficiency and accuracy of the planning process across lines of business (LOB), geographies, etc. 

In this case, instances of Hyperion Planning are deployed as Java Virtual Machines (JVM) on an application server such as Oracle WebLogic.

at

Fronted by a load-distribution capable web server like Oracle HTTP Server (OHS), users accessing the Hyperion Planning system are directed between the instances on a round-robin basis. However, since not every user is the same, the load that each user adds onto the instance will vary. This is because the objects or data set accessed by the users will vary, depending on their role or function.

In such cases, if one of the instances gets overwhelmed to the point that it cannot recover, the instance will crash and OHS will redirect the users to the additional instance(s). This shields the majority of users from the habits of the few “high volume” users. The users who are not on the instance that crashed will never know the difference.

Support personnel can then restart the failed instance and add it back into the load-distribution pool. So, if a large user-base is accessing just one or two applications, system availability can be increased by having multiple instances with memory allocations that are not too large.

However, with the budgeting process converging across the enterprise, this also leads to multiple Hyperion Planning applications or solutions introduced onto the same system. Users may access different applications, based on their role or LOB. Such a system would have more than two applications, often times leading into the double-digit application count.

This introduces a different problem. Since the web server distributes user traffic on a round-robin basis, and not based on what application is being accessed, two users accessing the same application may get directed to two different instances. In such a case, each of the instances will need to load a copy of the application in their memory spaces. And when another application’s user connects to the instance, another application will get added on to this, and the memory space fills up quickly. The more applications in the system, the faster this grows.

In such a scenario, simply having multiple instances does not help. Individual instances need to be sized properly to have enough memory allocation (also known as heap space) to potentially load all of the applications. So it is important to weigh this option when deciding how many instances of Hyperion Planning to deploy for the system. This is also a factor if horizontal scaling is used, as it does not matter if the instance is on the same machine or not.

So how much heap size are we talking about? Your consulting partner should be able to help you out based on your application size. But there is almost always an element of performance testing involved to determine this. When dealing with systems with copious amounts of RAM, like Oracle’s Exalytics machine, it is not difficult to go into double-digit heap size allocations (when taking in gigabytes). Also, there are other settings or flags available to tune the characteristics of the instance, based on what application server technology is used (e.g., Oracle WebLogic, IBM WebSphere, Oracle HotSpot). This is not meant to make your life as a system and/or application administrator more difficult, but to give you options to cope with an ever-growing user-base and system load. Having one large instance or multiple small instances is not a silver bullet for performance issues, but more of a shield to protect your well-built application from the barrage of future growth, and to make your system last longer and to make your users stay happy for longer.

Author: Andy Tauro, Performance Architects


© Performance Architects, Inc. and Performance Architects Blog, 2006 - present. Unauthorized use and/or duplication of this material without express and written permission from this blog's author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Performance Architects, Inc. and Performance Architects Blog with appropriate and specific direction to the original content.

Using “Selection Steps” to Filter Data from Essbase Hierarchy Dimensions in OBIEE 11G

April 10, 2014

Is it possible to select or filter on any specific members after a query is applied to your business intelligence (BI) analysis?  Absolutely!  You can apply “Selection Steps” to filter any specific members from your BI analysis even after a query has been executed.  The next few diagrams demonstrate exactly how this concept can be applied.

For this specific example, we’re demonstrating a sample of Essbase integration with Oracle Business Intelligence Enterprise Edition (OBIEE) 11g:

jk 1

The “Time Dimension Column” displays five options to choose from:

  • Sort. This option lets you sort your results in either ascending or descending order
  • Edit Formula. This option allows you to write your own custom formulas
  • Column Properties. This option lets you customize various data property options
  • Filter. This option permits you to apply various conditions
  • Delete. This option lets you delete a specific data column  

jk 2

Unlike the “Time Dimension” column displayed above, there are only three options to choose from in the below “Products Dimension” hierarchical column.  You are probably wondering how you can filter or if it is even possible to filter on a specific “Products Dimension” hierarchical column.  The quick answer is that you can’t apply filters directly to hierarchical columns, as demonstrated in the image below.

jk 3

Once you click on the “Results” tab, all products show up from the “Products” hierarchy.  The “Selection Steps” marked in red below will allow you to filter on any specific “Products” hierarchy.  The next few diagrams demonstrate how to filter on hierarchical columns.

jk 4

Once you expand on the “Selection Steps,” you can locate both “Facts” and “Dimensions” underneath the “Menu” option.  You can scroll down and locate the “Products” hierarchy.  Double-click on the “Start with all members” area located right underneath “Products –> Products Hierarchy.”

jk 5

Once you are in the “Edit Member Step” menu panel, you can select one or all products from the left-hand “Available” panel and move them to the right hand “Selected” panel.  For this specific exercise, we only opted to go with “BizTech” from the “Products” hierarchy.

jk 6

Notice now “BizTech” is only displaying from the “Products” hierarchy.  Using the “Selection Steps” option is the easiest way for anyone to filter hierarchy dimensions in the Essbase hierarchy structure in OBIEE 11g.

jk 7

Author: Jon Kim, Performance Architects

 

 

 


© Performance Architects, Inc. and Performance Architects Blog, 2006 - present. Unauthorized use and/or duplication of this material without express and written permission from this blog's author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Performance Architects, Inc. and Performance Architects Blog with appropriate and specific direction to the original content.

Optimizing the Supply Chain with Oracle Hyperion Essbase

April 2, 2014

As modern day business leaders are constantly seeking new ways to add value for stakeholders, focus has turned much more intently on supply chain management (SCM). The overarching goal of SCM is to boost profitability by managing resources and costs efficiently to maximize returns on capital employed. Within this realm, there are multiple areas to target for analysis and optimization, including demand planning/forecasting, inventory management, procurement, and transportation/logistics.

For many companies, inventory represents one of the single largest items on the balance sheet. Cost associated with poor inventory decisions (holding too much or too little of specific products or raw materials), as well as product obsolescence and holding costs can significantly eat away at company profitability. Product demand is inherently variable, and companies must absorb this variability through some combination of inventory, capacity, and customer service, all of which are costly and consume valuable resources.

One of the biggest challenges a company faces is trying to accurately forecast product demand to mitigate the costly effects of demand variability. Being able to produce a good demand forecast is a foundational element that impacts many downstream supply chain decisions. As technology has advanced, companies have been able to collect and analyze larger amounts of sales and customer data to build more accurate forecasts.

For more sophisticated demand forecasting models, analysts are often most comfortable building models offline in some complex array of spreadsheets linked together and to a range of disparate data sources. As business needs grow and change, and businesses become more global, spreadsheets are simply not able to scale to meet the challenge. Businesses need to take a centralized and systematic approach to creating demand forecasts, one that is directly tied in to the larger finance planning process.

Oracle’s Enterprise Performance Management (EPM) suite is a powerful set of tools that is already an integral part of the finance planning process for many companies. In many cases, though, companies using these tools are not taking full advantage of the massive analytical capabilities that are built into Oracle Essbase, the multidimensional database engine at the foundation of the EPM suite.

Essbase can accomplish anything that a spreadsheet can, plus a great deal more. The product has many powerful functions that are often overlooked that can build sophisticated demand forecasts. Analysts can build simple models that derive forecasts based on some trailing average of specified periods of historical sales data, or build something much more complex that predicts future demand based on historical trend. For example, the @TREND function allows for choosing a forecast method (simple regression, or single, double, or triple exponential smoothing) and gives analysts the ability to weight historical data based on their understanding of the data characteristics (market changes, seasonality, etc.).

For executives interested in the discipline of lean supply chain management, Essbase can be an invaluable tool for integrating large volumes of data and producing actionable insights to make better business decisions that will reduce costs and ultimately increase the bottom line. Essbase is scalable to accommodate the needs of growing businesses that are global in nature, and allows for an integrated approach to all levels of planning and forecasting that must be done across an organization.

Author: Shane Hetzel, Performance Architects


© Performance Architects, Inc. and Performance Architects Blog, 2006 - present. Unauthorized use and/or duplication of this material without express and written permission from this blog's author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Performance Architects, Inc. and Performance Architects Blog with appropriate and specific direction to the original content.