Monthly Archives: March 2014

Oracle Higher Education User Group (HEUG) Alliance 2014 Conference Recap

March 25, 2014

Earlier this month, several Performance Architects team members attended the Oracle Higher Education User Group (HEUG) Alliance 2014 Conference in Las Vegas.

Community Sessions

We had a chance to sit in on a handful of the Product Advisory Groups (PAG) Community Sessions, including the Budgeting and Reporting & BI sessions. PAGs are charged with advocating to Oracle on behalf of the Higher Education community and with helping to facilitate product enhancements.

Budgeting PAG Community Session Discussion Outcomes 

The “hot topics” at the Budgeting PAG meeting were no surprise, given the amount of press these have been getting lately.

The first topic was the launch of Oracle’s Planning and Budgeting Cloud Service (PBCS) (Performance Architects was one of the few Oracle partners selected for the Oracle Preview Program, and if you’re interested in learning more, please send us a note at sales@performancearchitects.com).  This is an especially attractive option for institutions looking to get started with Hyperion Planning, but without committing to a full-scale, on-site implementation.
The second theme was related to the continued enhancements to Oracle’s Public Sector Planning and Budgeting (PSPB) solution, most commonly used for workforce planning (both position and employee planning).  Many institutions involved with the Budgeting PAG have now successfully implemented PSPB, and many had “lessons learned” and best practices to share from their implementations (Performance Architects has successfully implemented PSPB; if you’re interested in networking with peers or in learning more about what works in this arena, or learning more from Performance Architects about how to avoid pitfalls, send us a note at sales@performancearchitects.com).

Reporting & BI PAG Community Session Discussion Outcomes

The Reporting & BI PAG community meeting was comprised of a mix of two main user communities:

  1. Those implementing or using Oracle Business Intelligence Enterprise Edition (OBIEE) or Oracle Business Intelligence Applications (OBIA).
  2. Those working with legacy PeopleSoft products.

One hot topic on the OBIA front is the newly released Student Information Analytics (SIA) application. This is pre-built ETL, data warehouse, and OBI content specifically geared towards analyzing student data at Higher Education institutions. Several institutions of varying sizes are in the process of implementing SIA, and the PAG group held lively discussions about the implementation process as well as the potential for enhancements (Performance Architects is one of the few partners currently implementing this new product with our clients and we are happy to share our thoughts around what is working and what isn’t in the current solution set; if you’re interested, send us a note at sales@performancearchitects.com). Another key theme of the Reporting & BI PAG community meeting was the concept of pushing embedded analytics within the institution, which is really about ensuring that analytical processes are part of everyday life at both the administrative and academic levels.

Oracle Roadmap Discussion Outcomes

Oracle also presented separate sessions on the Business Intelligence (BI) and Enterprise Performance Management (EPM or Hyperion) product roadmaps. Some of the key takeaways from those presentations include:

Oracle EPM (Hyperion) Roadmap Discussion Outcomes

  1. As we mentioned in the Budgeting PAG discussion summary, Oracle’s Planning and Budgeting Cloud Service (PBCS) has launched. The next Oracle EPM (Hyperion) capability Oracle will offer in the cloud is Hyperion Financial Reporting.
  2. Optimizing existing BI and EPM implementations with engineered systems, such as Exalytics. Hyperion Financial Management (HFM) will soon be certified on the Exalytics platform.  If you want to understand why this is important for the Higher Education market, check out our recent blog post entitled, “The Case for Financial Close and Consolidation Solutions in the Higher Education Market: Why Your Institution Should Evaluate Hyperion Financial Management (HFM).”
  3. Enhancing the mobile capabilities of the Hyperion solution set. Coming soon will be the ability to review Hyperion Financial Reporting, submit web form data to Hyperion Planning and Essbase, and manage the planning cycle workflow.
  4. The support for PeopleSoft EPM (not to be confused with the Oracle EPM or Hyperion product suite) is evolving and institutions will want to evaluate what to do in relation to their EPM environments as a result of these changes.

Oracle Business Intelligence (BI) Roadmap Discussion Outcomes

  1. “Oracle Financial Planning Analytics” is an upcoming OBI application that will integrate more tightly with the Hyperion toolset to provide pre-built OBI content for reporting on top of Hyperion’s Public Sector Planning and Budgeting (PSPB) module. This will include enhancements for reporting on commentary from Hyperion!
  2. A new version of the Oracle BI Mobile App Designer is coming soon. There will be two main apps in the iTunes and Android market: BI Mobile HD (for end users) and BI Mobile App Designer (for developers).

Other Discussions and Presentations Outcomes

In addition to the community sessions and Oracle roadmaps, we were pleased to see that the topic of unstructured and semi-structured data has also finally made its way onto the agenda at this conference. A number of institutions mentioned that they are adopting the Hadoop framework, mostly in pilot projects, to explore topics like learning management system (LMS) behavior, student chat feedback, and student portal log mining. Other institutions are using Oracle Endeca Information Discovery (OEID) to perform data exploration on student success and grants productivity; one institution was able to get an application up and running in a matter of weeks with little vendor support!

Finally, we really enjoyed the discussion and debate at the two Performance Architects presentations at the conference.  Our first presentation was with client Harvard University and was entitled, “Exalytics: Adding Value to the Harvard University Budget System,” where we talked about the 2x-6x performance gains Harvard has seen in their Hyperion Planning environment by implementing Exalytics.  Our second presentation with client The University of Chicago, “Integrated Strategic and Operational Planning at The University of Chicago,” was a review of the budgeting and forecasting process and solution transformation (using the Oracle EPM or Hyperion suite of tools) at the institution.

If you missed either or both of these presentations and want to learn more, we are hosting live webinars on both topics; the Harvard Exalytics webinar will be held on March 26th at 3 PM EST and The University of Chicago webinar will be held on April 14th at 3:30 PM EST.  The Alliance presentations, as well as the webinar decks and recordings, are also posted on the Performance Architects Learning Center, a free community and forum that provides access to functional, technical and industry-specific “How To” 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.

Authors: Michael Bender, Jonathan Etkin, and Kirby Lunger

 


© 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.

Calculation Manager 11.1.2.2 Fails to Export Member Block Variable

March 20, 2014

During a recent project involving Oracle Hyperion Enterprise Performance Management (EPM), we encountered an issue migrating Calculation Manager Rules via Life Cycle Management (LCM). While LCM did a good job of re-creating most of the variables used in the rule on the target, one variable didn’t make it. On further investigation we found that the variable is different from the others. Consider the rule of the format:

FIX ({var1}, {var2})
    {var3} (
     …………………
    )
END FIX

In this case, we found that the variables {var1} and {var2} would get exported from the source just fine, including their properties like limits, prompts, etc. However, {var3} does not seem to get exported.

When we tried exporting the rule via the Calculation Manager “System View” interface, we found that there too it would not export this variable. So the problem is not with LCM, but with Calculation Manager itself. On examining the XML file that is exported, there is a <variables> section that can be found at the beginning of the file. This section lists {var1} & {var2}, but not {var3}. However, {var3} is mentioned in the code block later on.

Oracle Support has been able to re-produce the issue in-house and have registered a BUG for this: Bug 18227617 – MEMBER BLOCK VARIABLE DOES NOT GET IMPORTED USING LCM; NOT SEEN IN VARIABLE LIST.

As a work-around to this, either create the variable manually on the target (a one-time thing if the variable is “Global”), or export/import the variables from Calculation Manager directly, and separately from the rules.

The good news is that this issue does not seem to exist in Calculation Manager 11.1.2.3. I guess just another reason to keep up with upgrade cycles.

Author: Andy Tauro, Performance Architects

*Please note: bug has been fixed as part of the Patch Set Update 106 for Calculation Manager*


© 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.

So…What is FDMEE? And Why Should I Care?

March 19, 2014

Several months ago Oracle released a product called Oracle Hyperion Financial Data Quality Management Enterprise Edition (FDMEE) in Version 11.1.2.3 of EPM– hallmarked as the more athletic, and handsome, brother of the old standby Oracle Hyperion Financial Data Quality Management (FDM). After working through a couple implementations of the tool I’ve come up with a short list of reasons why you should care about this product, and what it means for the EPM space.

Let’s start by talking about what FDMEE is, exactly. FDMEE is a data integration tool that helps transform data and metadata from a variety of sources into a consumable format for Hyperion products like HFM, Planning and Essbase. It allows you to apply mapping logic though a friendly user interface, and manage the data load process via a web console in Hyperion Workspace. For those of you who are thinking, “Tom, that sounds a lot like FDM,” you’re correct!  It does sound like FDM. What differentiates FDMEE are a few changes that make it a superior option to what FDM can do on its own – and this is why you should care.

First and foremost, FDMEE leverages the 64-bit architecture seen in products like ERP Integrator (ERPi), another Hyperion product that used to be paired with FDM to get data out of particular source systems such as EBS and PeopleSoft. This move to a 64-bit architecture has brought noted stability and scalability when it comes to implementing the tool; I haven’t been able to break the tool by throwing “too much” data at it, or running too many processes in parallel. Since FDMEE is Java-based as well, we’re able to tune the application to improve things even more.

Second, FDMEE integrates directly with source systems and target applications. When you need to source data from PeopleSoft and load it to Hyperion Planning, there isn’t a two-step process to jump from ERPi to FDM – instead we go straight from one to the other. This reduces complexity and removes an integration point – helping to both shorten development time, and also give folks like me the opportunity to add value to Finance users with a bit more free time. FDMEE interfaces directly with EBS, JD Edwards, SAP, flat files, and the open interface table – a custom source system that lets FDMEE receive data from a data warehouse, an ODS, or other custom sources.

Third, I’ve found FDMEE to be friendlier for Finance users than FDM and ERPi – and this is a huge plus. The tool now sits in Hyperion Workspace, so users don’t have to navigate between ERPi in Workspace and FDM on a separate webpage when making changes or running a job. The tool also has a built in scheduling process to run automation routines that is managed through the tool’s web interface, making it simple for a user to schedule a reload job.

Fourth, FDMEE lets you source metadata from a source system and load it into target applications. This capability exists in ERPi, but for those folks who only have FDM this is a huge step in a new direction. Now, reporting and planning systems can be paired with metadata sourced on a regular basis from a general ledger; adding new accounts, entities, departments, etc. on the fly. This reduces maintenance for the administrators, and when paired with creative mappings, can ensure that data always finds its way into the Hyperion world.

Finally, users can now start in a retrieve form, a web form, or a report and drill back to FDMEE to see source records, mapping logic and process times. From there, users can continue on and drill all the way back into a source system, such as EBS, to review individual transactions and see the detail. This is a deadly combination.  When paired with mappings that bring data to a summary level for reporting, users can quickly drill back and analyze data – reducing the need for all the detail existing in Hyperion.

For example, we could map all GL travel expense accounts to a member called “Travel Expense” in Hyperion, creating a summary number for reporting. If the need arises to see detail, users can quickly drill back into FDMEE to view the balance on every source Travel Expense account, and if interested, continue on into the ledger to review all the transactions that make up a single chart-string balance. In the old world, this was a clunky process that took several steps, and FDMEE has reduced the complexity making life easier.

I could continue on for several pages about everything else I’ve learned, but instead I’ll wrap things up and shamelessly plug my upcoming presentation at Collaborate on FDMEE on April 8th, 2014 from 3-4 PM PST entitled “What is FDMEE? And Why Should I Care?” (Session ID: 14801; a full list of Performance Architects’ eight sessions and booth information is available here). In short, FDMEE certainly is the natural progression of FDM and ERPi, bringing together the best of both worlds, and adding a little extra. For folks considering integrating source systems with Hyperion, FDMEE is a serious contender.

More to come in future blogs, but for now, have a good one.

Author: Tom Blakeley, 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.

How to Fix “208 Days” Bug in Oracle Exalytics

March 10, 2014

Have you ever gone on vacation only to come back and realize that you cannot access your Oracle Exalytics machine anymore? Your ILOM (Integrated Lights Out Manager) shows that system is up and your network team says that ports are open in the firewall…but it isn’t working?

It turns out that there is a bug in the system where the Oracle Exalytics Linux operating system may crash or become unresponsive after 208 continuous days of use.

To fix the issue you will have to restart your system. You can also refer to the following document on Oracle Support if you run into this issue: “Oracle Exalytics Stability and Performance Issues on Compute Nodes Running Linux After Uptime of More Than 208 Days (Doc ID 1579194.1).”

For your reference, the Oracle documentation states, “To avoid further occurrences of the 208 days issue; it is recommended that you pro-actively plan to reboot the server before 208 days of continuous up-time has been reached.  By doing this, you will avoid any unplanned server outages.”

If you want permanently fix this issue, you’ll need to apply Patch Set 4 (PS4).

If you have run into this issue or would like to chat more about Exalytics, please email communications@performancearchitects.com.

Author: Suresh Vupputuri, 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.

Why an Operational Data Store (ODS) Provides Important Capabilities in an Analytic Architecture

March 5, 2014

The Operational Data Store (ODS) concept is a viable architectural option when considering complex requirements that deal with data transformation and mapping logic for large data sets.  At times, data feeds from multiple source systems require manipulation through the use of various reference data that may exist either as look-up tables or master data tables residing within the repositories of Oracle Enterprise Performance Management (EPM) – also known as Hyperion – solutions.  You can gain flexibility by building in a relational layer between your source systems and downstream analytic tools.

As an example, a recent client chose a solution that requires both an Essbase Aggregate Storage Option (ASO) analysis cube and a Block Storage Option (BSO) Planning cube. The sources of data are the same for each of these; however the requirements, functionality, and data needs are quite different when comparing the two. Furthermore, there are multiple years of data in the source systems that include many attribute-type data elements. The files themselves are also large with many exceeding 1GB in size per each year of data being sourced. These characteristics require an agile solution that can adapt with regards to the handling and distribution of data.

By importing source data initially into a relational layer (tables, views, and stored procedures), we can quickly modify outbound data files according to the changing needs of the receiving analysis systems (Essbase cubes in this example). Also given the increasing size of the data sets involved, the ODS proves to be a flexible and effective tool as data is turned around quickly to meet needs downstream.

On the master data front, the ODS also provides benefits. We are able to generate Oracle Hyperion Enterprise Performance Management Architect (EPMA) load format (.ads) files directly out of the ODS layer for loading into EPMA. This is a required step in the administrator-owned data load process for aligning the ODS exported data files with the base-level hierarchy members available in Essbase dimensions.

The EPMA .ads files can be generated in two ways for loading EPMA. We first are able to generate a full file of all base members required in the dimension (a complete product listing, for example). Secondly, a subset file can be created which contains just newly added members (e.g., products). The latter is accomplished by joining to EPMA repositories and essentially conducting a compare to see what already exists in the downstream Essbase analysis cube. This file is then used for incrementally updating the members in the hierarchy versus conducting a full replacement. This is an added benefit to the administrator who maintains parent/child relationships within EPMA and only needs to add new members.

The SQL code built into the ODS views also allows the ability to control member and alias naming conventions via string length and prefix/suffix handling, which eliminates the need for global transformations in the Essbase load process.

In summary, an ODS can provide benefits that include:

  1. Offers scalable and flexible control of data per constantly changing business requirements
  2. Handles large data sets effectively, sourcing from one or multiple systems/databases
  3. Enables fast turnaround, with ownership on the business applications side of the house
  4. Offers data scrubbing and transformation, with the ability to modify both data and master/meta-data
  5. Provides the capability to build queries against tables and views for quick answers to large, complex data sets

Keep warm this winter and stay tuned to our blogs for additional discussions regarding BI solutions! If you would like to learn more, contact us at communications@performancearchitects.com.

Author: Jason Sawicki, 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.