As part of a recent PBCS implementation, I had to design an Intercompany Elimination solution within a BSO plan type.  This is a relatively uncommon and somewhat complex requirement in a PBCS implementation, and as such, I was pretty excited to blog about it.  Unfortunately, the whole post nearly ended up in the garbage.

Every once in a while, you write a great post, only to find that someone else has covered the topic sooner than you, and better than you.  In my case, that person was Martin Neuliep.  In Developing Essbase Applications:  Hybrid Techniques and Practices, Martin shared a single formula that elegantly calculated Intercompany Eliminations in a BSO cube.  In addition, Martin was scheduled to discuss the topic of eliminations at Kscope in his presentation on June 28^th, titled Eliminating the Eliminations Problem in Essbase.

Luckily, I had an ASO solution in addition to my BSO approach.  As it turns out, Martin was planning on sharing several elimination calculation methods in his presentation (some developed by him, and some developed by others).  Perhaps my ASO solution will make the cut?  (hint, hint)  Either way, you should check out his presentation – I will definitely be there.

With all of that said, here’s my approach to intercompany eliminations in an ASO plan type.

Why Eliminate Intercompany Transactions?

Most of my clients create their Budgets and Forecasts in a “fully eliminated” manner.  Within their financial plans, they pretend that certain transactions don’t occur, because these transactions don’t affect the company’s bottom line.  If one subsidiary has a planned (future) transaction with another subsidiary, these clients may not bother recording the transaction in their Budget or Forecast at all.  While this approach simplifies the planning process, it leaves out financial details that may be useful.

When a client’s requirements indicate that these transactions are relevant to the planning process, we need to build Intercompany Elimination logic into our PBCS applications.  The accounting behind these transactions can get pretty complex, but what we’re going to focus on today are the technical mechanics that facilitate these eliminations in PBCS, specifically in an ASO plan type.

So why do we eliminate these transactions?  Because depending on where we’re looking in the Entity dimension, we need to pretend that they never occurred.  As far as most investors are concerned, these intercompany transactions don’t count on a consolidated basis.  Imagine selling a car to your spouse . . . this transaction doesn’t change your combined income or net worth.  This leads us to an interesting question.  When and where do we eliminate these transactions?

The Land of Make Believe

Let’s start with a simple example – an imaginary company that mines raw materials and sells those materials to its manufacturing units.  These plants then sell finished goods to a distribution company.  All of these entities are part of the same vertically integrated company.  The Entity dimension in such a company might looks like this:

To facilitate transactions between two entities in an OLAP database, it is generally beneficial to have separate Entity and Customer Dimensions.  The only unusual thing about this design is that many of the “Customers” are internal customers.  As such, our Customer dimension might look something like this:

Note that all members under Intercompany Trading Partners correspond with a member from the Entity dimension exactly, but with an “ICP_” prefix.  This ASO solution will not work if there are discrepancies between Entity dimension members and the members underneath Intercompany Trading Partners.

Planning Intercompany Transactions

Intercompany transactions can occur across a variety of accounts within the Income Statement and Balance Sheet.  The simplest example is one subsidiary selling something to another subsidiary within the same company.  Let’s assume that US Mine 1 sells something to Plant 1 for $100.  Our level zero data would look something like this in PBCS:

If we were to look at Total Sales for US_Mine_1, we would want to see $100 in sales.  But what if we wanted to see Total Sales for Total_Company?  Assuming that this was the only sale, we would want to see $0.  This is because the transaction must be eliminated at the first common ancestor between the Entity and the Intercompany Trading Partner.  Total_Company is the first common ancestor between US_Mine_1 and Plant_1.  What makes this calculation interesting is that the transaction should NOT be eliminated at ancestors before we arrive at the first common ancestor.  So we would definitely expect to see that $100 in sales to show up in the parent member Mining.

The Dreaded “E-Company”

The ins-and-outs of these transactions can get tricky, especially when mixing many transactions together with both internal and external customers.  Developers will likely have multiple sets of accounts being eliminated.  (E.g., Sales & Cost of Sales, Receivables & Payables, etc.)  Ragged hierarchies and alternate hierarchies can add additional complexity.  For this reason, it can be helpful to introduce “E-Companies” into the Entity dimension.  These are basically fake companies where we only calculate and store elimination data.

Adding E-Companies to the Entity dimension might look something like this:

Unfortunately, E-Companies can make an Entity dimension convoluted.  If your company structure is particularly large or volatile, E-Companies can create a significant amount of maintenance.  They can also be confusing to end-users who might not understand their purpose.

** NOTE– Most intercompany elimination solutions in BSO databases require E-Companies!

ASO to the Rescue!

One of the nice things about PBCS implementations is that they often include an ASO reporting database.  In fact, some clients never consolidate their BSO databases at all, and instead, simply map their data to an ASO cube that rolls everything up on the fly – no business rule required!  And here’s where things get really awesome – in an ASO database, we can calculate intercompany eliminations without the need for E-Companies.

Here are some things to consider when designing your ASO plan type:

• Both an Entity dimension and a Customer dimension are required.
• The Intercompany Trading Partner hierarchy (within the Customer dimension) must match the Entity dimension exactly, with the exception of “ICP_” prefixes. This includes intermediate parents.
• A “Data Source” dimension of some type is required to separate regular data from eliminations.
• Account dimensions in ASO databases are automatically set to Dynamic. The Data Source dimension will also need to be dynamic to support member formulas.

The Data Source Dimension

In this solution, all of the logic associated with eliminations lives in a dimension called Data Source (or something similar).  In this dimension, all base (non-eliminated) data is loaded into a member called Amount.

This dimension also includes a special “Do Not Use” system-only type of member.  Here, it is called Do_Not_Use_Elim.  We generally do not want users querying this member.  It’s just a temporary holding spot for storing the inverse balance for any intercompany accounts.  This member can be populated with a procedural calculation.

It is important to note that the “Do Not Use” member should be populated in the same member combinations as the original budgeted amount, with the exception of the Data Source dimension.  Remember – this “Do Not Use” member is simply a holding spot.  Users should not query this member.

Abracadabra

The “real” magic happens in the member above called Elim.  In this member, a formula is added that filters when eliminations (stored in the “Do Not Use” member) are displayed and subsequently added to the base data.

When the Elim member above rolls up to the Consolidated_Amount member, we see that the intercompany sales amount goes to zero.  In other words, it is eliminated.  (See row 15 below)

The example above shows our original sale (row 3) and the inverse amount stored in the “Do Not Use” member (row 4).  Rows 7-10 show our original Entity and its ancestors.  We can see that the eliminations are only displayed at the first common ancestor and above (row 9 – 10).  Finally, in rows 13 – 16, we see the view that most users will retrieve, using the member Consolidated_Amount.  This member takes the base data in the Amount member and layers in the data in the Elim member.  As such, we can see that the sale exists in rows 13 & 14, but is eliminated in rows 15 and above.

Wrap-Up

Like most calculations in PBCS (and Essbase in general), there are multiple options for solutions, each with its own pro’s and con’s.  This solution works well against very large databases and has the added benefit of not requiring E-Companies.  Happy Eliminating!

This post corresponds to the webinar “Connecting the Value of IT: A Disciplined Solution for Service Costing and Chargeback,” the last in our “Let Your Profitability Soar” webinar series. You can access the recording here.

Within an organization, technology is mission-critical to most business strategies, and IT costs represent a significant portion of back office spend.

Among their many responsibilities, the CFO and the CIO must make sure that:

• Technology spending is aligned with business strategy
• Business applications and end-user services are delivered efficiently and cost-effectively
• Coherent project portfolios that grow and transform the business are created and nurtured

Within this new economy, a key ongoing goal of the CIO is to make sure that IT is aligned with business strategy.

Generally, this IT-to-Business Strategy alignment is achieved in two ways:

1. Running the business: Providing a cost-effective level of internal services necessary for sustaining business activity.
2. Building the business: Managing and delivering portfolio development projects that are prioritized and aligned with all key business initiatives aiming to improve efficiency and aid in gaining competitive advantages.

The Nature of the Problem

One challenging pattern we see time and again is the ongoing disconnect between the CIO and the CFO.

Some might say this disconnect is an inevitable result of the fact that technology is moving so fast and we don’t always have the time to stop and assess its value. Understandably, it can be difficult for a CFO to get away from all the checks and balances just to get the financial books closed, let alone turn attention to the books that measure performance at greater depths, like line of business.

In general, as a function of the role, the CFO does not talk servers, desktop deployments, applications or other semantics of the technology business. Conversely, with many companies establishing Technology Shared Service Centers, pressure is placed on the CIO to operate the business of IT with the same financial disciplines the CFO requires of all lines of business. The CIO must connect the value of IT services and capabilities to internal business partners. To achieve this, IT Finance teams require performance management solutions that are IT-specific, yet are connected to Finance, to ensure efficient allocation of resources and effective delivery of internal services.

Part of the CFO’s role is to look at the technology projects and initiatives and think about how all of this technology is adding value. CIOs have to fill information voids, while also having to build their own financial models and performance management book of record using their own resources.

Two seemingly differing views of value can be hard to navigate and leverage. If two divergent approaches are not connected in a common view among the key stakeholders, then—more often than not—there is ongoing value-related confusion. Ultimately, the dissonance between the line of business owners can stall or even paralyze decision-making.

A Better Language Is Needed

For the good of your organization, it’s imperative that the CIO and the CFO speak the same shared language of value and that they connect in an effort to move forward in the most aligned and productive manner possible.

Speaking a shared language—one that offers a unified financial model view and is based on shared definition of value—is a key to finding a solution. The disciplines of ITFM (IT Financial Management) is about equipping both of these executive-level offices and their teams with a better language.

With an ITFM solution, you are able to:

• Reduce the time that IT Finance spends on managing the business processes, providing more time for value-added analytical activities
• Give IT Managers more detailed, timely, accurate data to better understand the cost & effectiveness of the services and projects they are delivering
• Provide Line-of-Business managers with cost transparency into IT allocations and chargebacks, allowing them to better align their consumption of services with their business goals

ITFM focuses on these finance business processes:

• IT Planning: Budgeting & forecasting of IT Operating and Capital Spend
• IT Costing: Linking supply side financial cost structures with demand side consumption for services and projects
• IT Chargebacks: Equitably charging lines of business for internal services and projects performed (or Showback)

IT Finance Organizations typically manage these processes through a series of multiple systems and offline spreadsheets. These processes are not ideal, as they create pain as far as inefficiencies and ineffectiveness in terms of results.

Our preferred solution for IT Service Costing—co-developed with Oracle—is based on PCMCS (Profitability and Cost Management Cloud Service). Oracle’s PCMCS is a cloud-based, packaged performance management application. It offers, in one package, a rules engine for cost allocations, embedded analytics and data management platform.

When developing the solution with PCMCS, the following were top priorities for our team:

• That it required no large initial investment
• That it was accessible to all
• That it was always updated/up-to-date
• That limited IT involvement was needed

Oracle IT Financial Management Solution Overview

The ITFM solution, a joint development effort with Oracle and based on valuable feedback and results from multiple Ranzal customer implementations, offers all of the following in one package:

• Pre-Packaged Content for Cloud or On-Premise
• Pre-Built Data Model
• Pre-Built Costing Model & Reporting Content
• Pre-Built Interface Specifications

A key component of the PCMCS IT Costing & Chargeback Template is its approach to modeling IT Like a Service Business, which includes the following modules:

• Model Financials & Projects: This first step is focused on modeling financial projects, allowing you to combine multiple data sources, perform cost center allocations and, for those customers without an existing project costing system in place, to perform basic project costing and project allocation functions.
• Complete Costing of IT Operations: This second pillar of the solution provides a flexible framework that allows you to combine data from multiple sources, perform resource costing and perform service costing.
• IT as a Business Service Provider: This third leg of the solution service considers catalogue & bill rates, contribution cost trace, consumer showbacks and consumer chargebacks.

 We Have Options, You Have Options

Our Flexible Maturity Model allows customers to start where they feel most comfortable, and progress in a way that is focused on maximum flexibility for maximum effectiveness. No one size fits all, and we believe in starting right where you are.

For more information or to request a demo, email us. Be sure to ask if your company qualifies for our one-day complimentary PCMCS assessment of your IT Service Costing needs.

Author: Margaret Motsi, Performance Architects

One of the ways to enhance data visualization is by enabling a fully interactive drill up and down experience in the data hierarchy. Oracle’s Data Visualization Desktop (DVD) uses the
“Row Expander” custom visualization plugin, available at the Oracle Public Store, to offer the capability to dynamically drill up and down attributes that may not necessarily belong to a hierarchical column.  This means that a user can switch back and forth between summary and detail data to form intermediate subtotals and quickly analyze data.  This blog post provides instructions on how to get started with this plugin.

From the Oracle BI Public Store, find the “Row Expander Viz” plugin box and click to download.

This action will create the following notification:

Click the “Download” link and copy content into the “plugins” folder; if your installation doesn’t offer  “plugins” folder, simply create one. Additional plugins should also go into this folder.

Restart DVD. Once restarted, you can view the plugin in the visualization options.

Select desired columns and then right-click on “Pick Visualization.”

Select the “Row Expander” plugin

The attributes display as rows in the canvas along with the selected metrics. The default view shows the summary data.

You can click on each attribute to drill up/down to the next level.

You can also add or remove attributes as you go. The number of attributes is equal to the number of levels you can drill up and down.

You can also create a filter on an attribute by right-clicking the attribute and selecting “Create Filter.”

The plugin will filter accordingly.

You can drill up and down on the filtered data.

When it comes to measures, this first version of the plugin has a couple of limitations.  First, the plugin is only capable of performing the “Sum Aggregation” function on metric values. It will be able to perform other calculations in future releases of DVD.  Second, the input dataset is limited to 500 rows for the plugin to perform accurately.

Need help or advice on data visualization plugins?  Contact Performance Architects at sales@performancearchitects.com and we’d be happy to help.

When comparing two roughly-equivalent environments, UAT and Staging, we kept running into a performance discrepancy we were unable to account for. We had the same data, rules, etc. in each HFM application, the same tuning parameters applied, and similar server host capabilities. We could not run a complex Task Scheduler instruction set in the same amount of time in each environment. UAT would take 90 minutes longer to complete the few dozen steps in the Task than Staging would.Examination of the log entries showed that for nearly every task in the list, it was taking approximately 150 seconds longer to perform each step in UAT than in Staging! We recreated the task list from scratch in Staging and found this extra time between steps was reduced to almost nothing. The same list was recreate in UAT and still ran 150 seconds longer between steps. The 11.1.2.4 Task Flow was being run by the same users in each environment, but was set to RUN AS a different user. When we ran as a simple ‘admin’-type user instead, as was used in the Staging environment, the extra time between steps vanished.

Review of the differences between the simple ‘admin’ user account and the LDAP user account being leveraged for the job in UAT showed the LDAP account had 7 PAGES, or about 3500 Native Group rules, being applied for provisioning. We advised the client of the cause of the issue, and saved them 90+ minutes by their provisioning a user with more straight-forward security.

What price Essbase?

I can(‘t) get it for you wholesale

Money makes the world go round

Can’t means won’t and won’t means jail

What does it take to get to the cost of an Essbase Cloud instance?

I’m not a lawyer but I play one on TV

• Oracle Cloud Services home page, specifically the Oracle Platform as a Service and Infrastructure as a Service – Public Cloud Service Descriptions-Metered & Non-Metered pdf.  This link is as of the date of writing, 17 May 2017 so be aware that it will change.
• Oracle Analytics Cloud pricing
• Oracle Database Cloud Service pricing
• Oracle Database Backup Cloud Service pricing
• Oracle Compute Cloud Service pricing
• Oracle Storage Cloud Service pricing
• Oracle Compute Cloud Service - Block Storage pricing

Non-metered

• B87390 Oracle Analytics Cloud– Standard – Non-Metered – OCPU which includes:  Essbase, BICS Mobile, 50 named users of Data Visualizer desktop per OCPU, Smart View for all users, and however many OCPUs you buy.
• B87389 Oracle Analytics Cloud - Enterprise- Non-Metered – OCPU which includes:  BICS Mobile, 50 named users of Data Visualizer desktop per OCPU, Smart View for all users, one BICS administrator (I believe the significance of this is that Enterprise OAC is full BICS), and however many OCPUs you buy..  

Metered

Pricing

OCPUs explained

List pricing

Is it worth it?

Wowzers.

Don’t be afraid to correct me

Author: Tony Tauro, Performance Architects

In 2017, we’ve seen an expansion of business intelligence’s (BI’s) scope, changes in consumption, and shifts in the roles of BI consumers and creators. Traditional and fundamental BI practices and processes, however, remain more important than ever.

As a result, the three major market drivers of BI trends so far in 2017 include:

1. Get more from your data
2. Do it faster and cheaper
3. Make your data better

These are not mutually exclusive, but instead tend to reinforce each other and the general direction of BI trends.

1.     Get more from your data

Data is just bytes (or even bits) till someone can process it into information. Ideally, all the data sitting in our data warehouses has already been processed into information…of course, there is always better information if only we could read the data correctly. Data discovery and visualization are currently the hot tools to help us achieve more complete and better analysis of our data.

These tools are especially relevant because of the advent of another hot trend: big data. An easy way to understand big data is to think of the progression from to-do list to contact list to spreadsheet to relational database, and try to fill in what comes next: a solution that can handle data sets that are too big for traditional databases. And we are seeing more and more of such data sets now.

Once upon a time, manual data entry was the primary way to build data sets. Now data is introduced to data storage solutions automatically. Transactions are mostly electronic, and we have sensors producing data as well. It’s no wonder that our datasets are doubling in size every 2-3 years!  Big data tools are getting more and more prominent as companies realize the need to harness the power of this data.

Data discovery at its core is about interacting with your data the way you would with a search engine: ask a question and get an answer. Unlike a search engine, your data discovery solution gives you an appropriate (contextual) answer, considering items such as your role and permissions inside your company.

Visualization is about…visualizing your data, but it’s also about moving beyond the traditional graphs and charts that have always been used for BI. If data discovery is like using a search engine, visualization is a little like Wolfram Alpha, where you can query on a general topic, get in-depth information and find answers to questions you did not even know to ask.

Essentially data discovery and visualization techniques and solutions allow the consumer to create and discover the information needed, which brings us to the next topic.

2. Do it faster and cheaper

Since the days when humans fought velociraptors to win the evolutionary wars, “business people” have fought “IT people” for control of the reporting and analysis (BI) environment. Actually, one of those two things is pure hyperbole, but that is not the point.

“Self-service BI,” while not a new concept, is getting more traction now. While a diverse group, “business people,” are getting more savvy with BI solutions. At the same time, BI environments are getting more complex, making it even more important to get architecture and processes right.

Self-service BI is the concept that BI can be centrally managed, while also allowing “business people” to create their own set of reports, charts, graphs: basically, have their own BI and let IT manage it, too.

The savvy reader will note that data discovery and visualization are also forms of self-service BI, though that is not what is usually implied in general usage of the term “self-service.”

3. Make your data better

Data discovery pushes the boundaries for how we source data, going beyond the limits of the traditional data warehouses and bringing in data from more and newer sources (hence the search engine analogy earlier). this introduces questions about how to control data quality and how to improve data context.

Sometime after the Dark Ages, we came to the realization that the shiniest of dashboards get their credibility from boring old data quality and master data management processes. Transactions (e.g., sales orders, invoices, material movements, accounting documents) are great. They represent action and contain numbers that can be put into reports (like financial statements) and (gasp) glorious visualizations! However, without tying back to master data, the transactions are just business data (not information), and certainly do not provide context.

Ultimately data quality and management is about ensuring that the consumers of the data have a solid set of assumptions to use while translating that data into information. Keeping those assumptions true in the light of growing data sets and sources is a challenge (or opportunity… which one is it?), but is essential for the data discovery, visualization and self-service capabilities to stay relevant.

We discovered a few interesting things about consolidation tasks and HFM 11.1.2.4 at a recent client engagement. First, there is no “keep-alive” between HFM and the Oracle HTTP Server, so anything relying on a keep-alive will time out and cause issues later. Load balancers would need to have their keep-alive times adjusted, and OHS would need the entry for WLIOTimeoutSecs in the mod_wl_ohs.conf file adjusted upwards too.We saw this issue with our install due to Task  Flows that incorporated consolidation tasks. If a task in our environment took more than 5 minutes to complete, a new identical task would be spawned. The consolidation CPU requirements were exceeding the capabilities of the HFM servers because of this. We adjusted the LB timeout and got our test task to not duplicate. However, it only took 23 minutes to complete. We found the OHS timeout (7200 seconds, or 2 hours) in our environment when we ran longer consolidation tasks. We upped the timeout to 16 hours in OHS and have not seen a repeat of the issue.

In addition to the extra task being spawned, the task takes up a connection in the EPMSystemRegistry pool in WebLogic - default is 30, so if you have a LOT of consolidation tasks, you can exceed this value and cause failure for other tasks that require a connection to that pool.

The plan is for Oracle to introduce a keep-alive in HFM in version 11.1.2.4.300, scheduled for release in November 2017.

The post Export Substitution Variables, The PBCS Edition I appeared first on The Unlocked Cube and created by Vijay Kurian.

These days, I find myself working with on-premises applications quite a bit. So, once in a while, it is refreshing to look at what some Planning’s cloud counterpart can do. One of my absolute favorite features that Oracle has rolled out over the last few years is the advent of the REST APIs. I thought……

Continue reading

The post Export Substitution Variables, The PBCS Edition I appeared first on The Unlocked Cube and created by Vijay Kurian.

Data Maps is a great feature in PBCS that allows you to seamlessly move data between plan types and to your reporting databases. You can map smart lists to dimensions in ASO reporting cubes to convert the accounts to dimensions so that you can easily report in Smart View and Reports. It works great when you have similar dimensionality but if you really have extensive mapping you should use Data Management to move the data between cubes. As we use EPBCS, and take advantage of the out of the box content we have hit a few performance issues with data maps. Here is a recent solution we devised to resolve slow transfers of data between the Workforce cube and the ASO Workforce reporting cube.

OVERALL PERFORMANCE

Essbase is not great at extracting members that are stored as dynamically calculated. You will see this if you execute a data map and all of the sudden a quick map is now taking a considerable time to run. Most likely there are dynamical calculated members recently added to your selection. Luckily you can disable dynamically calculated members in the map. To enable this option edit the Data Map and click the Options button. In the window that displays click the option to Exclude Dynamic Calc Members.

Then run the map again and it should be back to what you are use to. However you may be missing some data that you expect to be there. If you cannot discover what the dynamic calc members are, create a data export calc script with the same source point of view and the option DataExportDynamicCalc OFF. Run the script in calc manager and the log tab will inform you what members where excluded from your export. You can then investigate how you can include them in your data map as stored members.

OWP_Headcount Data for Reporting

As your plan and forecast grows you will see that this map will get slower and slower. It is created out of the box for the Workforce Planning module. Its purpose is to move the headcount data from the WFP cube to the ASO reporting cube that is supplied with the solution. The main reason this is slow is because the source of the headcount data is a dynamically calculated account.

The reporting cube and the WFP cube share the parent OWP_Total Headcount. In the WFP cube it is a dynamic calculated parent account and in the ASO cube it is a level 0 account. The data map is not optimally designed, and you cannot select the exclude dynamic calc option because no data will move over.

Here is the solution I found.

Step one edit the data map. Then click the red x to the right of the accounts dimension.

This will move the Account dimension to the Unmapped Dimension section. Click on the hyperlink OWP_Total Headcount and open the member selector.

Clear the current selection and the select every level 0 member under OWP_Total Headcount except OWP_Departed Headcount.

Then on the Target select the account OWP_Total Headcount.

Then click the Save and Close option. You will then get an error message that you cannot save the map because there is an invalid member OWP_Total Headcount.

This error is a defect because in the ASO cube the member is a level 0 and without a formula .

Since that member was created by WFP you cannot edit any part of it. To work around this error temporarily point the map to the account OWP_Compensation Expenses. This should let you save the map.

Now that it is saved navigate to Variables. Create a substitution variable at the All Cubes level called TARGET_HC_ACCOUNT and set the value to OWP_Total Headcount.

Click ok, and then edit the Headcount map again. Edit the account dimension on the target side of the map. Enable the Substitution Variables selector and select the substitution variable TARGET_HC_ACCOUNT.

Now when you click Save and Close the Data Map will save!!!

THIS MAP SHOULD NOW FLY!!!!!!

Take a look at some results from a client app below:

The out of the box method took almost 12 minutes mostly all extracting that dynamically calculated data. With the modifications it ran with the same results in 9 seconds!

CONCLUSION

Dynamically calculated members are great for reporting but they have no place in data exports, and obviously do not belong in Data Maps! I hope that this solution helps you with your implementation of EPBCS or even in your own custom apps.

Thanks for reading. Have a great Memorial Day Weekend and remember:
“A hero is someone who has given his or her life to something bigger than oneself.“ Joseph Campbell

Per the Oracle technical reference for IMPLIED_SHARE the following steps must be performed any time the IMPLIED_SHARE setting is changed in essbase.cfg:

1. Add IMPLIED_SHARE FALSE to essbase.cfg.
2. Restart Essbase Server.
3. Create a new application and database, with the IMPLIED_SHARE setting in place.
4. Rebuild the outline, with the IMPLIED_SHARE setting in place.
5. Reload the data.
6. Run aggregation or calculation scripts.
7. Restart the application.

However, the technical reference does not say what would happen if the above steps are not performed. May be the implied share would not work as expected. However, on one of my recent projects we did not do the above steps and updated essbase.cfg with the implied share setting. We encountered a data loss one on of the Essbase cubes, specifically, data part of a partition area was cleared and some of the data was literally shifted to weird combinations. This event occurred when the Essbase service was restarted. Note that the data loss did not occur every time the Essbase service was restarted.

It took a bit of time to figure out what was going on. I found a document on Oracle support for an issue having the same symptoms i.e. after adding the setting IMPLIED_SHARE TRUE/FALSE value to the Essbase.cfg file and restarting the Essbase service, data is lost or shifted in Essbase.  Oracle support referred me to the same document id. The document id on Oracle Support is 1539305.1.  Per this document, the cause of the issue is an unpublished Bug 14258058 –  IMPLIED_SHARE setting change to Essbase.cfg requires database restructure.

This is a serious issue and would expect that Oracle would add more information to the documentation covering system behavior if the IMPLIED_SHARE settings are done without the required steps.

The Essbase version on which this issue was encountered was Essbase 11.1.2.4.012

The post Essbase data loss and data shifting due to IMPLIED_SHARE setting appeared first on epminsight.