The release candidate contains the new front end with all features the final release will have and some new demo reports, which demonstrate the new layout possibilities of 3.0.

In difference to our original plan, the Workbench still contains a report editor and users will be able to create and edit reports in the web front end OR in the Workbench.

Complete, 32-Bit (Windows Installer), Version 3.0.rc, 08.02.2012, 74.9 MB
Workbench, 32 Bit (Windows Installer), Version 3.0.rc, 08.02.2012, 26.1 MB

Server (Linux Installer, RPM) , Version 3.0.rc, 08.02.2012, 69.6 MB

Frontend (WAR-Archive), Version 3.0.rc, 08.02.2012, 19.9 MB
Backend (WAR-Archive), Version 3.0.rc, 08.02.2012, 4.9 MB
Workbench (WAR-Archive), Version 3.0.rc, 08.02.2012, 16.0 MB
Complete (EAR-Archive), Version 3.0.rc, 08.02.2012, 40.5 MB
instantOLAP Console (JAR-Archive), Version 3.0.rc, 08.02.2012, 0.1 MB

• A new block structure was introduced:
  • There is no difference between inner and outer blocks and they can be nested as deep as wanted
  • Each block can contain a selector bar, attached to the top, bottom, left or right of the block.
  • Blocks will load their content individual and parallel to other blocks
  • Tabbed blocks load their content on demand when switching to another tabulator
  • The orientation property of blocks was replaced by the new block formats “vertical”, “horizontal”, “tabbed” and “animated”
• A new designer mode was added and replaces the prior pivot-designer:
  • The designer offers different panels for different report elements, for the report itself, blocks and queries The block designer offers a couple of block designs
  • The query designer allows to add dimensions and levels to axes with a single mouseclick
  • With the table designer, you can switch blocks into tables by using one of  the preformatted layouts
  • The chart designer turns a query into one of the predefined charts
• A new edit mode allows to edit reports inside the browser. In difference to the simple design mode, the editor offers the same ability to build the same complex reports and to edit all existing properties – like in the workbench.
• The frontend has a new look&feel

• The query editor was adopted to the new block structure

28.06.2010: Release 2.6.1 Update 3

• Two new properties (“First Day Of Week” and “Min Days In First Week”) allow to fine-tune the TimeKeyLoader – e.g. you can generate ISO compatible calenders now
• Instead of “= NULL”, the SQL Generator now generates “IS NULL”
• Faster communication between the server and the frontend or Workbench
• Better detection of the return type for the LOOKUP method (which is eg. important when using a LOOKUP in the “Options” property of a selector)
• Bugfix: Dimensions and offline stores with the same name could cause trouble in distributed environments

• Dynamic loading of suggestions for the SuggestSelector – the number of possible options is nearly unlimited now
• For SuggestSelectors, the correct text is now displayed when exporting a report
• Better caching-strategy for files, the browser will immediately reload changed resources from the repository
• Better performance for the ExpressionEditor, the model-tree at the left side now loads async
• Bugfix: The default of a HierarchyTreeSelector was not visually selected if it was another key than the root key

• The default Tomcat installation is now configured not to expand the WAR archives for easier updates

We cleaned up and gave instantOLAP 3 a fresher and clearer design with less buttons and tabulators but more functionality. Not much to write here, just have a look onto the following screenshots (of course, the appearance can slightly change until the final release).

instantOLAP 3 allows to interactivly change content and the look&feel of your reports while watching them. A couple of new “designers” can be opened while watching a report, including a designer for the pivot-table content, a designer for the table-layout, a designer for charts and others.

The Query Designer allows to add axes, facts and selectors to your report with single mouseclicks. You can also rearrange your reports, add Top 10 filters and do many things more inside the designer without having any knowledge of instantOLAP queries.

The Table Designer offers a large number of predefined table designs. Simply change the look & feel of your report by selecting one of them. Of course, after you selected on the designs, you can change it afterwards by changing the table properties.

The Block Designer is very similar to the Table Designer but changes the look & feel of the containing blocks. Like the Table Designer, it offers a number of predefined layouts where you can choose from.

Another very useful is the Chart Designer. It automatically turns your query into one of the predefined charts. A large number of chart layout are offered, including 3D versions of every available chart type.

First of all, we broke the fixed block structure of queries and created a new, more flexible alternative. Now you can nest as much blocks as you want and as deep as you want. E.g. it is possible now to nest complete reports in tabulator block and to build huge executive reports which contain all neccesary data in one page.

Also, the selectors are no longer limited to stay at the top of the query, they can also be placed in any other block and every block can have its own selector. You may even place the selector at the left, right or bottom border of your blocks.

Of course, the old query structure will automatically converted in the current one and every existing query will be compatible with version 3.

We also introduced two new block types, the “IFrames” and “Query-Imports”. IFrames allow to display other HTML pages within your reports and to enrich them with other content. Query-Imports display other queries inside the importing query as if they were part of it itself.

And – you can include other queries as a block by using the “Query Import” format. This allows to reuse and to build dashboards by putting multiple queries together!

Another great change is how the queries load their content: Until now, every query was executed and displayed as a monolithic block, but now every block can update its content without the need to refresh other blocks. E.g. drilldown inside a table can become much more comfortable, because only the block containing the table has to be refreshed.

Also, blocks now do not load their content until it is needed. E.g. tabbed blocks will only load the data of their visible tab and dynamically load the other tabs content when the user selects them. Additionally, blocks can be collapsed and only load their data when the user expands them. This allows to build fast dashboards which load their data only on demand.

A much requested feature was to use multiple models in a single query. In instantOALP 3, each query can use a different model and you can generate reports with data from multiple models. Even the selectors work with different models if the dimensions have the same name. The only limit is that a single table or chart may only use a single model.

This is surely one of the greatest features in instantOLAP 3: This version will contain a fully functional and powerful query editor, even more mighty than the editor included in the Workbench of previous versions.

The new editor used the same layout and look & feel like the Workbench version – including drag & drop – but contains a couple of new improvements:

• First of all, it reflects the new block structure, this means you can nest blocks as you want and create selectors everywhere. A large number of new context menus will help you to arrange blocks and to change their format.A
• Also, the new Designers from the Query Viewer are available in the editor. This means, you can also design tables or turn them into predefined charts within the editor with a few mouse clicks.
• The property editor has the same functionality as the Workbench version but now groups the properties by themes to give you a better overview.
• The query editor will use the powerful expression editor of the portal which allows a much more powerful editing of expressions, including Top 10 requests, dimension searches, saved expressions and much more. We also improved the expression editor between version 2.6 to version 3.
• A zoom function allows to keep the whole report in sight when editing large reports.

The new editor will replace the existing version in the Workbench.

instantOLAP 3 will be available at the end of 2010. A beta release is scheduled for August 2010 and existing customers are welcome to test the new version!

• Expression now accept the escape characters \\’ or \\” to use delimiters in strings
• TimeKeyloader Attributes are now also adjust the years to weeks when the owning KeyLoader has its property “Adjust Years To Weeks” set to “true”
• Better performance for list reports
• Better performance for the MATRIX function

• Large logos are scaled to fit when exporting PDF documents
• Bugfix: Interval selectors now accept two equal default values
• Bugfix: The SuggestBox selector now selects values when using the tabulator key
• Bugfix: Better screen layout for the Automation Manager when editing email automations

• Better performance when rendering configurations with many dimensions and cubes

• The automatic synchronization of dependent and indirect-dependent dimensions now ignores the cron patterns of the other dimensions and always synchronizes them.
• KeyLoaders now always set the name of their target level, even if they don’t load any key from the datasource. This only works for loaders nested under the root key of the dimension structure.
• Bugfix: A possible deadlock of model data files was eliminated. This deadlock could have cause non-removable locks on models.
• Bugfix: TimeKeyLoaders with combined weeks and years in their pattern didn’t correct the year properly for end 2009 / beginning 2010.
• Bugfix: Stores using Where-Only mappings threw exceptions while building.

• The model tree of the Analyzer now offers some standard expressions like the Query Editor.
• The server now disallows the Browser to cache the startup HTML file of the Portal.
• Bugfix: Rendering-problems with overlay charts have been fixed.
• Bugfix: Reports with complex selectors did sometimes not appear before using the “Execute” button.
• Bugfix: The Expression Editor can now be closed with empty expression without displaying an error message.
• Bugfix: The dropdown fields of classic Hierarchy selectors are now expanded to the full selector size in IE now.
• Bugfix: Interval selectors did not display two equal default values correctly

• The Workbench offers a new property “Adjust Year To Weeks” for TimeKeyLoaders which allows to disable the automatic year correction for patterns combining both weeks and years.
• Bugfix: A possible deadlock while displaying properties for a currently loading model was removed.
• Bugfix: The webstart version of the Workbench killed existing named sessions even if the user denied to kill them.

You can download the version from our downloads page. You will also find a update of our manual, reference and online documentation on our documentation page. Read this article to learn some details about version 2.6.1 and to read its release notes.

The new version contains new and faster renderers for the HTML, Excel and PDF output. In difference to the output of the PDF renderer, which generates output equal to its previous version, the HTML renderer now generates a much more compressed result. Especially for the Internet Explorer, this reduces the load and display time for HTML reports significant.

Additionally, the new HTML renderer offers some new formatting options:

• Scrollable tables with fixed headers: Before in the only available in the analyzer, tables in normal reports can now also contain fixed headers and scrollbars. The new HTML renderer will automatically display every table as scrollable table which is placed inside a block with fixed width and height properties set.
• Free positioning of blocks: For each block you may assign a fixed X/Y position and place them everywhere on the screen you want to.
• The automatic expansion of blocsk to make them all equal with can be manually switched of now.
• Charts are now automatically expanded to the size of their owning blocks and therefore zoomed when blocks are expanded because of other and larger blocks in the report.

The Excel renderer also contains some new features. E.g. automatically generates foreground and background colors for negative numbers or exceptions are now included in the Excel export and all original datatypes of the cells are preservered when exporting a document.

The report viewer now contains a new “Print preview” button which shows the report in a printable PDF version directly inside the browser. This allows to print a report without exporting it into the PDF format before. To use this feature, your browser must have installed the Adobe PDF plugin.

A new button in the upper right corner of the portal allows to hide the explorer bar at the left side of the portal and to gain more space for the reports or other viewers.

The expression editor, which is used e.g. to edit formulas or header iterations in the Analyzer, has widely changed and offers a number of new dialogs for searching, sorting, navigating etc. in dimensions without using the keyboard and manually typing expression.

There is a detailled description of the expression editor in our also new manual for the version 2.6.1.

When creating a store you can now choose between two storage models for their data: The known “Condensed” format and the new “Indexed” format, which is much fast in building high-dimensional stores but slower in reading data out of stores. With a good combination of condensed and indexed stores, you can now create offline version of really high dimensional cubes by creating the most used combination of dimension as condensed stores and additionaly storing the “complete” view of the cube in a indexed store.

To learn more about the new storage type, read the new store property description in our new reference for version 2.6.1.

22.10.2009: Release 2.6.1

Server

• A new function ISLEAF was added
• Less logging, e.g. for unknown tables while editing configurations
• Now match filters for cubes can contain expressions of any complexity
• Imported dimensions now use less memory and CPU than before
• The limits for number keyloader have been lowered, both limits can be negative now and the start-limit has no longer to be less than the end-limit
• Unknown keys are now mapped to the root key when loading a cube into an offline store (but they still generate warnings). This brings the behavior closer to online cubes, because the root key of dimension always reflect the total of the database content, independent of all keys being mapped to the dimension.
• A new storage model for stores, the “Indexed” storage, allows to create stores for high-dimensional cubes. Though the indexed storage is slower than the previous “condensed” model, its build time does not grows exponential with the number of included dimensions.
• Line breaks in escaped SQL statements (@”…..”) are now allows and will no longer irritate the Database server.
• Unmappable keys are mapped to the root key of a dimension when loading a cube into a store. This equals the root value of the dimension to the online version of the cube.

• Bugfix: A possible deadlock while using a corrupt dimension was fixed
• Bugfix: Mappings with “mode = select” did not appear in the result sometimes
• Bugfix: Multiple mappings for the same fact now also work with DLOOKUP
• Bugfix: The GROUP BY clause was constructed wrong if a statement contained
  equal expressions and used the “Group by indices” option.

Web-Frontend

• New and faster renderers for HTML, Excel and PDF output
• A strong compression for the HTML output was added, now cells with similar styles are clustered into CSS classes
• A new “Print preview” was added to the report viewer now. The preview displays reports as PDF files inside the portal and allows to print reports without exporting them before.
• User can now hide the explorers at the left side of the portal
• Now ThinUsers can also delete their private files and snapshots
• Now negative numbers are also colored red when exported to Excel
• New placeholders for filenames in automations have been added (%USER% etc)
• Empty rows and columns are now removed from the result set and no longer displayed as rows or columns with the height or width 0
• Input fields and links became much more performant, reports with thousands of input fields are now possible
• Assertions are now also displayed in Excel and PDF exports
• The selected tab in the tabbed orientation is now highlighted
• Dialogs are movable now
• Window titles are limited to 30 characters now
• Improved drag and drop behavior which now also works if the user drags an item very fast or if the browser reacts slowly
• The table editor in the Pivot viewer does no longer disappear when changing the format to “chart”.
• The “save as” dialog now also shows where the user has no write access if one of the subfolders is accessible.
• The Excel export now preserves the data types and exports all texts as text, even if it is convertible to a number.
• There is no need to create access rules for the bookmarks or snapshots folder any more.

• Bugfix: Headers nested under spanned headers are now no longer too wide or too heigh
• Bugfix: The snapshot and automation folders no longer need manual set access rights
• Bugfix: The font-color was not exported to Excel correctly
• Bugfix: Removing sorts in the Analyzer now works
• Bugfix: Like for other formats, the XML export is now only offered after the report was displayed
• Bugfix: Long URLs caused problems in IE and are now replaced by simple numbers
• Bugfix: EMail attachments always had “.pdf” extension when send directly out of the report viewer.
• Bugfix: Sorting a row or column in the pivot viewer (using the context menu of a header) now also works with scrolled tables.
• Bugfix: The content of search selectors and input fields can now be selected.

Workbench

• The Workbench no longer locks when waiting for a model load or reload
• The ERM view became faster and uses a cached version of the database-schema when displaying the links
• Invisible keys and facts now appear in the cube editor
• Invisible dimensions now appear in the dimension preview
• A new background property was added for queries
• The column widths in the property editor are no longer reset when a new element is selected
• The previewer now also works on all Windows 64 bit systems

• Bugfix: The password for automations can now be edited and changed

Console

• A new argument now allows to kill an existing session for a named user

If you haven’t read about 2.6 before, have a look into our previous article Whats new in 2.6?. At a glance, the version has some nice frontend improvements, it is faster and uses less memory, it is easier to use and it contains less restrictions. Read the release notes in this article for a complete list of all improvements and fixes.

And version 2.6 contains a free administrator license for each new or existing customer. This also means, you will no longer need to request an evaluation license if you want to download and test instantOLAP. Just download the software, install it and start using the it without any time limit.

19.08.2009: Release 2.6.0

Server

• The rebuild of dimensions and stores (former offline cubes) is now synchronized and executed in the same transacation, now stores are immediatley rebuild if any of their dimensions change and are publish at the same time as their dimension. Therefore, stores can now longer become “outaged” as in previous versions.
• The size of ToDo-lists is no longer limited as before, the “Query too complex” can no longer appear. Also, the ToDo processing now is 100% faster and uses exponential less memory.
• Imported dimension use less memory now and are synchronized faster. Imported models are therefore automatically started whenever an importing model is started.
• Models can now be shut down while they are synchronizing their dimensions or stores
• Less overall memory consumption
• File caches are now automatically deleted an re-initialized whenever they are corrupted
• New MDX-Functions and operators: UniqueName and StrToElement
• Now, calculated MDX members with other return types than “Set” are also supported

• Bugfix: The JOIN operator returned an empty result if the first argument was a single value and the user had no access-rights for this value
• Bugfix: The stores can now also contain non-aggregated facts and aggregate  them manually
• Bugfix: Sometimes the ROWNUM function displayed “0″ in each row

Web-Frontend

• Two new selector types have been added to the frontend: Calendar ranges and SuggestBox (which offer possible options while the user types into the box)
• Selector groups have a new property “Popup” which allows to convert a group into a space saving popup selector
• Negative values are now also colored red in the Excel export
• Better performance for Input selectors
• The repository no longer collapses when beeing refreshed
• New properties have beed added for overlay charts and allow display multiple ranges within the same chart
• The metrics for tables (needed cell-width and heights) are now calculated more accurate
• A fixed snapshot folder can now be set up for the frontend. If used, the users automatically save their snapshots in this folder without beeing asked for a file path before (like in version 2.2)

• Bugfix: The Excel export threw an error if the left padding of a cell was defined as double value
• Bugfix: The current value of input fields was not exported to PDF or Excel files
• Bugfix: Sended files did always have the extension “.pdf”, independet of their type
• Bugfix: The save as dialog now only shows folder with admin access for the current user
• Bugfix: The login dialog now also forgets the last user name even if the “remember user” option is not selected.
• Bugfix: The access-rights for the bookmark folder “.boomarks” are now ignored, every user can store data here
• Bugfix: The bookmark editor was messed up for arguments with multiple values
• Bugfix: The input selector now matches all result-elements again its options
• Bugfix: The Browser shows no longer endless “loading” after an empty query was executed
• Bugfix: The frontend offered edit rights for a folder if only some of the contained files was editable but not the folder itself

Workbench

• A complete new structure for offline cubes was introduced: Now cubes can contain as many “stores” as you want, each with a different combination of dimensions and fact. The configuration of stores is much easier than before,  just add dimensions and facts using checkboxes to the store and determine  their aggregation type (if wanted) – thats it.
• Cubes can now optionally still exist even if an offline version (store) of  them is active. In this case, the online cube is listed *behind* its offline versions.
• A new “Notes” was added to the configuration editor and allows administrators  to add notes to their configuration files
• The process manager now shows the current progress of a progress
• A new “Edit” button was added to the model manager which allows to open the configuration editor for a specific model directley from the manager
• The model-info now displays the number of warnings and error for a store in the cube tab
• The auto mapping for dimensions is no longer stored in the dimension itself and still keeps valid even if a dimension structure has changed
• Virtual tables (aliases) and virtual columns (expression) now have their own  symbol in the database explorer to make them visible for the administrator
• A new refactoring was added which deletes all references (links etc) using an  table-alias after the alias was deleted

• Bugfix: If you start a model out of the model-info panel, the panel now waits until the model was loaded
• Bugfix: The preview inside the Workbench displayed “Unknown host” if the  server wasn’t able to resolve its own domain-name.
• Bugfix: The mousewheel now also works while the mouse stands over the mapping panels in the cube editor
• Bugfix: Drag&Drop did no longer work after a user was moved in the user  manager
• Bugfix: When a configuration was changed and saved, the new dimension version have published to the productive model. This still happens but cannot block  the Workbench any more when waiting for a model lock.
• Bugfix: The mouswheel function in the ERM editor now works vice versa (moving the wheel forward zooms into the chart)
• Bugfix: The properties of the currently selected element are still displayed after clicking onto a database or repository element

If you are a new user, you can now download and use instantOLAP without the need to request an evaluation license. If you are an existing customer, you will have an additional and free license in your license pool after you updated to 2.6

The free license is an unlimited administrator license and never expires or contains any other restriction. With this license, you can build OLAP models and reports, evaluate instantOLAP for an unlimited period of time or set up your single user OLAP server without any costs.

instantOLAP 2.6 will be available at the end of september 2009, but you don’t have to wait and can download the preview version right now. Use this link to download and test the preview version.

Two new types of selectors were added to the frontend: Interval calendars and suggest boxes.

Interval calendars allow to select a time-range (from date / to date) using the calendar known from the original calendar selector. The interval selector shows two calendars, on to select the start and one to select the end date of the interval.

The suggest box offers the user valid input options while he types into its text box. This is well known from other web 2.0 applications.

Selector groups can now be turned into popup selectors by setting their new property “Popup” to true. Popup selectors only appear as a small and space saving button. By clicking onto this button, the user can open or close the selector group and use its contained selectors.

One of the main features of the new version is the much easier handling of stores. Unlike before, where cubes could be converted into offline cubes by setting some properties and setting the wanted dimension into the offline state, the cubes now can contain one or more store. Each of this stores defines a number of contained dimensions and facts, whereby different stores can contain different combinations of dimensions and / or facts and there is no need to copy cubes for different combinations in future.

The configurations of stores with the Workbench became easier and more intuitive: Simply add a new store to a cube by using its context menu and add dimensions and facts to the new store by clicking onto it. The Workbench will now automatically do all needed settings for your, e.g. by loading the lowest level of a dimension if you want to roll it up.

The rebuilding of dimensions is now more connected to cubes than before. Everytime a dimension is rebuild, all stores being dependend on this dimension will be automatically also rebuild in the same transaction if the dimension has chaned. “In the same transaction” means, the user will not see changed dimensions and cubes before both are finally loaded and commited. The advantage of this new transaction model is, that cubes can no longer reach the “outaged” state if dimension changed and the cube was not scheduled to rebuild directly after.

The internal ToDo lists – which instantOLAP uses to determine where to load which result cell from – became much faster and less limited. The overall performance improvement is round about factor 2 when generating and resolving the list. And, because the new implementation nearly uses any memory, there is also no limit for the maximum size of ToDo lists from version 2.6. Therefore, the “query too complex” message, which appeared when building reports with too much cells, will no longer appear.

The beta version of the connector will be available for free as a part of the following releases unless it reaches the production standard in version 2.6.

You can download the new release 2.5.3 from our download page. Beside the connector, it contains a small number of bug fixes and performance-improvements.

The instantOLAP XML/A connector supports XML/A 1.1, a web-service based query protocol which is supported by a large and increasing number of clients. The new XML/A server is part of the usual instantOLAP server and is reachable under “http://<server>:<port>/iolapWS/xmla“. It uses basic authorization (which must be supported by the client) and only operates with user accounts having the “iolapPowerUser” role or upper (from version 2.6 a new and different role will be mandatory).

Windows clients using ODBO as their connection standard can connect to the instantOLAP XML/A connector using an ODBO provider which talks “XML/A” to the server. There are currently two bridges available, the commercial O2X Provider from Simba Technologies and the Microsoft SQL Server 2008 provider from Microsoft.

The connector transforms the instantOLAP concepts into the XML/A world:

• instantOLAP models are visible as “catalogs” for XML/A, schemes are not supported.
• Each dimension contains a single hierarchy with the same name as the dimension. Future versions of instantOLAP will support multiple hierarchies.
• Keys are visible as “members” for XML/A
• Key-Attributes become properties in XML/A. The member properties MEMBER_UNIQUE_NAME, MEMBER_CAPTION, CHILDREN_CARDINALITY, PARENT_UNIQUE_NAME, LEVEL_NUMBER and UNIQUE_LEVEL_NAME are supported and delivered for each key, even if they are not defined in the model.
• All queries via XML/A use a single cube, which must be specified in the query. Future versions will allow to query all cubes or groups of cubes.

Both XML/A and ODBO use MDX as their query languge. MDX was invented by Microsoft and is comparable to SQL but specialized for multidimensional queries. The beta version of the instantOLAP XML/A connector supports the core syntax of MDX and a number of operators and functions.

Supported MDX syntax in the beta version:

• SELECT statement (with an unlimited number of axes and dimension properties)
• WHERE clause
• WITH MEMBER clause
• WITH SET clause
• Operators: (), {}
• Functions

Supported functions in the beta version:

• AddCalculatedMembers
• Aggregate
• Ascendants
• Children
• CrossJoin
• Descendants
• Distinct
• DrilldownLevelBottom
• DrilldownLevel
• DrilldownLevelTop
• DrilldownMember
• Except
• Hierarchize
• Intersect
• Join
• Union
• VisualTotals

Before you can connect instantOLAP from Excel, you must install an XML/A compatible ODBO driver on your local workstation. In this example, we use the O2X driver from Simba. After you successfully installed the driver, you can start Excel and create your first pivot table:

1. Click into an empty cell of your worksheet and use the “Pivot Table” button from the “Insert” tab to create a new table at this location.

2. Now, Excel wants you to define the datasource for your pivot table. Because instantOLAP is an external data deliverer, you must choose the “External datasource” option and use the button “Choose connection” to continue.

3. The following dialog shows all existing datasources stored in your current workbook and on your computer. Because this is the first time you connect an instantOLAP model, you must use the “Search for other elements…” button to create a new connection. After you successfully created an connection, it will be displayed here and you don’t have to redefine it.

4. This button opens the Explorer and shows the folder where all stored connections (ODC-files) are located. Now you can use the “New source…” button to create a new ODC file (and datasource):

5. The new dialog wants you choose the type of the new datasource. It contains a list for all standard types and an element “Others” for unknown types like instantOLAP. Choose this option and click the “Next >” button.

6.  Now you can choose the driver for the new connection. Here you must choose and XML/A compatible driver like the SimbaO2X Provider or the Analysis Services Driver 9 or 10. Then click onto “Next >” to open the next tabulator in the dialog.

7.  The next page wants you to input the URL of the server and the username and password for the connection.

The URL for an instantOLAP server is usually “http://<address>:<port>/iolapWS/xmla”. E.g. this example connects to an instantOLAP server running on the local machine and on port 8080. The username and password are the same as when you log into the instantOLAP Web Portal or into the Workbench, but you need at least to be a “Power user” to connect to the XML/A server.

We also suggest to save the password and to select an initial Catalog for your connection (a catalog is the same as models in instantOLAP). After you successfully entered and tested you connection, you can use the “OK” button to continue.

8. Now you can choose the cube of the model on which you want to execute your queries. Here, instantOLAP shows all cubes and an addition cube named “ALL” which acts as a hypercube containing all other cubes (this is also the cube all instantOLAP reports use). The use the “Next >” button to enter the next page.

9. Here you can edit the filename and display name of your new connection. Also, if you want to store the password in the connection, you must check the “Save password” option here, again.

Now finish all dialogs. If everything went right, a new and empty Pivot table should be visible and you can add facts or dimensions to it by clicking them in the field list at the right border of your Excel sheet.

08.06.2009: Release 2.5.3

Server

• A new XMLA server is available as backend for XMLA and ODBO clients (e.g. Microsoft Excel)
• Supported MDX Functions: AddCalculatedMembers, Aggregate, Ascendants, Children, CrossJoin, Descendants, Distinct, DrilldownLevel, DrilldownLevelBottom, DrilldownLevelTop, DrilldownMember, Except, Hierarchize, Intersect, Join, VisualTotals
• Better performance for File-Caches
• Bugfix: The Omit-Factor of dimension mappings is no longer used for list reports

Web-Frontend

• Patterns (%USER%, %TITLE% and %NAME%) for mail topics introduced
• Bugfix: Problems with included images in PDF-mails solved

Workbench

• Other levels with the same name in the same dimension are no longer automatically renamed when a level is renamed
• Visible and Description-Properties introduced for Keys, Attributes and Cubes
• Bugfix: A missing java archive for Linux was added