It’s finally here. SAP released Feature Pack 3 (FP3) for SAP BusinessObjects 4.0 yesterday for ramp-up. Since the release of Business Objects 4.0 almost everyone has been anxiously awaiting the promises of fixes, enhancements and usability in FP3. Yesterday I had the opportunity to install FP3, and I found that there were a few surprises or features I was not aware of. There have been numerous blog postings, presentations and webinars that discussed all that FP3 had to offer. As a result, I had a good idea of what to expect. Now it’s my turn to get hands-on with FP3. The following article will outline my environment setup, the top new features and the surprises I discovered.
My lab quality installation environment was setup utilizing two Windows 2008 R2 server on a VMWARE ESX 4.1 host. The first server hosted the web applications, intelligence tier and processing tier components. Basically everything server related was on the first server. It was configured with 4 vCPU and 16 GB RAM. The second server was deployed to host the client tools, Crystal Reports 2011, Crystal Reports for Enterprise, Dashboard Design, live office, analysis for MS Office and the HANA Studio client. It had 2 vCPU and 2 GB of RAM. I recommend deploying the client tools separate from the server. Patching a server hosting both, can take a very long time and result in a long duration server outage.
Because I had a HANA server at my disposal, I decided to host my CMS and Audit repositories on HANA. This is a new capability of FP3 that appears to work great. I setup the SAP HANA 64bit ODBC client on the first server to allow the CMS to connect to HANA. I created two schemas / users on HANA, one for the CMS and the other for the audit data. Post installation, when I investigated the HANA tables for both repositories, I discovered that there was a mix of both Columnar and Row tables created. This new feature will be interesting as it relates to the manageability of large repositories. However, I would not expect any company to invest in HANA just to host these repositories.
Below is a summary of some of the new features of FP3. There are more, but you can read about all of them in the Feature Pack 3 What’s New guide.
The top features I discovered were the following:
- Explorer received several enhancements in FP3. User can now create and save Exploration Views. These are views that allow one or more charts, tables and facets to be organized on a custom workspace. You can also add a second dimension to many charts to display multiple groupings of dimensional data. There is also a Geo Aware component.
- The LCM (Life Cycle Manager) and VMS (Version Management System) are integrated into the home page of the CMC. There is no longer a need to access the LCM using the http://<Host>:<Port>/BOE/LCM URL. This fixed several issues I had in base 4.0. In 4.0 I was not able to manage the system name on the logon page nor utilize Active Directory authentication.
- The CMC has a new option to allow the administrator to delegate access to the various “Tabs” or links on the CMC Home page. For example, I can prevent LCM users from accessing “User and Groups”, “Applications”, or “License Keys” when accessing the CMC home page. This will significantly improve the manageability of the system when assigning delegated rights to power users and lower level administrators.
- The CMS and Audit Database can be hosted on SAP HANA. This will be a great option for environments that need to scale their repositories while maintaining peak performance of the system. It will also allow you to create better analytics over the audit data.
- The Administrator account password cannot contain some special characters. This only seems to be an issue during the installation. Post Installation I changed this password, utilizing special characters, and found no issues. Hopefully SAP will create a note to explain this new behavior.
- There is now a Semantic Layer Java SDK. This will allow partners, developers, and 3rd parties to create applications to manage Universes. I can already think of a few applications or programs where this will be beneficial.
- Analysis for OLAP now supports SAP HANA as a direct connection source. However, HANA Hierarchies are not supported. The functionality is not enabled by default but you can follow my article here to enable it.
- The upgrade management tool has options to make it more like the Import Wizard. This was a much needed addition. Migrators can now choose whether universes or connections on the target are overridden. It also allows you to choose how a user’s favorites and inbox documents are moved over. Migrations from previous version have been painful using this tool in 4.0. This makes migration a little better. However, I hope that someone at SAP will get a clue and make sure that “ALL” the options from the Import Wizard are carried over to the upgrade management tool. I would also like to be able to use this to move from one 4.0 system to another 4.0 system. The LCM is just not my favorite tool when migrating 10,000+ objects during an Enterprise level migration from a DEV environment through a PROD environment.
- The LCM has a few other enhancements as well. You can now promote a user’s inbox and favorites. There is a test promote filter and LCM Job versioning. I also found that you could import a .biar file that was created with the .biar command line tool (biarengine.jar). This will help with my low tech backup solutions but I still can’t pick and choose the object to import. Hopefully SAP will also get a clue on the biarengine.jar options as it make for a great scripted backup tool. There are several other enhancements in the UI as well.
- The Dashboard Analytic Server and Dashboard Server have been removed from the CMC. These services were the holdover from the XI R2 and XI 3.1 performance management stack. They have been replaced with the “Java BI workspace bundle”. I don’t know exactly what that means but it sounds like they are imbedded in the Java Application server (Tomcat) now. If you know, please leave a comment.
- Crystal Reports for Enterprise (CRE) can now connect to ODBC, JDBC and other data sources without the need for a Universe. This helps to bridge the gap between Crystal 2011 and CRE. There are also new charts, drill down options and bind options to report objects.
- Dashboard Design can now directly bind to BEx and new components to work with Hierarchies. On the development side, Adobe Flex 4 is now the only supported version for custom add-ons.
- The IDT (Universe) tool can now create connections directly to SAP ERP, Essbase OLAP cubes, and Apache Hadoop HIVE. There are also new options to bind additional LDAP attributes to Row and Column level security. These attributes are created in a new section of the CMC called “User Attribute Mapping”. This should help enterprises maintain both their authentication and data security in a central source.
- Web Intelligence has several new features. There are new charts, better BEx integration, and several other minor changes.
My overall assessment of FP3 is that is has most of the options that 4.0 was lacking in comparison to XI 3.1 and to customer’s expectations. It also has made integration with BW a little better. Legacy BusinessObjects customers will find several enhancements to make the transition to 4.0 much better than it was in September 2011 when the 4.0 GA was released. I have a suspicion that SAP released 4.0 under mounting pressure to better integrate with BW but forgot to mention that it was not exactly ready for legacy BOBJ customers. Feature Pack 3 seems to have filled the gap between XI 3.1 and 4.0 for legacy BOBJ customers. However, both SAP ERP and non-SAP customers will find plenty to cheer about with FP3.
Doc talks about Java Bundle : http://help.sap.com/businessobject/product_guides/boexir4/en/xi4_whats_new_FP3_en.pdf
I saw that as well. I was hoping for a more technical explanation though.
Hi Jonathan,thanks for this great review
Are there any news about Webi reports dealing with the #UNAVAILABLE# message when using Bex cubes ?
There are several listed fixes to the #UNAVAILABLE# issue in Web Intelligence. Look for the following ADAPs in the Release Notes:
Release Notes: (Requires SMP Logon)
Great post. I have additional comments on the IDT SDK included in FP3, according to our meetings with SAP, the APIs is limited to those that support their new multi-tenant onboarding processes. We are building .unx’s but can’t generate metadata like we did with the Universe Designer. Also the IDT is a mess, any updates on it from your testing are welcome!
Would love to hear more about LCM and VMS when you have time. Having some issues with job server children failing to terminate normally and LCM UI requireing a Tomcat restart in SP2 Patch 10.
Anything new with respect to SharePoint Integration?
The new REST API is a welcome addition, if you drop your custom sdk app on BOE’s tomcat 6, or use a proxy from a second tomcat, much easier to do log ins, folder/category/inbox retrieval. This is huge for shops doing custom apps.
I have not had a chance to work with the SPIK in FP3, but I had several open notes (Issues) that were slated to be fixed in FP3. Once I have a chance to test it, I’ll post feedback.
If we upgrade from BOBJ 4.0 sp02 Patch 14 to FP3.Do we have to do end to end report testing for Webi,crystal & dashboard again as we did when we upgraded from BOBJ 3.x to BOBJ 4.0 sp02 Patch 14.
In my experience updating from one Support Pack (SP) version to another SP version caries more risk then when applying patches. Due diligence would dictate that that you validate the SP in Development prior to moving the SP into Production. Concerning the testing of each report type, it would be wise to test the top 25% most critical reports of each type. If you are using Crystal, Web Intelligence, Dashboard Design (Xcelsius) I would pick the top 25% from each area and test. In my experience there are some issues and subsequent risks as a result of the SP upgrade due to the new features and functionality. However, I would not expect the testing processes for a SP upgrade to be as thorough as the upgrade process from one major version (XI 3.1) to another (BOE 4.0).
Hi Jon, Can you expand a bit on the cleanup process after installing SP4. I see my Installcache is at 10 gigs now and I found out the hard way that just deleting the files are a really bad idea.
According to SAP, the only way to remove the patch’s installdata cache is to un-install the patche before applying the newer patch. You can un-install a patch using the Windows (Add/Remove Programs) or Unix (Using the modifyorRemoveProducts.sh ) script. I discussed this in a previous article as well http://wp.me/p2868w-q . If you delete the directory the update processes will no longer be possible but it sounds like you found that out the hard way.