Category Archives: essbase

How are you using the Essbase Outline Extractor?

The venerable Tim Tow and his crew have recently released a next generation version of the Essbase Outline Extractor. As many of you know, this is a tool that has been around for years and is used to ransack a Hyperion cube’s outline an generate a text file. I have talked to people for years that absolutely swear by this tool and attest to it saving their lives.

And yet, I’ve never used this tool in a professional context. I mean, I’ve used to the tool to see that it works and otherwise experiment with it, but I’ve never had an occasion where I had to use this tool to unlock some of my outline data on my Hyperion servers. 

Generally speaking when I have been in situations where I needed the data that was in an Essbase outline, there has been some upstream system that had the data that I really wanted and could be used to build the dimension I needed. So from where I sit (which is apparently an architectural ivory tower, but I digress), the use case for needing the outline extractor is that metadata is trapped in the outline itself.

Business metadata being trapped in the outline isn’t inherently bad. However, I suspect that the propensity to manage the system this way has a strong correlation to the slow and steady migration of Hyperion system management from the laissez-faire finance department to the fortress of the IT department. In other words, finance users had the Hyperion server at their knees and could manage it by shooting from the hip, wild west style – meanwhile, IT wants forms filled out in triplicate in order to dole out EAS access. 

A question for my Hyperion readers and enthusiasts

That all being said, I am curious to hear about your  real world and practical usage of the tool. I am very curious to hear about the context you are using it in, such as:

  • Is it part of the normal automation process?
  • Did you use it to import the data into some other system that would then be used to update the outline (EIS, Studio, etc.)
  • Did you just need the data for something else?
  • Was ODI available, if it was, why didn’t you use it?

I don’t need any particulars unless you care and are able to share them. I am just very curious about the context that you use or have used this tool in. And for the record, there’s nothing wrong or indicative of a bad environment if you have used this tool, it’s just that I haven’t personally been able to use it, owing to having other options available. Please leave comments or email me, thanks! 

Cool Hyperion Infographic

Just a quick post for this beautiful Monday. A colleague of mine, Daniel Poon, has put together a really cool Hyperion infographic showing the development of Essbase over the last 23 years. The timeline lays out when various versions came out and what the major features of the time were.

It’s especially cool for me to see since I am a relative Hyperion/Essbase “newbie”, having started working with it in 2005. At the time I was taking care of a few cubes on a server that had just been upgraded to Essbase 6.5.1. My server hardware was something of a corporate “hand me down” and sported quad 200 megahertz CPUs. Yes, that’s 200 megahertz, not 2000. Sometimes you didn’t know if a calc was just taking a long time (but still working) or had just completely run off the rail. Of course, I know many of you out there have been at this much longer than I have and will have stories about your even more quaint-by-today’s-standards hardware – and I’d love to hear about your first Hyperion server in the comments.

Hyperion Essbase rejected record automation one-liner

I’m just cleaning up some old files here and came across a post from quite some time ago that never got published. Whoops. This was before I officially released the Rejected Record Summary Java routine for analyzing/summarizing a Hyperion data load rejected record file.

So imagine row after row of something like the following:

\\ Member Ac.0170001 Not Found In Database
09 0170001 900 11 .00

\\ Member Ac.0170001 Not Found In Database
09 0170001 904 11 .00

\\ Member Ac.0170001 Not Found In Database
09 0170001 905 11 .00

\\ Member Ac.0170001 Not Found In Database
09 0170001 906 11 .00

You could run the following one-liner on it:

grep \\\\ sample1.txt | sed -e 's/\\\\ Member //' -e 's/Not Found In Database//' | uniq -c | sort -nr

And get something like the following:

24 Ac.0453902
24 Ac.0397511
24 Ac.0171026
24 Ac.0170926
24 Ac.0170126
23 Ac.0909100
23 Ac.0901100
23 Ac.0201220
23 Ac.0170326

Now, hopefully you aren’t getting any rejected records (and certainly not this many, but then again, it’s just test data), but no matter what, your Hyperion automation practices should include regularly inspecting your rejected records, if any, and this might help if you happen to work it in to some automation.

This example of course presumes your Hyperion automation server is running Unix command tools, so alternatively you could install Cygwin or something similar on Windows if you script there. And for complete platform independence, check out my Java library!

EIS, IES, and things that make you go hmm…

This will be another one of those “fun” posts that I get to occasionally do. Every now and then I check out the web traffic stats for this blog and other online endeavors and take a look at who is coming and how they are getting here.

I wrote an article years ago about my experiences with EIS (Essbase Integration Services). It has turned out, hit-wise, to be pretty popular. There’s just not a lot of information about this wonderful little tool out there.

So, naturally one of the search terms for those looking up information on EIS is either Essbase Integration Services or just EIS. Curiously, though, one of the major traffic contributors to the EIS page is actually from people typing in “Boobeis” (presumably a misspelling of “Boobies”). When searching on the former, Google seems to helpfully break up the search term into constituent parts Boob + eis and suggest my site as a result.

So, yeah. According to Google, my EIS posts get two types of visitors: those looking for information on how to build a cube from a database schema and those wanting to see naked females. Hopefully those two demographics don’t overlap… at least during work hours.

Hyperion Essbase wish list: Import a compressed file

I thought this up while attending Dan Pressman’s Kscope presentation How ASO Works and How to Design for Performance, a presentation that definitely appealed to my inner Hyperion geek. Dan did a crazy deep dive on performance tuning with particular respect to loading ASO. He had some pretty bangin hardware to play with too.

Long story short, and many of us have known this for awhile, but there are ways to format your Essbase load files so that they load faster. Basically what you are trying to do is make things easier on Essbase: stream in less data, don’t repeat things you don’t need to repeat, don’t thrash blocks in and out of memory, and so on. That’s all well and good.

The advent and proliferation of SSDs in the enterprise has done wonderful things for Hyperion performance by  eliminating a lot of the performance quirks with rotational media and penalties from fragmentation. But at the end of the day we are still looking for ways to pump ever-increasing amounts of information into our cubes even faster than we were the day before.

For instances where we are loading a file that resides on the same machine as the Hyperion apps/cubes or even across the network, I wonder what, if any, performance benefits are to be had if we had the ability to import a zip file?

Zip files can get awesome compression on text files. They can also have their uncompressed contents streamed. In other words, it’s not necessary to extract the contents of a zip file before you can read the contents (starting at the beginning). In theory, if one achieved moderate to decent compression on their zip file and handed that to Essbase (say with a specialized import data MaxL command), it would be saving time on the disk-read aspect of the data load, at the expense of some additional CPU usage. Many Essbase load operations are disk I/O bound anyway so this seems like a reasonable tradeoff to make.

As an additional benefit or elaboration on the concept, perhaps multiple text files could be placed into the same zip file, perhaps with a “load manifest” or options on the load command, and Essbase would attempt to parallelize the data load to the extent it can. This would likely be an add-on feature once the basic support is in place. In all you would need to augment the data load process with a zip file reader routine (this would be an off-the-shelf library that is quite common), a couple new MaxL import data variants, and an augmentation to the Java API. I suppose you could leave the MaxL command alone and just program the interpreter to look for a .zip extension and treat it accordingly, but it seems like it’d be the better choice to specifically indicate the data load is from a compressed file.

Of course, if you’re loading just from SQL this whole thing wouldn’t apply to you. Loading data files may seem low-tech but it’s incredibly common and often times I prefer it as I have an exact text file to tie back to, if need be, versus a possibly changing SQL data store (but that’s a conversation for a different blog post). This feature would cater to the performance nuts out there – and if Kscope is any indication, there are plenty. I’d be curious to hear anyone’s thoughts on this.

Calling all Seattle Hyperion/Essbase/EPM enthusiasts!

Are you a Hyperion/Essbase enthusiast, analyst, consultant, or all of the above in the Seattle area (and the greater Pacific Northwest for that matter)? This includes Seattle, Bellevue, Tacoma, Federal Way, Everett, Issaquah, Kent, Kirkland, and everything in between (even you, Portland!). I am working on something special and I would love for you to say hello if you are in the area.

Please drop me a line!

Viva La Database Notes

I originally titled this post “The best little Hyperion Essbase feature you’re not using” but thought better of it. So, database notes. Have you ever wondered what that little Notes button in the Excel add-in does? Or rather, what it did (since even yours truly has finally made the official jump to Smart View)?

Essbase Excel add-in Database Note dialog

Essbase Excel add-in Database Note dialog

It is basically a facility for letting the Hyperion administrator set a “message of the day” on a per database basis. This message can be set manually from EAS, set via automation from MaxL, and of course, updated via the Java API. Since 99% of the cubes I am used to maintaining are wrapped with automation, the notes are a a great place to put some information about the automation process, even if I’m the only one that is going to refer to it. I honestly don’t know of any environments where even a significant portion of users is aware of or uses this feature, but I am sure there are some.

Just off the top of my head, here’s some ideas for information that can be put in the database notes:

  • Last cube refresh time
  • Rejected record stats (such as those generated by the Rejected Record Summary tool)
  • Automation stats (duration, errors, etc)
  • Next cube refresh time
  • Notes about business process calendar/schedule
  • Link to cube status information website
  • Timestamp/version information from file(s) used to load the cube
  • Indicate date of archival for an “archive” cube

As an example, I created an automation routine once that depended on files from various divisions. Each division was on its own schedule, which was denoted by an integer, typically between one and 40. It was useful to me to know which was the most recent file loaded in. I programmed the notes to contain this information, as well as some rejected record stats. So, on a day to day basis I didn’t need to refer to the notes very often, but when things went a little haywire, it was a great first place to be able to check for information before having to dig in further.

Continue Reading…

Kscope13 Session Thoughts: Joe Aultman sprinkles magic Groovy dust on the Essbase Java API

There were lots of great sessions this year, as always. I tried to get outside my comfort zone a little bit and take in some new content. I thought over the next few posts I’d shine a little bit of light on a few of the sessions that were notable to me.

Fixing What’s Broken in the Essbase JAPI and Reaching New Heights with Groovy

As a programmer, this was right up my alley, naturally. This session was presented by Joe Aultman. Joe is using Groovy to do some automation that would otherwise be done with the venerable Java API. Joe, if you’re reading this, super cool presentation. In my mind I saw this presentation as being about two things: issues with a particular API and issues with a language – namely a lot of “boilerplate” Java code.

As an aside, there is something of a renaissance happening in the Java world with respect to the JVM. As a computer science guy (go Dawgs!) and general purpose programming nerd I have been keeping abreast of this language and several others, particularly Clojure. In a nutshell, what’s going on is this: Sun (now Oracle, of course) created Java many years ago. Java runs inside the JVM, or Java Virtual Machine. Whereas with a language like C or C++ you might compile your C source code into an executable meant to run on a particular system, in Java you just always compile down to the same code that in turn runs on a JVM. In other words, in theory, as long as you have a JVM available for a platform (Windows, Linux, OS X, etc), then you should be able to run the same old byte code. Hence the original notion, “Write once, run everywhere.” In practice there were and are many quirks to this, but by and large the statement is true – which is why generally speaking you are able to download Java JAR and class files that work irrespective of the underlying operating system (rather than separate downloads for Linux, OS X, and so on).

As it turns out, the JVM is useful for more than just Java. New languages that are able to compile down to code than runs on the JVM are able to stand on the shoulders of giants and leverage an incredible amount of infrastructure that has already been written and battle tested. Some of the more notable languages running in the JVM are Groovy, Scala (kind of a streamlined Java), Clojure (a “Lisp” dialect), and Jython (Python running inside of a JVM).

Getting back to Joe’s presentation, I think it’s fair to say that Joe is enamored with what’s called “syntactic sugar” in the programming world. This essentially means that a language provides features or inherent abilities that reduce the need for verbose boilerplate code. Groovy more or less delivers the goods in this regard. Furthermore, Joe has created some idiomatic Groovy enhancements specifically for the Essbase Java API that reduce the need to include some “clutter code”.

I definitely liked what I saw, although as an apparently diehard Essbase Java API guy I don’t think I’ll be switching anytime soon (you’ll convert me yet, Joe!). If you didn’t make the presentation then definitely give his slides a look. Joe is working through the red tape in his legal department to be able to post his code under an open source license. Joe, if you haven’t picked a license yet, give me a call and I’ll help point you in the right direction if I can.

In any case, nice job on the presentation, man.

In other news, I have been working a little bit on the side on a wrapper for the Essbase Java API that repents for some of its sins and modernizes the usage a bit more. This is tentatively called Java Essbase Antikythera Layer (JEAL). If you use the Essbase Java API at all you might really like how this simplifies your life. It’s unfortunately a labor of love that I can’t dedicate a lot of time to so if you want to help please drop me a line!

 

Hyperion Papercut: Cross-server cube copy in EAS doesn’t carry over data

So, this might not be a Hyperion papercut in the sense that I have been having some fun with. And it’s not something that I have even thought was an issue until it came up today. When you copy a cube onto the same server as the source cube, you get a full copy plus all the data. But when you copy the cube to a different server, you get everything excluding the data.

I’ve always assumed this was by design and have taken it as such. I mean, if the source cube is huge you could potentially do some bad things when you copy it to the target server. But wouldn’t it be nice if EAS just provided a checkbox to copy the cube with the data when you are going across servers? Obviously it knows how much space is available since the destination server can provide that information.

Of course, the process for getting over the data isn’t too onerous – you just export the data (raw) and load that up as-is to the target. All things considered, is this really a big deal? Of course not. But if you approach the construction of your software with a craftsmanship mentality, would the design be at least a little different? I think so.