The Essbase Book Review to End All Reviews: Developing Essbase Applications

I received my copy of Developing Essbase Applications – Advanced Techniques for Finance and IT Professionals some time ago and spent every spare minute I could find eagerly devouring the book. I was eagerly anticipating the arrival of the book to my doorstep, hoping and wanting it to be the Essbase book to end all Essbase books. While it has its flaws, it is a must-have for the dedicated Essbase practitioner’s bookshelf. As I always say in these reviews, there are precious few Essbase books out there, so anything that helps the cause is welcome as far as I am concerned.

Developing Essbase Applications sports an impressive and diverse array of contributors. Most of the names are easily recognizable to anyone that has taken a break from writing a calc script to seek out an Essbase blog article or Google for some help on the Essbase forums. Seriously, check out this who’s who list of Essbase folks that helped create this thing: Cameron Lackpour, Dave Anderson, Joe Aultman, John Booth, Gary Crisci, Natalie Delemar, Dave Farnsworth, Michael Nader, Dan Pressman, Robb Salzmann, Tim Tow, Jake Turrell, and Angela Wilcox.

While this is probably the greatest strength of the book, it also inevitably contributes to my complaints about the book, which is not so much that it is a book, but more, a collection of books. This divide and conquer approach is probably the only realistic way to get so many people together to create such a thing, and the necessity of doing so is perhaps a reflection of the increasing breadth and scope of the Essbase ecosystem itself.

So for those of you looking to start on page 1 and work your way front to back (as I am want to do with a good programming book), you can do that, but it’s not necessary since the chapters don’t really build on each other. They just really do feel like a dozen books bound together, each with their own table of contents and style. The voice of each chapter author really shines through. In this regard, the book can be thought of as more of a reference, and indeed, some chapters are so packed full of information that there’s no reasonable way to absorb it all in one reading.

That all being said, what about the actual content of the book? It’s impressive, if slightly disjointed. In order, the chapters cover the following: Essbase infrastructure, tackling bad data, Essbase Studio, BSO, converting BSO to ASO, MDX, ASO and performance, Essbase Java API, system automation with Groovy, Advanced Smart View, and how to successfully manage an Essbase system.

Some chapters are more useful than others, and some chapters are definitely “stronger” than others. I can easily see anyone reading the book and being asked to rank their chapters in order of preference or usefulness coming up with completely different rankings from anyone else.

The way in which chapters convey their information to the reader also varies significantly from chapter to chapter. There are various degrees of difficulty that the reader has to go through to extract the useful information from the chapters of the book. This thing is fill with gold and diamonds but its up to the reader, in no small part, to extract it, as it is rarely handed over on a silver platter.

For example, starting things off is John Booth’s chapter about Essbase infrastructure. Clearly, John is a very smart guy witha lot of experience, a recent Oracle ACE who graciously donates his time in the form of numerous forum posts and even creating Amazon EC2 images of Essbase servers. However, his chapter , but I found that his chapter  reads more like an animated brain dump of Essbase infrastructure. I imagine having a beer with John and saying “Tell me everything you know about infrastructure” and Chapter 1 is the result of that conversation. At the end of the day, I learned a lot, but as far as book form is concerned, would have appreciated more structure and progression to the manner in which the information was doled out.

Cameron picks things up in Chapter 2 which is ostensibly about slaying bad data in Essbase, a ubiquitous problem for any database administrator. While this is ostensibly the focus of the chapter, after a few pages, Cameron takes an immediate left turn into all things ODI and covers that. While also useful and insightful, as pedantic as it sounds, I found myself wishing the chapter had just been made to focus on ODI itself.

The “voice” in each chapter also varies, and in many instances while a chapter ostensibly wants to explain some concept to you, seems to talk as if you already know the concept. Going along again with my “you have to mine the information” analogy, I would say it’s also the responsibility of the reader to have been a somewhat astute observer of various acronyms and terms in the Essbase ecosystem even if they aren’t familiar with them beyond being able to simply unwind the acronym.

The book covers the Essbase Java API and brings in Mr. Essbase Java API himself, Tim Tow to help out. As an also experienced Essbase Java API guy myself, I certainly read this chapter very enthusiastically, and Tim does not disappoint. That being said, however, in my experience I haven’t found a lot of overlap between the Essbase administrator and Java programmers (much to my chagrin). So while I welcome the content, in some ways I question its usefulness to the average Essbase administrator. Of course, I could be mistaken and for the least part, exposing more people to the Java API in the hopes that they will pick it up and do something cool with it (as I have tried to do myself with various Essbase-related open source projects) is nevertheless extremely welcome.

The rest of the chapters round out topics from MDX to Groovy. Gary Crisci is the quintessential MDX practitioner, having presented on it and posted many an Essbase forum comment on it and does a nice job, as always, of conveying information on a topic that can be challenging. I have looked at Groovy in the past but haven’t made the jump just yet (too busy with Clojure and Go, I guess) but was found it to be an enjoyable read for what surely must be an uncommon, if esoteric, way to automate Essbase.

Lastly, on a purely aesthetic and non-functional note, I would like to say that I would have enjoyed more and better visuals in the book, in fact, the formatting of the book leaves something to be desired. For a tech book I would have loved to see something more modern than Times New Roman (for example, the formatting in O’Reilly and Apress books is second to none). Many of the tables and figures are laid out so as to be unhelpful and awkwardly cram information into a weird presentation, and a perennial pet peeve of mine is inconsistent framing, cropping, and styling of screenshots. Of course, this is all pretty predantic stuff and has more to do with my almost closet obsession with fonts and data visualization.

All that being said, all of the little things that I didn’t like about the book or thought could be better are  easily and handily overshadowed by all of the things that I did like about the book. This thing is jam-packed full of content and, perhaps more importantly, content that is based on the vast experience of some of the top Essbase people in the industry. These aren’t people pretending to know their stuff or trying to fool you, this is people at the top of their game in the top of their field that are sharing some top shelf knowledge with the rest of us. You just have to dig a little to get at it. At $69.99 list price on Amazon, this book is a little bit pricey for the average tech book. But you know what? Even if you only got one or two useful things out of it, it would absolutely be worth the price. This is an absolutely essential book for anyone that is serious about using Essbase and being a successful administrator or consultant. On a classic scale from one to five stars, I easily give this four stars.

Book review: Oracle Essbase 11 Development Cookbook

I was recently given the opportunity to review another Essbase book from Packt: Oracle Essbase 11 Development Cookbook by Jose Ruiz. Overall I would say I am pleased with the book. It covers a lot of ground and a lot of disparate tools, many of which are scantily documented elsewhere.

Before I really get into the review, I must say that I have never been a big fan of the approach that technology cookbooks take. I’m also not a huge fan of having a book for a specific version of software. Of course, in order for the cookbook approach to work you don’t have a choice but to tie to a version of software. This is because the recipes are sequential and very explicit — as with cooking  a recipe in real life — and rely on the exact version of the software in order for the detailed steps of the recipe to work. I’ve grown up with software, and am a cross between a visual and a kinesthetic learner, so my preference is to have concepts and goals explained to me, then to go exploring on my own. To this end, I find technology/recipe books to be tedious as they laboriously lay out the steps: click this, then click that, enter this text in, and 15 steps later you have a result.

So, my personal preference for book styles aside, this book largely succeeds for what it is: specific, methodical ways to perform a certain task. You won’t get a lot of explanation on why you might do something a certain way. In this regard, the book is useful as a complement to your Essbase literature rather that the place you would go to understand why you might want to accomplish some task.

Okay, now that I have beat up on that horse enough.

As I said, I enjoyed the breadth of content in the book. There are detailed recipes for setting up your relational data store to load a cube with EIS and Essbase Studio, building load rules and loading data to BSO/ASO cubes, writing calc scripts, working with Star Analytics, using EAS, HFR, writing MaxL scripts, and provisioning security. It even covers working with the revered Outline Extractor tool.

All of this content was really nice to see in book form. One of the upsides to the recipe format book is that it won’t spend a lot of time laboring over what a cube is and your first steps retrieving data with Excel. In fact, the book even says it’s not for beginners. It just jumps right in. I think this book can be a very handy reference for someone that needs something a little more guided than the technical reference (and less heavy).

On my arbitrary rating system, I would give this book a four out of five star rating. And again, that’s me trying to be fair to the book even though I’m not in love with this format, but it largely accomplishes what it sets out to do. I’d say it’s a great addition to the pragmatic Essbase developer’s library, but certainly not the only book in it.