John Myers: April 2010 Archives

The team from Calpont came to the BBBT on Friday to discuss their DBMS product – InfiniDB.  The group included:

• Jeff Vogel – President & CEO
• Robin Schumacher – VP, Products
• John Weber – VP, Marketing & Business Operations
• Jim Tommaney – Chief Product Architect

Product Fits and Starts

Calpont has had several different company starts since its beginning in 2000.  It was interesting to see the evolution from proprietary hardware to Open Source software only solution.

That being said, the product release plans for 2010 seem to be well organized and following the stated goals of their business and marketing plan.

All Things to All People?

As I said as part of the BBBT Twitter Stream, I liked several parts of the Calpont positioning messaging relating the differentiation between row-based and columnar-based DBMS platforms for both functionality and market positioning.  However, the technology messaging seemed to stray from this positioning to an “all things to all people” message.  It seemed that InfiniDB was being positioned as something that could handle all challenges regardless of situation.

This could have been the result of the early stages of the InfiniDB messaging or the nature of the BBBT forum.  In any event, I feel that this needs to be clarified in the future for InfiniDB to find its footing in the market.

Telecom Applications

As with most of the columnar DBMS providers, I find great use for columnar DBMS platforms in the telecommunications space.  The nature and amount of data in telecommunications is particularly well suited to the columnar DBMS technology.  For Calpont’s InfiniDB, I see the same promise.

I see challenges in maturity and product development before InfiniDB can be accepted into the official telecom data centers where other DBMS products generally hold sway and political clout. 

However, in Calpont’s stated efforts to begin at grass roots level(s) with their InfiniDB Community edition, I see openings and potential placement within the various business organizations of telecoms.  The “shadow IT” departments may come to the rescue via unofficial ( read “rogue”…  ) and official proof of concept initiatives.

In Conclusion

As many of my BBBT colleagues stated via Twitter and in person, Calpont has a promising offering with InfiniDB.  But there is work to be done in relation to refining the messaging for some of their stated target markets and linking their technology to that messaging.

Overall, I like columnar DBMS technologies for their promised and actual performance.  I look forward to hearing more in the future about how Calpont matches their intentions with performance to differentiation their solution from others in the marketplace.

++++++++++

Update April 26 – Also check out Jos van Dongen’s comments on Calpont and InfinDB.  Good analysis and insight.

Following up on the “group” BBBT posting from the recent Microsoft SQL Server 2008 R2 release, I decided to write a “concurring opinion” on the SQLServer 2008 R2 announcement(s).

I liked what I saw from the web materials and “press releases” from Microsoft on SQLServer 2008 R2.  However the briefing(s) and related materials didn’t bring those points to light.

Possunt Quia Posse Videntur

There are plenty of positive technical aspects to like ( … or hopefully like… ) about SQLServer 2008 R2.  I like the Parallel Data Warehouse functionality for large data sets via the DataAllegro acquisition / Project Madison. Telecoms will embrace the ability to have a SQLServer “certified” appliance in the data center.

I also believe that the telecom industry will like, and embrace, the CEP aspects of SQLServer 2008 R2.  As more and more of the Telecom industry moves toward automated processing for:

• Provisioning
• Activation
• Maintenance

Understanding how these processes and associated events relate to the bottom line will be key for both the business side and the IT side of an organization.

Errare Humanum Est

However, the apparent lack of “solutions” ( … or point-to-point integration mentioned in the ‘majority opinion’ … ) for SQLServer 2008 R2 show some of the errors of the Microsoft approach with this product.

Telecom IT departments are looking for ways to eliminate costs and streamline operations.  I fear the “flexibility” in the lack of point-to-point integration solutions will ‘scare off some implementation teams in favor of more mature solutions.  While these solutions may come from the Microsoft partner ecosystem, SQLServer 2008 R2 leaves some of those to the imagination…

And at the moment, telecom IT departments are looking for imaginative solutions and not blank slate technologies.

Verbum Sat Sapienti

I agree with Merv Adrain’s individual blog posting on SQLServer 2008 R2 when he said:

“There is a great deal going one in this release, far more than was discussed at the briefing. And Microsoft’s communications efforts will no doubt ramp up in the weeks ahead.”

For now, I will take Microsoft at their word on, but I would like to see more information in the coming weeks before the official launch.

“Returning Champion” Lyzasoft made another visit to the BBBT.  Unfortunately, due travel conflicts, I was unable to participate in person with the Lyzasoft team of:

• Scott Davis, CEO
• Brian Krasovec, CTO
• David Pinto, VP Engineering

Yet, I was fortunate enough to follow the #BBBT twitter feed remotely ( … as anyone can…  ) and I had a recent briefing from Scott Davis.

Social Concepts

Lyzasoft brings to the table an interesting perspective.  Not that BI needs to have social/collaborative elements “bolted” to it or that social/collaborative engines need to have BI “bolted to it…. But rather there needs to be a blend of these concepts.

In particular Lyzasoft brought the concept of the 5Cs of user roles:

This idea brings about a particularly valuable concept in that “maturity” ( …probably the wrong word, but good enough… ) of data/analytics creation-consumption roles within an organization.

Much like previous models associated with BI roles, this 5Cs of user roles takes into account the differences between consumers and creators of information.  The key aspect that changes with this 5Cs model is that the collaboration is defined in “21st century” or “twitter-sphere” concepts rather than the cubie bullpen or meeting room table.

More Things Change More They Stay The Same 

With this move toward a more collaborative philosophy; Lyzasoft, and their offerings Lyza and Lyza Commons, bring some of the tried and true aspects of business intelligence.

Of course you need to acquire and analyze the data… The key new addition is the “from the ground up” integration of social/collaboration aspects and having those aspects cascade back throughout the metadata ( … again probably not the ‘right’ word, but good for this conversation… ). 

Linking the data lineage with the social aspects enables something that current social media and to a certain extent “old school” web applications have lacked.  For many years, people asked about the validity of “Wikipedia” entries due to the lack of an ‘evidence’ trail associated with the information.

Now with Lyzasoft’s offerings it is possible to not only link datasets together and provide context “socially” and distribute “collaboratively”… AND… detail the history or lineage of the data. 

Telecom Applications

Unfortunately, the culture of the average telecommunications organization will have many problems with the social/collaborative concepts behind Lyzasoft’s offerings.  The idea of releasing centralized control may be too much for most telecom IT departments to understand/comprehend.

However once embraced/understood, telecom data governance professionals ( data stewards, etc ) will find the ability to spread data creation and consumption duties in a crowdsourcing manner ( … with associated “unproven” wikipedia reference… ) to be an excellent way to extend their reach into the business areas of the organization while providing additional value to their roles.

Specifically, I see great value to the ares of revenue and expense management within telecoms for Lyzasoft.  These are two areas that often require a great deal of consensus and coordination between disparate groups.  Being able to utilize social/collaboration concepts into the BI analysis will allow for the expansion of those who can contribute to these activities from core groups to those interested within the organization.  Yet, there will be the controls and visibility to the data for IT stakeholders  that will make the efforts meet some of the core data governance requirements and concerns. 

In Conclusion

Lyzasoft is doing an excellent job of making both a conceptual and technical leap into social/collaborative BI.  Rather than appending one concept to another, Lyzasoft is ( …in my opinion… ) growing an entirely new kind of BI application.