Is NoSQL So New?

March 7, 2010

A comment on the idea that the new “NoSQL” databaes may mean the era of the relational database may be over, at least in high volume OLTP. Despite what RDBMS verndors like Oracle may want you to believe, that era of the relational database never really began. To this day the world’s largest and busiest OLTP databases still run on legacy products like IBM IMS, CA-IDMS, and other CODASYL style databases. Data is hard to find but it is possible these systems run more transactions per day today than all relational products together.

IMS and IDMS first went live in late 1960’s and predate the existence of SQL, they are certainly “NoSQL”. The IBM data access method DL/I (or DL/1) for IMS is a put/get type architecture using keyed data segments, much like the NoSQL key/value pairs with structured values. There are differences, for example DL/I can store child keys in the data segments, which is what makes IMS a “heirarchial” database. I haven’t seen that in the new “NoSQL” databases but it could be done.

In a way it is shame to see all this effort to essentially re-invent something IBM, Cullinane and others built 40 years ago but then not everyone wants to buy an IBM 390 mainframe to run it. So we have NoSQL on more modest hardware and with more modern programming methods. But NoSQL is not so new but more another flavour of the way things have always been done in high volume OLTP.


Oracle Exadata 2

September 22, 2009

Yikes it been a long time since my last post. Let’s get started catching up

Just watched Larry Himself unveil the new Exadata 2 database machine.

Clearly this will put an end to the idea that Oracle wants to sell off Sun’s hardware business. Creating a fully integrated stack from hardware to OS to database to middle tier to applciation is the Oracle mission now and IBM is the target. Now we see why IBM made that desperate and hopeless bid for Sun back in the winter.

Oracle stacks up pretty well against IBM all through the stack and owns the Java technology they both rely on, so IBM has some need to be worried.

And the Exadata 2 is definitely a beast, big enough that Larry can at least claim it as the world’s fastest database machine. Now has Larry ever exaggerated before?

Sun Sparc/Solaris customers probably aren’t feeling so rosy though as the giant database machine runs Intel/Linux. Hopefully Oracle will quickly offer them a nice upgrade path from Sparc/Solaris before they all bail out onto IBM and HP.

I don’t see where MySQL fits in all this. Already the MySQL community is splintering and reforming around some of the old core, I don’t see that Oracle is going to have any claim to having more than a leftover piece of the “real” MySQL. Time to score some open source points and donate what’s left to some foundation. Larry though is having none of that so far reportedly.


Oracle & MySQL

May 7, 2009

Yes it remains to be seen what Oracle will do with MySQL. Will Oracle see it as a threat to Oracle Database and try to smother it or will Oracle see MySQL as more of a threat to things like SQLServer and use MySQL as a way to keep Oracle shops on the Oracle brand.

While MySQL may cost Oracle some revenue on the low end of Oracle Database, I’ve seen many Oracle shops use SQLserver for smaller , so-called departmental loads to save on licenses and admin costs. (Or at least a perceived saving on costs, that a whole other discussion.) Using MySQL would “obviously” save even more and cut MS out of the picture. There is a lot of value in keeping the competition’s sales person out of the building.


Kickfire

September 12, 2008

Kickfire is impressive, I like it. They’re setting both price/performance and pure preformance TPC-H records with it. Ever since I saw the open, pluggable architecture of MySQL I thought there could be a lot of innovations with it. And here is one example
This kind of stands my database world on its head though. I always saw MySQL as a little database for small web systems, not a real serious database like Oracle. . Being able to run big performance on big data warehouses with MySQL takes some getting used to. Certainly takes away the argument that only things like Oracle can scale up.

I’ll watch for Kickfire to be bought out by Sun or make a licensing deal with Sun. For Sun, being able to deliver low cost high performance database platforrms with their own database and own hardware puts Sun in a very good place. The MySQL purchase starts to make sense now.


New Directions

September 12, 2008

I’ve been looking at several new things lately.

My SQL
I remember when MySQL first hit the scene, all of us who used “real” databases like Oracle sneered at it. It didn’t have tranasaction control, referential integrity etc, how it coold it even be called a database we thought. Well, it has come a long way. I especailly like the pluggable transaction engine architecture which allows me to plugin the complexity I need only as I need it. This is a cleaner solution than Oracle’s policy of offerig several incompatible products or trying to pretend that the 11g monster is a one size fits all solution.

I’d really like to see a column-store plugin. It would be an easy way try out column stores.

Ruby on Rails
This is a framework I can work with. I’ve seen too many frameworks that just bury me in syntax. The proverbial 400 line Java “Hello World” program problem. Rails certianly enforces a set of restrictions but the benefit is way less code. The remaining code is much more focused on implementing the business rules rather than the syntax to drive the framwork.
Clearly there are limitations especialy noting that it really works best with new applications and doesn’t work nearly so well with legacy data models. You do have to reconsider the conversion costs, if you are making substantial changes to a cumbersome legacy system anyway, a complete rewrite in Rails could be cheaper in the long run.

Why
Also a special mention to WhytheLuckyStiff and his Camping and Shoes projects. For those times when even Rails seems too heavy, something even lighter. Brilliant little frameworks, almost pieces of art in their brevity and simplicity..


Data Warehouses

March 27, 2008

I never really liked those star and snowflake schemas, they seemed to be building too much of the supposed application into the data model. To me, a data warehouse should be as neutral as possible to support future possible applications. I’ve always preferred that the data warehouse be close to the OLTP source and maybe even more normalized to remove the OTLP performance hacks. We understand the OLTP data model because it supports the real business, if the data warehouse is more or less the same, it is also easy to understand. Those star schemas just confuse me, I can’t tell if the data is in any way correct. If you need a star schema as a hack for a particular problem, it is fine to roll one up as a copy (or data mart) but don’t build the warehouse on that sort of thing.

Column oriented databases  are interesting. See <http://www.databasecolumn.com/&gt;. The fact that the legendary Michael Stonebraker is involved in this alone makes it interesting. I think they’re onto something and it seems to be a way to get around some of the performance issues without the complexity of star schemas.


My Sun SQL

January 19, 2008

The bottom end of the database market may get a shakeup now that Sun has bought MySQL. We’ll have to see what sort of support models Sun comes up with but this should remove enterprise class support as a concern, especially for customers already on Sun systems.

For price or license sensitive customers,  paying no license at all is obviously appealing. Just removing the hassle of tracking licenses is an incentive itself. Will the bottom now drop out of the bottom of the market? Of the big three, I would suspect SQLserver is most vulnerable since it is sold on price more than Oracle or DB2. Anyone buying Oracle Enterprise licenses can’t very price sensitive, I don’t see much impact there and Oracle already gives away low end versions.

If Sun has success with MySQL, will it drag PostgreSQL up along with it? PostgreSQL is already “commercialized” as EnterpriseDB and RedHatDB but I haven’t seen much traction in the market so far.

I can’t help thinking this is a bit of a shot back at Oracle. Oracle now supports Linux, undercutting Sun’s preeminence as the Oracle platform of choice. So Sun shoots back by supporting MySQL.

I never understood why a software company like Oracle would encourage open source software. For Sun (or IBM or HP), using open source software to possibly drive hardware sales makes some sense but for Oracle? I don’t get it.