Getting to Know DATAllegro, Part I
To me, DATAllegro has always been the black sheep of the new database vendors. They've never done anything sexy (that I'm aware of), they're on the west coast (yes, that is a sin) and they don't really make any noise about performance. Actually, they don't seem to make any noise at all, save comments by CEO Stuart Frost on various blogs.So when the opportunity to learn more about DATAllegro popped up a few weeks ago I was both thrilled and apprehensive. On the one hand, I knew nothing about them, so I was glad to have the opportunity to fill in one of the (very few and small
) holes in my knowledge of the new database industry. On the other hand... I wasn't really expecting to hear anything all that interesting.It turns out that even though DATAllegro's not really making much noise, what they are doing is very, very interesting.
This post, as well as the one or two future posts, will cover what I learned about DATAllegro, why I think it's interesting and what I think it will mean long-term. As always, if you see anything amiss or a-missing, please let me know.
Continue reading "Getting to Know DATAllegro, Part I" »
Teradata Appliances: Me Too
Teradata's announcement of their new "appliances" a couple weeks ago was, to me... well, uninteresting, quite frankly. So I put off writing anything about it. And I'd pretty much forgotten about it altogether, to be honest. (Like I said, uninteresting.)This morning, however, Phillip Howard's comments on Teradata appliances reminded me of said announcement and my lackluster opinion of it. And though I do my best to avoid ever saying "me too", Phillip's comments express how I feel about the subject surprisingly precisely. So rather than rehashing things, I'm simply going to nod, smile, and say...
Yup, me too.
Really Cool Stuff: Teradata Express Edition
Ok, so, I'm not exactly a huge Teradata fan, but I have to admit that the idea behind Teradata Express is pretty cool. Nothing like being able to download and learn hands-on about serious database software.I've requested my copy... I'll post more when it arrives and I've had a chance to try it out.
Webinar Recap: Data Warehouse Appliances: An Update on the State of the Art
Quick notes, thoughts and comments from today's Data Warehouse Appliances: An Update on the State of the Art webinar:• According to Mr. Russom, the definition of "data warehouse appliance" is shifting from a full hardware/software stack to a packaged or bundled system using commodity hardware (my words, not his). I think it's hard to argue with that, despite my belief that for most systems "appliance" is a misnomer.
One thing that became especially clear was how much DATAllegro needs this shift in definition to take place. They are one of only two vendors I can think of, Kognitio being the other, who started with a proprietary system and moved to a bundled system based on commodity hardware. Appliances are sexy, so everybody wants to have one, but DATAllegro and Kognitio need the terms to change so they don't have to change their message.
• In one slide, it was said that Dataupia is "usually" added on as an accelerator. As far as I know that's the only way you can use it.
• One of the recommendations made was that data warehousing professionals should "stick with the data" and not spend their time dealing with hardware configuration. While this clearly favors the new breed of database vendors in general, and not least of all the talk's sponsor DATAllegro, I think this is sound advice. A bit utopian, but a worthwhile goal and a good yardstick by which to measure your solution.
• In a nice reality check, the point was made that hardware re-use is unusual in practice and that there isn't really much reason to get hung up on it when selecting a vendor. In my experience it is fairly common for old systems to get moved to test environments or simply disposed of, as was suggested in the talk, so I think this is a valid point.
• Maybe my favorite point of the entire talk was the assertion that data warehouse appliance vendors will continue to become more diverse. Amen to that.
• If you're interested, a recording of the webinar was promised, though I don't think it'll be available until tomorrow.
• Lastly, one nit that I need to mention just to vent: "precedence" is not an alternative spelling for "precedents". They mean different things. Grr.
All in all, a pleasant way to spend a lunch hour. I look forward to more.
UPDATE: Slides from the presentation are available, as well as an On-Demand Replay.
More on Dataupia Compatibility
Immediate mea culpa regarding my last post: Dataupia has already announced compatibility with "others", namely SQL Server and DB2. Well, maybe announced is the wrong word - I don't remember seeing an announcement, but they aren't hiding it either. From their Product Overview page:Omniversal Transparency™ — Supports applications running on Oracle, Microsoft SQL Server, and DB2 databases natively. Accepts data from a range of legacy systems (including mainframe), making previously isolated data available for reporting and mainstream analytics.
Now, as far as I'm aware, they're the first company to support all three major RDBMSs. That's huge.
But since I'm never satisfied with anything, here's a thought: it would be a real coup if Dataupia added MySQL support as well (via a MySQL storage engine plug-in, of course). Talk about taking data warehousing to the masses! And they may be uniquely positioned to do that, both as a matter of technology and cost.
Crazy, maybe, but possible. And fun to think about.
