Data warehouse appliances
Analysis of data warehouse appliances – i.e., of hardware/software bundles optimized for fast query and analysis of large volumes of (usually) relational data. Related subjects include:
- Data warehousing
- Parallelization
- Netezza
- DATAllegro
- Teradata
- Kickfire
- (in The Monash Report) Computing appliances in multiple domains
The four horsemen of data warehousing
I’ve been talking a lot to text mining vendors this week, as per a series of posts over on Text Technologies. Specifically, I’ve focused on the two with exhaustive extraction strategies, namely Attensity and Clarabridge. (Exhaustive extraction is Attensity’s term for separating the linguistic-analysis part of text mining from the DBMS-based BI/analytics part.)
So I asked each of Attensity and Clarabridge the side question as to which data warehouse software or appliances they were seeing. The answers were almost identical — Oracle, Microsoft SQL*Server, Teradata, and Netezza. One also mentioned MySQL and 2 HP prospects — but the HP sites were running NonStop SQL, not NeoView. Amazingly, there were no mentions of DB2. There also weren’t any mentions of the smaller specialist startups, such as DATAllegro, Greenplum, or Vertica.
Calpont finally has a multipage website
Calpont’s website is finally more or less real. It still doesn’t say much except that the company is in alpha test with a Type II appliance, and that the product has a columnar DBMS architecture and Oracle transparency (with DB2) promised. Oh yes; it has 32 employees. The “Customer” tab doesn’t list any customers, but I guess they saved site design money by having it all ready to go when that situation changes.
Philip Howard’s recent article has a lot more meat than that, including the perplexing bit of info that Calpont is starting out with a shared-everything architecture. Based on that, as well as the company’s prior technical efforts, we can probably conclude they’re focused on rather small warehouses.
| Categories: Analytic technologies, Calpont, Data warehouse appliances, Data warehousing, Emulation, transparency, portability | Leave a Comment |
Oracle sincerely flatters DATAllegro
Actually, I’m kidding with the post title; I doubt that Oracle’s new deal with DATAllegro partners Dell and EMC has much to do with DATAllegro at all. Rather, I think it’s an example of a trend I’m also sensing* from other major hardware vendors — doing deals with multiple data warehouse software suppliers to cover different hardware size ranges. This just happens to be the first one to be announced.
*How’s that for a nice, vague euphemism?
DATAllegro is targeted at warehouses sized, at a minimum, in the tens of terabytes of user data. Oracle’s technology works well enough up into at least the multi-terabyte range — unless you’re looking to get the best possible price and/or performance on your system — but then things start getting dicey. So there isn’t a lot of overlap between the two Dell/EMC offerings. Read more
| Categories: Analytic technologies, Data warehouse appliances, Data warehousing, DATAllegro, EMC, Oracle | 1 Comment |
Database management system architecture implications of an eventual move to solid-state memory
I’ve pointed out in the past that solid-state/Flash memory could be a good alternative to hard disks in PCs and enterprise systems alike. Well, when that happy day arrives, what will be some of the implications for database management software architecture?
- Compression will be even more important. Cost per terabyte of storage will spike up for that storage that is moved from disk to solid-state.
- The sequential-rather-than-random reading strategy of data warehouse appliance makers may become less relevant. The one way to get rid of the disk-speed bottleneck is to get rid of disks.
- DBMS will need to write data as rarely as possible. Solid-state memory tends to wear out if you keep writing over it. Assuming this problem gets better over time (if it doesn’t, this whole discussion is moot) but isn’t totally solved, architectures which have fewer writes are on the whole better.
| Categories: Data warehouse appliances, Data warehousing, Database compression, Netezza, Solid-state memory, Theory and architecture | Leave a Comment |
Four anonymous Netezza fans
I just found a blog post asking about Netezza that elicited quite a few responses, including at least four that purported to be from people whose companies had selected Netezza in a POC (Proof Of Concept) bake-off. One says Netezza was super-fast, even over DATAllegro, and DATAllegro’s professional services were lacking. One says Netezza is 50X faster than traditional alternatives on some queries, but up to 2X slower on some others. Two others just expressed love (or at least commitment) without giving details.
I haven’t yet looked through the rest of the responses in the thread.
| Categories: Analytic technologies, Benchmarks and POCs, Data warehouse appliances, Data warehousing, DATAllegro, Netezza | 3 Comments |
The Netezza Developer Network
Netezza has officially announced the Netezza Developer Network. Associated with that is a set of technical capabilities, which basically boil down to programming user-defined functions or other capabilities straight onto the Netezza nodes (aka SPUs). And this is specifically onto the FPGAs, not the PowerPC processors. In C. Technically, I think what this boils down to is: Read more
Notes from the Netezza user conference
EDIT: Big whoops, and apologies to Philip. I didn’t check the date, and what I linked to was last year’s article. That said, it read as if it could have been this year’s, which tells us something about the pace of Netezza’s information disclosure. Resulting errors of mine are left in place.
Netezza perennially annoys me by the secrecy with which it surrounds its information disclosure, especially at the annual user conference (just concluded). Essentially, except for what has also been separately disclosed, the whole thing is under NDA beyond the generality “We told you that we intend to improve our product by making more use of the FPGA.” Blech. That said, Philip Howard* has a long and — no surprise there! — upbeat article. So I’ll link to that, saving me some worries about what I myself am or am not allowed to say. E.g., I wouldn’t dare suggest — as Philip does — that Netezza’s zone maps (essentially, one-dimensional partitioning) could be enhanced going forward. And while I think Netezza has made strong efforts to tell the marketing stories Philip describes as being “hidden under a bushel,” I agree that — largely because of its self-defeating mania for secrecy — Netezza hasn’t done nearly as good a job of getting those messages accepted as it could have.
*Just to be clear — notwithstanding how much I tweak him for his exuberance, Philip seems to be a great guy, both in his publications and in person.
In general, much of what Philip wrote I would agree with. That said, let me hasten to point out some exceptions, including: Read more
| Categories: Data warehouse appliances, Data warehousing, DATAllegro, Netezza | 2 Comments |
Oracle promises to respond to the data warehouse appliance makers
On Oracle’s quarterly conference call September 20, Larry Ellison said:
There are some interesting niche players. Sybase gets smaller every year. Teradata, a database machine and now there’s some new database machine players, Neteeza, and let me say that Oracle is a very innovative company and I think you’ll see us with a response to some of these niche players some time at the end of this year or next year.
How important this is depends hugely, of course, on just what form Oracle’s response takes.
Oracle already does a great job of accelerating complex queries within the severe limitations of its SMP/shared-everything architecture. If it just does more of the same, perhaps adding in some hardware optimizations and vendor relationships, it will be a big ho-hum. At best, such moves will improve Oracle’s price/performance somewhat and garner some favorable publicity, and postpone the serious bleeding for a while as Oracle tries to find a better way of dealing with the specialist threat.
Much more significant would be a new engine, whether developed inhouse or acquired. Read more
| Categories: Data warehouse appliances, Data warehousing, Oracle | Leave a Comment |
Market reach tidbits from Netezza’s conference call
I’ve been slow to notice a very useful service being provided by Seeking Alpha, namely transcripts of quarterly earnings conference calls. For example, the Netezza call on August 23 revealed that Netezza sells approximately as many systems per year as it has quota-carrying sales teams. Or maybe it’s closer to 2 sales per team, especially for the more experienced ones. More precisely, the numbers discussed were 6-15 sales per quarter, and 35 sales teams. Average deal size was $2.3 million; based on the earnings press release, that suggests 10-11 deals depending on how much service revenue (if any) was included.
And by the way, if Netezza does 6-15 sales per quarter, and has a much smaller average sale than DATAllegro, and has much more revenue than DATAllegro — well, it’s easy to understand why DATAllegro isn’t exhibiting a very long list of customers.
Keep getting great research about data warehouse appliances and related technologies. Get a FREE subscription by RSS/Atom or e-mail!
| Categories: Data warehouse appliances, Data warehousing, DATAllegro, Netezza | 1 Comment |
Some pushback from DATAllegro against the columnar argument
I was chatting with Stuart Frost this evening (DATAllegro’s CEO). As usual, I grilled him about customer counts; as usual, he was evasive, but expressed general ebullience about the pace of business; also as usual, he was charming and helpful on other subjects.
In particular, we talked about the Vertica story, and he offered some interesting pushback. Part was blindingly obvious — Vertica’s not in the marketplace yet, when they are the product won’t be mature, and so on. Part was the also obvious “we can do most of that ourselves” line of argument, some of which I’ve summarized in a comment here. But he made two other interesting points as well. Read more