Infobright’s Release 3.4
Infobright called a couple weeks ago to discuss, among other subjects, its subsequently-released Infobright Release 3.4. I made no effort to distinguish between community/open source and professional/chargeable editions, but leaving that aside, it seems fair to characterize Infobright 3.4 as having two overlapping primary themes:
- Performance and bottleneck cleanup.
- “Omigod, you mean you didn’t have that feature before?” cleanup.
That said, the traditional release for cleaning up the last huge gaps in an analytic DBMS product seems have become 4.0; recent examples include Aster Data, Vertica and Greenplum. Infobright seems on track to be another example of that rule.
Ack. Now that I’ve said that, other vendors are going to be tempted to accelerate their numbering so as to reach the 4.0 mark sooner …
A lot of Infobright performance enhancements are in the vein “We used to rely on generic MySQL for that, but now we do it ourselves, and it works a lot better.” Examples include:
- Infobright now does DELETEs all at once, vs. the previous row-by-row way. This makes DELETE performance similar to SELECT performance, when previously there was a big difference.
- Ditto, if I understood correctly, INSERTs and UPDATEs.
- Each release, Infobright covers more SQL functionality itself and passes less through to the generic MySQL engine.
- UTF-8 Unicode data can now be loaded via Infobright’s parallel loader. Previously, you had to use MySQL’s load.
Infobright has also added workload management in 3.4, and this is intertwined with multicore parallelization, apparently because the workload manager decides when a query should use multiple cores to execute. Infobright further says that multi-user INSERT performance has increased a lot more than single-user, but I have forgotten why that is.
Infobright now streams data back to the client faster. E.g., unless there’s some good reason not to, partial query results are pipelined back as they become available.* Finally, loading data no longer locks tables from being read (in my notes I wasn’t sure whether that was a current or future feature, but Infobright’s marketing seems to indicate it’s current). For some reason, Infobright is positioning this as a major, innovative feature.
*A good reason not to do this would be an aggregate that requires full materialization of the table before Infobright can carry it out.
Comments
6 Responses to “Infobright’s Release 3.4”
Leave a Reply
[…] Infobright (so far as I can tell, just because it was time for a product release, and also to get ahead of the summer doldrums) […]
Yes Query while Load (and Query while Insert/update/delete) is a current feature in Enterprise Edition 3.4 which will ship this week. We weren’t positioning it as remarkably innovative but very useful as are the significant performance enhancements for query, insert/update and delete.
As you know we also stress that users get great query performance without any tuning. Per your tweet saying that other vendors claim the same thing (which they do), by “no tuning” we mean more than just no need to create indexes – we also mean no need to create projections, partition data, and other work that many databases require for performance.
Ease of administration, low cost, much less hardware and fast query response are the reasons we continue to grow in our targeted markets. For companies looking for high end enterprise data warehouses, there are plenty of other vendors they can turn to.
Susan,
How long would you say it now takes to “warm up” an Infobright installation with queries before the performance gets great?
Curt,
One thing I will point out is that we are very customer driven for every release. During our internal processes, we prioritized a set of key defining features of the release based on all the great customer feedback we get. For V3.4 we heard our customers wanted even greater performance gains, deeper support for UTF-8, especially while loading or unloading UTF-8 data and everyones favorite, query while load, in no particular order. These were considered high priority by our customers. We tend to try to get these frequently requested features into our customers hands as quickly as possible. Of course, we do apply levels of innovation that are sometimes not quite as visible. Every release we want to ensure the architecture stays fresh and is stable and reliable and of course high quality. Wish we could add a lot more features, but we do our best every release cycle to help prioritize based on our available resources.
As far as your warm up question. We think that lots of diverse queries run really fast when the Knowledge Grid has context. We do everything possible to make sure the Knowledge Grid has the level of metadata necessary to execute queries and sometimes not even having to hit the storage. This is where things really fly. For example, summary or min/max, etc.
Bob,
That all makes sense.
But my point is that the desirability of warming up the Knowledge Grid illustrates that Infobright, just like every other DBMS in the history of software, calls for at least SOME effort to administer or tune. 😉
[…] to all remain as independent companies. (Edit: Infobright becomes a full member of that list if its Release 4 goes well.) CurtMonash: Some will buy each other. HP needs to buy somebody at some point. Dell and […]