Обсуждение: Metapa - PostgreSQL/Linux clustering for BI

Поиск
Список
Период
Сортировка

Metapa - PostgreSQL/Linux clustering for BI

От
"Ned Lilly"
Дата:
Saw that these guys had raised some more VC money, and made an acquisition.  Anyone from the company on the lists?
(Probablybest to reply only to -advocacy, but wanted to cast the net wide). 

From their website (www.metapa.com):

--

Metapa CDB is an enterprise Java software application running on Linux. Key aspects of the CDB architecture include:

* Built on open standards.
* Based on a shared-nothing architecture.
* Advanced SQL parallelization engine supports predicate-based partitioning and query acceleration.
* Based on a database-agnostic framework allowing clustering capability for different RDBMs, initially supporting
PostgreSQL.
* Standards based client interface (JDBC/ODBC/DBI).
* Custom low-latency, preemptable protocol on top of TCP.
* Purpose-built from the ground up to increase the performance of "star-schema" data models typical in today's
demandingdecision support systems. 

Re: Metapa - PostgreSQL/Linux clustering for BI

От
Dennis Gearon
Дата:
I'll take a look at it,
    but my experience in reviewing  many applications on JAVA,
BRING MORE HORSEPOWER (in your server), at least double what it would
normally take. Java on the server side IS a great development package,
lots of power, typechecking, and all sorts of goodies, but at the cost
of extra horserpower required.

Ned Lilly wrote:

>Saw that these guys had raised some more VC money, and made an acquisition.  Anyone from the company on the lists?
(Probablybest to reply only to -advocacy, but wanted to cast the net wide). 
>
From their website (www.metapa.com):
>
>--
>
>Metapa CDB is an enterprise Java software application running on Linux. Key aspects of the CDB architecture include:
>
>* Built on open standards.
>* Based on a shared-nothing architecture.
>* Advanced SQL parallelization engine supports predicate-based partitioning and query acceleration.
>* Based on a database-agnostic framework allowing clustering capability for different RDBMs, initially supporting
PostgreSQL.
>* Standards based client interface (JDBC/ODBC/DBI).
>* Custom low-latency, preemptable protocol on top of TCP.
>* Purpose-built from the ground up to increase the performance of "star-schema" data models typical in today's
demandingdecision support systems. 
>
>---------------------------(end of broadcast)---------------------------
>TIP 7: don't forget to increase your free space map settings
>
>
>