Rocky Castaneda wrote:

> Is this a combined recipe of MySQL and FirebirdSQL?
>
> I hope you'll setup many chain of stores for this recipe.
>
> On Sat, Jun 14, 2008 at 6:37 AM, Jim Starkey <jstarkey@...> wrote:
>
>
>> For those of you who didn't hear, today is my last day at MySQL, and
>> Monday will be my first day as Chief cook, CEO, and bottle washer of
>> NimbusDB, Inc., a software start up so young it doesn't exist yet.
>>
>> Nimbus, as the name hopefully suggests, is a relational / semantic
>> database system designed to run on clouds. It is a radical re-invention
>> of database technology, breaking with 30 years of disk-based
>> technology. Nimbus is designed to scale to almost arbitrary capacity by
>> simply plugging in more computers. It is also software and hardware
>> fault tolerant, including software upgrades. It's also an idea that's
>> been kicking around my brain for the last six months that needed to get
>> lose to the real world.
>>
>>
>
>
>
>

No, it isn't remotely similar to either. For example, nodes that do SQL
don't do disk I/O, nodes that do disk I/O don't do SQL, and all files
are written once and never updated.

A cloud is a new an different platform, though you wouldn't know it to
listen to pundits. In started programming in the era of batch with card
decks and I/O windows, and have observed the following:

* When timesharing was invented, the batch guys considered it a
special case of remote job entry
* When departmental computing (i.e. VAXes) came around, the
timesharing guys considered it just timesharing on a different scale
* When client/server came done the pick, the departmental computing
uys considered a PC just a special case of a smart terminal
* When the Internet showed up, the client/server guys considered the
Internet a special case of a LAN
* As clouds arrive, most every one considers them either as clusters
or warty SMP machines.

Nimbus is a total break from 30 years of continuous evolution of
disk-based database system. Pages: Just say no.

I'm also re-thinking other quaint holdovers from the past:

* Nimbus implements the semantic data model where on table can be
defined to "extend" another. A record inserted in extended table is
also visible in the parent table, albeit with fewer attributes
* Nimbus supports unbounded types. Fields can be declared as type
"string" or "number" rather than "varchar(37)" or "short int".
* Nimbus supports UTF8, period. Character set conversion is a client
issue. It does, of course, support collations.
* Nimbus will support an open ended set of aggregating interfaces,
so a single round trip to the server can return the results of many
queries