re: city dwellers
There will be sharding in Oracle Database 12c Release 2 but it remains to be
seen how it compares to Oracle NoSQL Database in terms of functionality, ease
of use, and price. My bets are on Oracle NoSQL Database.
re: Oracle is not just a relational database
Oracle sells two relational databases (Oracle and Rdb), one pre-relational
network database (CODASYL DBMS), two NoSQL databases (Berkeley DB and Oracle
NoSQL Database) and one in-memory database (TimesTen).
“Oracle Rdb is a full-featured, relational database
management system for mission-critical applications on OpenVMS platforms.”
(http://www.oracle.com/technetwork/database/database-technologies/rdb/overview/index.html)
“Oracle TimesTen In-Memory Database is a full-featured,
memory-optimized, relational database with persistence and recoverability. It
provides applications with the instant responsiveness and very high throughput
required by database-intensive applications. Deployed in the application tier,
TimesTen operates on databases that fit entirely in physical memory.”
(http://www.oracle.com/technetwork/database/database-technologies/timesten/overview/index.html)
“Berkeley DB is a family of embedded key-value database
libraries providing scalable high-performance data management services to
applications. The Berkeley DB products use simple function-call APIs for data
access and management.”
(http://www.oracle.com/technetwork/database/database-technologies/berkeleydb/overview/index.html)
“Oracle NoSQL Database provides a powerful and flexible
transaction model that greatly simplifies the process of developing a
NoSQL-based application. It scales horizontally with high availability and
transparent load balancing even when dynamically adding new capacity.”
(http://www.oracle.com/us/products/database/nosql/overview/index.html).
“CODASYL DBMS is
a multiuser, CODASYL-compliant database management system for OpenVMS operating
systems. CODASYL DBMS is designed for databases of all levels of complexity,
ranging from simple hierarchies to sophisticated networks with multilevel
relationships. CODASYL DBMS provides a reliable operating platform for
application environments where stability, high availability, and throughput are
essential.” (http://www.oracle.com/technetwork/products/rdb/index-086844.html)
From: rfreeman@xxxxxxxxxxxxxxxxx
To: tim@xxxxxxxxx; iggy_fernandez@xxxxxxxxxxx
CC: oracle-l@xxxxxxxxxxxxx
Subject: RE: Community Announcement: NoCOUG 2016 Spring Conference: Where SQL
and NoSQL come together (with hands-on labs and cherries on top)
Date: Mon, 2 May 2016 16:01:17 +0000
I was just thinking about the database landscape (ie: relational, NoSQL, etc)
earlier today in a different context.
Database ideas (I’m avoiding the word fad) come and go… look at the whole
object oriented database fad that I remember from the 90’s or so… Where did
that go?
Then there were the XML, etc databases, Hadoop, etc..etc… There are 1000’s of
potential one-off types of database applications that meet a very narrow set of
needs. It seems to me that many of these niche databases take off because they
do meet a very specific
set of needs that are identified. It might also be that specific sets of
technology become available to allow for these niche databases to be developed.
There is always a huge pack of these. I love the
Data Platforms Map from 451 research - it nicely makes my point for me.
Database here, database there, here database, there database, everywhere
database database.
I think there is a database product feature lifecycle… And I think Oracle is a
part of that lifecycle. In fact, I think it excels at the part it plays in the
overall product life cycle.
One of my mentors long ago used to liken this cycle to the following – I see
Oracle as among the City Dwellers…:
1.
Explorers
2.
Pioneers
3.
Settlers
4.
City Dwellers
Explorers are the true innovators – the Googles, the CERNS’s, the
military or universities, etc… with cutting edge data needs that need to be
addressed for which there is no current solution that is viable. This stuff is
bleeding edge. It’s not close to ready for commercial or public consumption but
it offers great ideas
for the pioneers to take advantage of.
The pioneers are those who follow up afterwards, taking advantage
of what was learned by the explorers and developing it into something more
useful. This could be the Googles, following up on their own products. Or, it’s
someone who’s just taking advantage of the experience of the explorers and they
start to produce something
based on that research or some theory. In some respects I think Oracle’s first
databases (pre-commercial and early commercial) were pioneer databases – where
Codd and his contemporaries were the explorers. Often Pioneers tend to produce
things that are very
niche.
The Settlers take the work of pioneers and produce nice commercial
versions of their work. Sometimes they are the pioneers just building on their
work, sometimes they are licensing the work of the pioneer or perhaps it’s just
someone who’s building a better mousetrap based on the work of the explorers
and pioneers. Sometimes,
the city dwellers have noticed what the pioneers are doing and attempt to
integrate their work into an existing product (ie: Oracle). These are often
still very much niche databases and nice functionality, but since they are
attempting to be a serious commercial
product, they begin to mature and add more enterprise features. In my
experience, I’ve seen the settlers often steal this niche business away from
the city dwellers at first, because they perform their niche work very well.
Sometimes the settlers become city
dwellers – more often than not, the settlers either sell to the city dwellers
or their product dies as the city dwellers move in.
The city dwellers eventually take the product that the settlers have
produced and integrate it into a more mainstream product. Usually these
products are provide a great breadth of services, and the new niche product
that has been integrated in from the settler stage is just part of a much
larger suite. This is where I see
Oracle excelling at. Often, in the short term, the functionality it integrates
in is not as good as that provided by the settlers. But, if that product has a
market, it will mature in the Oracle Database and it will overtake most, if not
all, of the niche
products.
When you think about it, this is why Oracle does not panic when something new
does not work perfectly. It know the lifecycle and it knows that it’s
implementation
of that product, if there is a demand for it, will improve, grow and at some
point it will tend to dominate that part of the market.
This is why, unless Oracle changes its vision, it’s Database will always
command a significant part of market share, in my opinion. Oracle is not just a
relational
database – but it’s many kinds of databases and it tends to perform each of
those different kinds of jobs well. Oracle knows that many customers will
eventually prefer to consolidate infrastructure and reduce complexity – and
when they can offer a product
that does what you want it to do, and a fairly fast speed and a cost that is
competitive compared to the cost of sprawling infrastructure – that is a
compelling argument.
Cheers..
RF
From: Tim Gorman [mailto:tim@xxxxxxxxx]
Sent: Monday, May 02, 2016 10:16 AM
To: Robert Freeman <rfreeman@xxxxxxxxxxxxxxxxx>; iggy_fernandez@xxxxxxxxxxx
Cc: oracle-l@xxxxxxxxxxxxx
Subject: Re: Community Announcement: NoCOUG 2016 Spring Conference: Where SQL
and NoSQL come together (with hands-on labs and cherries on top)
I'd guess that it is not a quote, but an interpretation of actions by Oracle.
Why else would they offer a Big Data Appliance based on Hadoop, and support Big
Data connectors, etc?
It is one thing to pivot the entire company in a single direction. Oracle
Cloud is not "some catching up", but direction for the entire corporate.
Oracle has made no pretense to make a "Big Data" their focus.
It is another thing to offer a capability to augment existing capabilities
(i.e. Big Data) - this is not a refutation of either relational technology nor
an embrace of Big Data and NoSQL, but rather an acknowledgement that the need
exists and the capability
should be supported.
So interpreting these actions as admission that "relational database management
systems have some catching up to do in certain specialized use-cases such as
event processing" is reasonable, in my opinion.
Just my own US$0.02... maybe there really is an underlying quote in context...
:)
On 5/2/16 08:54, Robert Freeman wrote:
Can you source this quote from Oracle? I’d be very interested in reading it in
context….
Robert
From:
oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx]
On Behalf Of Iggy Fernandez
Sent: Monday, May 02, 2016 2:43 AM
To: oracle-l@xxxxxxxxxxxxx
Subject: OT: Community Announcement: NoCOUG 2016 Spring Conference: Where SQL
and NoSQL come together (with hands-on labs and cherries on top)
The inventor of relational theory, Dr. Edgar Codd, had the last word on NoSQL
more than thirty years ago when he said “Only if the performance requirements
are extremely severe should buyers rule out present relational DBMS products.”
The bottom line is that
Oracle professionals need to learn about NoSQL since, as admitted by Oracle
Corporation, relational database management systems have some catching up to do
in certain specialized use-cases such as event processing. Our conference
director has therefore created
a fabulous
agenda combining the best of SQL and NoSQL (with hands-on labs and cherries on
top).
The conference is free for members and their guests, first-time NoCOUG
conference attendees, PayPal employees and students. Register at
http://nocoug.org/rsvp.html.
