►
Description
Lightning Talk: Shifting the Database Paradigm - Richard Douglas (Couchbase)
This OpenShift Commons Gathering was held on July 6th, 2022 live in London, England
https://commons.openshift.org
A
Foreign
Douglas
I'm
here
from
couchbase
it's
great
to
be
in
front
of
a
live
audience
instead
of
on
a
zoom
call.
Last
time,
I
was
at
a
conference,
was
for
Microsoft
SQL
server
in
Germany
two
weeks
before
the
pandemic
hit.
So
it's
been
been
some
time
so,
like
myself,
there
has
been
a
big
paradigm
shift.
I
mean
I've,
moved
from
a
relational
background
and
now
I'm
in
the
no
sequel
space.
Okay
and
that's
where
couchbase
really
excels.
A
So
you're,
probably
asking
yourself
what
is
couch
base
and
the
clicker
Works
Happy
Days.
So
couchpace
is
a
nosql
data
platform.
I've
only
got
10
minutes,
so
I
can't
really
go
in
depth.
We
do
have
a
exhibition
booth
just
outside,
so
please
feel
free
to
come
and
grab
one
of
us.
So
with
that
in
mind,
what
I
wanted
to
do
is
I
wanted
to
give
you
one
word
to
take
away
and
that
one
word
is
choice.
A
You
always
have
a
choice
with
whatever
it
is
that
you
want
to
deploy,
so
we're
going
to
be
talking
about
deployments
and
how
we
can
work
within
those
deployments.
So
the
first
thing
is
lock
in
now.
18
year
old
me
thinks:
lockings
are
great
in
fact,
16
year
old
me
thought
lock
into
a
great
as
well,
but
a
lock-in
isn't
necessarily
what
you
think
it
is.
A
Okay,
lock-ins
can
be
fun,
but
lock-ins
can
also
be
a
bit
scary
now,
I'm,
not
going
to
say
we're
going
to
send
you
to
jail
for
not
paying
support
and
stuff.
Like
that,
I'm
saying
that
you
want
to
be
able
to
transfer
from
platform
to
platform,
you
want
to
be
looking
at
something
that
is
platform
agnostic,
okay
and
we've
been
designed
from
the
ground
up
to
be
able
to
provide
that.
For
you,
another
choice
is
the
choice
of
location.
A
A
The
alternative
to
that
is
obviously
storing
things
in
the
cloud
or
some
co-located
data
center
So
within
this
I
think
this
demo
is
working.
Now
this
is
going
to
be
awesome.
You'd
be
pleased
to
know.
I
haven't
got
any
Demos
in
here.
So
on
premises,
you
do
have
the
ability
to
use
physical,
bare
metal.
If
anybody
remembers
that
you
know
plugging
in
machines,
beige
compact
boxes,
all
the
rest
of
it,
you
can
run
it
on
virtualized
platforms
or
some
of
you
guys
may
have
heard
of
containers.
A
No
anyone
yeah,
you
can
use
containers
as
well.
Okay,
and
we
were
the
first
nosql
platform
to
be
running
on
openshift,
okay,
we're
a
level
five
certified
operator,
so
you
can
be
sure
that
everything
will
work
as
you
would
expect
it
to
and
obviously,
within
the
cloud
you've
got
your
own
self-managed.
So
again
we
could
run
this
on
kubernetes
containers
within
AWS
gcp,
whatever
your
choice
may
be,
but
we
also
offer
a
fully
managed
service
and
that's
called
capella,
so
the
other
choice
you've
got
is
the
administration
path?
A
How
do
you
want
to
administer
this?
Do
you
want
to
administer
this?
You
know
these
are
the
kinds
of
questions.
Where
is
your
time
as
a
developer
or
an
architect
best
spent?
Is
it
best
spent
working
on
business
problems,
or
is
it
best
spent
running
backups
and
all
the
rest
of
that
kind
of
thing?
What
is
most
advantageous
to
you
and
your
business
so
again
lots
of
options.
You've
got
the
self-managed
option.
You've
got
the
semi-managed
option,
which
is
where
the
autonomous
operator
comes
in
again.
A
A
And
you
can
sign
up
for
a
free
30-day
trial.
Okay,
it's
not
going
to
be
fully
featured
because
we're
not
going
to
want
you
to
use
all
our
compute
power,
but
you
can
use
it
for
30
days
to
see
if
this
is
something
that
you
want
to
try,
and
you
can
also
hook
it
up
to
our
live
playground
as
well,
which
gives
you
examples
of
how
to
use
the
10
different
sdks
that
we
support
as
well.
Okay,
you
have
a
choice
of
how
you
want
scalability
to
work.
A
Every
platform
says
yeah,
we're
really
really
scalable
again
I'm
from
a
SQL
Server
background,
SQL
Server
scalability
is
the
fact
that
you
could
support
hot
ad
CPU
and
hot
ad
memory.
Great
I!
Don't
really
want
to
pull
a
server
out
of
a
rack
and
try
and
do
something
while
it's
live
risk
electrocuting
myself,
you
know
not
not
a
fun
place
to
be,
but
it
is
also
a
monolithic
application.
So
you
can
only
scale
it
up
or
scale
it
down.
A
I
can't
scale
in
or
out,
which
is
something
that
you
can
do
with
a
decentralized
platform,
and
we
run
it
in
a
masterless
configuration
which
means
there's
no
master
bottleneck
so,
as
you
scale
out
it
scales
in
a
linear
fashion,
there
are
others
in
the
nosql
space
that
do
have
a
Master
Slave
architecture,
because
this
is
being
recorded.
I
won't
mention
them,
but
that
does
mean
that
they
have
problems
when
they
scale
out.
They
don't
get
that
linear,
scalability.
A
You
have
the
choice
of
using
something,
that's
familiar
or
if
you
have
a
Netflix
subscription
doing
something
with
stranger
things:
okay,
so
who
here
has
used
SQL
server
or
Oracle
or
postgres,
or
any
kind
of
ANSI
language
yeah
within
five
minutes,
you
could
be
doing
whatever
you
want
within
couchbase
from
a
programming
perspective
at
the
database
layer,
because
we
support
all
of
that
with
our
SQL
plus
plus
language.
In
fact,
we
go
further
because
we've
got
support
for
arrays.
You've
got
all
the
indexing
and
all
that
kind
of
stuff
that
you
would.
A
A
Some
people
call
it
best
of
breed,
but
the
idea
is
that
you
choose
whatever
you
want
for
that
particular
area,
but
you
don't
always
know
what
you
need,
so
you
might
start
with
dynamodb
or
another
nosql
platform
and
you
might
find
out.
Oh
that's
not
really
scaling.
We
maybe
we
need
some
kind
of
in-memory
caching
layer
on
top
and
then
you
move
from
there
to
saying,
oh
well
as
this
application
has
grown,
we
need
to
do
full
text
search
because
there's
a
new
project
which
requires
us
to
do
x,
y
and
z.
A
So
maybe
we
need
to
bring
in
elasticsearch
or
a
another
component
for
that
and
it
goes
on
and
on
and
on,
and
the
problem
you
have
with
this
approach
is
that
if
you
have
a
problem
which
of
these
vendors,
do
you
go
to
because
they'll
all
blame
each
other
saying
it's
not
us
so
that
that
blame
game
kind
of
scenario
so
with
couchbase.
We
offer
all
of
this
within
one
data
platform.
If
you
want
lightning
speed,
we
can
do
everything
in
cash.
A
A
Coming
from
a
relational
background,
no
problem,
we
support
Scopes
and
collections,
which
are
the
same
as
schemas
and
tables.
So
you
could
map
one-to-one
if
you
wanted
to
we'd
recommend
that
you
read
model,
but
you
could
do
that
if
you
want
to
you've,
got
that
SQL
plus
plus
capability,
so
you
can
go
directly
from
one
to
another
again,
if
you
want
to
full
text
search
built
in
from
the
ground
up
again,
I
can
use
that
within
SQL,
plus
plus
one
defining
language
between
the
multiple
different
components
that
we
have.
A
Another
component
that
uses
SQL
plus
plus,
is
the
analytics
engine.
Anybody
used
Azure,
synapse,
synapse
kind
of
like
a
multi-parallel
processing
system,
similar
thing,
okay,
uses
MPP
architecture
and
we
can
pull
data
in
from
Azure
blobs
or
S3
buckets
Etc.
Okay,
there's
also
a
offline
first
mobile
embedded
database
that
we
offer
as
well.
It's
not
on
this
slide,
but
we've
got
lots
of
customers
who
use
that
as
well.
So
I
have
literally
70
seconds
left,
so
I'm
quite
impressed
with
that.
So
that's
it
for
me.