►
From YouTube: CNCF Storage WG Meeting - 2018-04-11
Description
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
A
I
get
Mormon
good
morning,
everybody
just
one
more
minute
and
we'll
get
it
kicked
off.
Actually,
it's
8:02
Kevin,
let's,
let's
get
going
here,
I've
got
a
you
know
terms
of
the
agenda.
We've
got
two
things
that
we
want
to
do
this
morning.
We've
got
first
of
all,
we've
got
Kevin
from
Tyco,
a/v
who's
gonna
be
presenting,
and
then
we've
also
got
the
follow-up
item
of
the
cube
con
sessions.
A
So
it's
getting
kind
of
hot
and
heavy
on
getting
those
done,
and
so
we're
gonna
make
sure
that
everybody
signed
up
and
we
set
up
some
some
groups
to
start
collaborating
I.
Think
that
the
next
session
that
we
do
in
two
weeks
cube
con
is
going
to
be
dedicated
just
reviewing
the
sessions
and
collaborating
a
bit
more
on
them
together.
So
don't
expect
a
presenter
next
week.
A
A
Great,
so
let's
let's
go
ahead
and
kick
it
off.
You
got
about
thirty
minutes
and
then
we
gotta
get
to
the
next
agenda
item.
Okay,.
B
Sounds
good
Thank,
You
Clint,
so
my
name
is
Kevin
Shu
I'm
from
the
company
pink
cap
and
the
company
is
ahead
of
us
strategy
and
operations
and
both
myself
and
Ed
Wong.
Who
is
our
co-founder
and
CTO?
We'll
be
doing
this
presentation
and
taking
your
questions
afterwards
about
Thai
kV
and
is
actually
dialing
in
from
Beijing
I.
Think
it's
like
11
p.m.
over
there
and
you
want
to
say
hi.
C
B
Alright,
so
I
will
just
launch
into
Thai
KB,
which
is
an
open
source,
distributed
transactional,
key
value
store.
This
is
again
one
of
the
projects
that
ping
captain
company
is
making
right
now
and
let's
get
started
alright.
So
a
quick
agenda
to
our
presentation
today
before
we
dive
into
the
whole
thing,
do
a
really
quick
history
of
the
company
and
the
product.
B
Kai
just
stands
for
titanium.
There
are
a
few
main
components:
the
Taibbi
platform,
one
is
Taibbi
itself,
which
is
a
stateless
sequel
layer
that
is
compatible
with
my
sequel
protocol.
The
other
one,
of
course,
is
the
focus
of
this
presentation,
which
is
ty
kV.
You
know,
ty
key
value,
essentially
is
what
it
stands
for,
is
a
distribute,
transactional.
B
Key
value
store,
we
also
have
a
project
called
tight
spark,
which
is
a
spark
plug-in
that
allows
a
lot
of
our
users
to
run
more
complex,
OLAP
queries
directly
on
top
of
ty
kV,
and
another
component
is
called
placement
driver
which
is
kind
of
this
meta
store
component
cluster.
That
does
a
lot
of
the
auto
scheduling,
as
well
as
load,
balancing
and
works
very
closely
with
ty
kV,
and
everything
that
we
have
been
doing
so
far
was
open
source
from
day
one.
We
open
sourced
hi
kV
on
April,
1st
2016
and
the
whole
projects.
B
Current
version
is
at
2.0
RC
4,
since
we
did
start
open
source
from
day.
One
we've
been
very
well-received
and
pretty
popular
in
the
community.
The
ty
BB
Club
repo
itself
is
more
than
half
thousand
stars.
Ty
kV
itself
is
closing
in
on
3000
stars.
We
also
have
quite
a
few
contributors,
not
just
folks
who
are
working
for
us,
but
people
from
really
all
over
the
world.
There
are
also
a
few
I
will
call
them
institutional
contributors.
B
That
I
want
to
highlight
real
quick
companies
like
Samsung,
like
$0.10
cloud,
which
is
one
of
the
larger
cloud
providers
in
China,
also
Alama.
The
use
case
that
we're
highlighting
is
contributing
to
you
know
the
type
Gabey
project,
as
well
as
companies
like
mobike,
which
is
one
of
the
largest
bike
sharing
companies
and
also
total
calm,
which
is
a
popular
news.
Aggregator
app
in
China
and
of
course
we
are
popular
I.
Think
because
we
are
trying
to
solve
a
pretty
pressing
pain
point
that
a
lot
of
organizations
and
companies
are
looking
for
as
we
accumulate.
B
So
we
actually
got
our
initial
inspiration
from
Google
spanner
the
spanner
paper,
which
I'm
sure
many
of
you
are
familiar
with.
Unfortunately,
the
spanner
was
not
open
sourced
and
still
isn't
open
source,
but
Thai
KBS
and
weari.
We
really
imagined
and
design
Thai
KB
to
be
a
fundamental
building
block
that
could
simplify
building
other
systems
more
interesting,
useful
systems
on
top
of
it,
and
so
far
we've
build
high
DB,
anti-spark
ourselves
and
Toto
calm,
which
is
this
news.
B
Aggregator
applicant,
that's
actually
valued
at
twenty
billion
dollars
in
China,
so
I
probably
shouldn't
call
them
a
start-up
anymore.
They
use
Thai
kV
as
their
metadata
service
for
their
own
kind
of
s3
architecture
and
oh
them,
which
I'll
go
into
a
little
bit
later
on
made
their
own
Redis
protocol
layer,
but
essentially
with
Thai
KB
being
the
building
block.
We
really
do
think
the
sky
is
the
limit
in
terms
of
what
you
can
build
on
top
of
them.
B
B
Now,
let's
dive
deeper
into
the
ty
kV
architecture,
we
have
always
imagined
it
to
be
a
standalone
distributed,
key-value
store
so
right
now
any
client
really
can
use
g
RPC
to
directly
connect
and
communicate
with
ty
kv
clusters.
Your
RPC,
of
course,
is
a
ciencia
project,
so
we
are
heavy
users
of
that
it
has
two
api's
in
the
top
one
is
a
transactional
kv
api.
Another
one
is
a
coprocessor
API
that
we've
designed
to
do
a
lot
of
push
down
to
push
down
sequel
logic
into
the
ty
kv
instances
to
be
processed.
B
We
also
apply
the
RAF
consensus
protocol
to
do
our
replication
to
provide
strong
consistency
as
well
as
a
high
availability
and
underneath
each
ty
kv
instance.
You
can
imagine
every
single
one
of
them
essentially
also
have
its
own
rocks.
T
the
instance
that
takes
care
of
the
storage,
and
this
is
just
one
example
of
how
we
build
heidi,
be
on
top
of
type
kv.
Tidy
beeping
the
my
sequel,
compatible
sequel
layer,
but
you
can
really
plug
in
all
kinds
of
different
or
ms
or
other
database
connectors
onto
it.
B
Or
you
know
the
connected
blob
storage
instances,
and
this
is
actually
how
Co
Telecom
is
using
ty
kV
for
their
own
metadata.
Again
raft
is
in
play
as
well,
so
a
quick
summary
of
the
technical
highlights
of
ty
KB.
So,
as
I
mentioned,
it
does
scheduling
and
auto
balancing
working
with
the
placement
driver
cluster.
It
has
this
multi
raft
implementation,
multiple
raft
groups
inside
one
ty
kV
instance.
B
It
also
does
dynamically
range
space
partitioning
using
splitting
and
merging,
and
this
is
how
we
resolve
hotspot
issues
and
I'll
go
into
that
in
a
little
bit
as
well
as
a
transaction.
Of
course,
we
use
the
two-phase
commit
with
optimistic
lock
and
the
whole
project
for
those
of
you
who
are
curious
is
written
in
rust.
So
no
garbage
collection
style
time
as
well
as
run
time
cost
either.
B
B
B
So
again,
this
is
one
of
the
features
we're
at
ikv
working
with
the
placement
driver
can
do
this
dynamically
and
kind
of
the
metrics
that
we
use,
or
the
configuration
we
used
to
do
splitting
is
if
a
region
size
exceeds
96
megabytes,
which
is
our
default
value.
Of
course.
This
is
something
you
can
change.
You
know,
depending
on
your
needs,
then
it
split
will
happen
to
avoid
a
region
being
too
large
or
forming
potential
hotspots
down
the
line
and
for
merging
it
kind
of
works
the
same
way.
B
B
You
know,
boost
the
performance
and
also
the
user
experience
of
the
entire
Tyco
heavy
deployment.
So
we
use
Prometheus
GRDC
like
I
mentioned,
etcd
grow
fauna.
We
are
actually
the
maintainer
of
Prometheus
and
G
RPC
implement
in
mutation
arrests.
So
we
have
contributed
that
implementation
back
to
the
community
as
well
and
are
very
actively
developing
those
two
things.
B
B
So
now,
I
would
do
a
really
quick
dive
into
how
a
llama
uses
Thai
keys.
So,
like
I
mentioned,
it's
a
food
delivery
platform,
kind
of
similar
to
your
door
or
your
post.
Mates
quar
by
Alibaba
is
currently
serving
about
260
million
users
and
the
bulk
of
his
data
is
in
key
value
format
and
before
they
adopted
hi
Katie.
B
They
just
used
a
hodgepodge
of
solution
from
Cassandra,
my
sequel,
Redis
and
so
on,
and
so
forth
made
it
to
make
it
work,
but
they
were
looking
for
a
single
unifying
storage
layer,
so
they
tested
out
high
kV
and
right
now
they
deployed
high
kV.
In
you
know,
ten
plus
clusters
spread
out
in
four
different
data
centers
more
than
a
hundred
nodes.
B
So
with
that
said,
I'm
gonna
do
a
really
quick
demo
to
kind
of
show
you
how
I
kb
works
in
action.
I
did
make
my
tribute
to
demo
God
this
morning,
so
hopefully
everything
will
go
smoothly
and
to
give
a
quick
context
to
this
demo.
So
what
I
would
do
is
launch
a
tidy,
B
cluster
and
kind
of
show
you
how
that
works,
which
is
very
simple,
using
docker
compose
and
then
also
launched
a
my
sequel
instance
and
a
spark
or
theis
Park
sequel
instance
on
top
of
this
cluster.
B
That
is,
of
course
undergirded
by
ikb,
and
we
will
just
do
some.
Real-Time
analytical
queries
on
top
of
it
to
see
how
it
enables
kind
of
this
hybrid.
You
know
real-time
data
data,
warehouse
experience
all
right,
so
I've
already
get
cloned
our
dr.
Campos
for
tyki.
So
let's
just
launch
this
thing,
so
buh-buh-buh-buh-buh
everything
is
done
and
before
we
write
some
queries,
I
want
to
show
you
guys
a
couple
of
things
that
are
really
cool
one.
B
Is
we
use
Cortana
as
the
the
monitoring
service
for
a
tightly
given
cluster,
so
we've
defaulted
back
to
42,000
and
admin,
the
username
and
password
for
this
is
just
admin
admin
very
secure,
and
here
you
can
see
how
you
can
monitor
a
type
of
your
cluster,
so
the
overview
not
really
that
interesting,
because
nothing's
really
happening.
But
if
you
go
to
Tyco,
hey
B,
you
can
check
out
the
cluster.
You
can
see
the
available
size,
the
storage
size,
the
capacity
or
things
like
that.
B
B
This
is
just
a
point
onto
my
local
machine,
so
you
know
that
whatever
the
numbers
here
probably
corresponds
to
that
and
another
cool
tool
that
I
will
show
you
real
quick.
This
is
something
we
made.
Ourselves
is
a
visualization
tool
of
the
whole
cluster.
So
that's
defaulted
on
to
port
80
10.
This
is
thing
called
high
BB
vision
see
if
it
works.
B
So
let
me
just
do
a
few
commands
here
to
JDM
it
a
spark
in
and
then
we
will
launch
our
smirk
so
we'll
give
that
a
minute,
and
we
will
also
launch
nice
equal
and
again,
all
this
instruction
and
commands.
We
have
on
a
github
and
I'll
share
that
link
with
you
afterwards
as
well.
So
we
launched
my
sequel
and,
as
you
can
see,
it's
directly
connected
to
the
current
version
of
Tidy,
be
that
I,
just
clone
down
to
my
local
machine.
Tidy
B
as
a
whole
can
essentially
right
now
serve
as
my
sequel
slaves.
B
So
you
can
keep
on
writing
your
my
sequel
code.
However,
you
want
and
enjoy
the
scalability
that
Kai
DV
brings
to
you
underneath
it,
let's
see
if
okay
so
spark
is
up
and
running
so
now,
I
just
need
to
import
ty
spark,
which
is
a
plugin
that
we
made
for
to
leverage
spark
on
top
of
Taiki,
be
alright
and
the
table
or
the
data.
B
The
sample
data
that
we'll
be
playing
around
with
is
in
this
thing
right
now,
so
let
me
just
show
you
what
is
databases
in
here,
so
we
will
be
playing
around
with
this
TCT
PCH
0:01
database,
which
is
just
like
a
set
of
sample
data
that
we
play
around
with.
So
if
we
do
actually
use
age-1
now
we're
using
that,
let's
bind
that
with
our
type
spark
instance
as
well
all
right,
so
that's
good!
So
now
both
of
these
instances
tie
spark,
and
my
simple
should
be
seeing
the.
B
B
And
you
see
the
same
list
of
countries
all
right
and
we
can
actually,
of
course
run
any
level
of
complexity
of
queries
in
here.
I
have
one
Pete
up
here.
That
is
particularly
gnarly
with
a
few
group
buys
and
order
buys
and
conditionals
in
here
on
the
my
sequel
side,
this
is
coming
from
the
line
item
table,
so
this
will
give
us
some
kind
of
data
and
let's
make
sure
that
the
spark
side
sees
the
same
thing.
So
this
is
the
sky
spark
equivalent
of
that.
B
Okay,
we'll
do
that.
This
might
take
a
couple
of
seconds
extra,
so
you
see
the
same
data
here
over
here,
just
format
it
a
little
bit
differently.
Okay,
now,
let's
say
we
want
to
change
something
on
this
data
set
again.
This
data
is
all
stored
in
PI,
K,
V
notes
right.
So
let's
do
like
an
update
of
the
nation
database
sets,
let's
do
the
animation
key
animation
e
equals
one
one
two
two
and
who
do
we
pick
on
today,
so
from
where
nation
equals
who's
the
to
two
to
Brazil
all
right.
B
B
Now,
let's
see
if
the
sparks
I
sees
the
same
thing
so
far,
sequel
select
star
from
nation,
so
plenty
and
voila
Brazil
is
also
two
two
two
because
they're
drawing
from
the
same
data
source
and
you
can
do
basically
simultaneously
transaction
and
analytical
processing
on
the
same
live
data
set
enabled
by
Thai
Kb,
alright.
So
that
is
my
demo
and
to
just
wrap
things
up
real
quick
again.
We
always
want
to
take
a
view
to
be
on
its
own.
B
A
E
C
C
B
A
B
Right
now,
the
the
in
production
deployment
tools
that
we're
using
is
a
ansible
deployment
which
can
be
used
on
really
any
cloud
service,
whether
you
know
San
cloud
or
AWS
over
here
and
we
are
still
working
towards
having
Thai
kV
deployment,
be
fully
fully
integrated
with
you
know
the
entire
toolset
that's
available
in
kubernetes,
so
the
Tybee
be
operator
that
I
mentioned
is
one
of
those
projects
that
were
working
on
right
now,
and
you
know
hopefully
in
this
project
were
to
be
accepted.
Then
you
know
that
could
be
accelerated
as
well.
So
we're
definitely.
A
Are
you
thinking
about
that
operator
like
what's
the
scope
of
that
in
terms
of
the
quote-unquote
like
day
2
operations,
because
I
can
see
it
being
beneficial
for
getting
it
up
and
running
and
just
replacing
ansible
for
day
one?
But
alright,
are
you
guys,
thinking
about
the
scale
out
capabilities
of
the
database
and
leveraging
the
operator
to
do
more
than
just
a
one
activities
and.
C
Ok,
yeah,
the
the
whole
idea
of
Tidy
B
operator
is
from
etcd
operator
it's
as
handled
deployment,
skill
out
and
out
of
favor
for
the
whole
tidy
B
project,
not
only
for
the
out
not
only
for
the
tightly.
We
part,
but
it
is
on.
Our
plan
to
you
know,
act
extract
the
deployment
attack
every
deployment,
operational
knowledge
to
another
operator
project,
so
it's
kind
of
like
EDC
the
operator,
the
tied
V
version
of
it
is
the
operator,
but
it
is
not
open
sourced.
B
Right
I
have
a
couple
points
on
that,
and
you
know
ed
you
can
chime
in
and
feel
that,
and
it's
really
any
more
details
as
well.
So
I
like
terms
of
the
architecture
of
cockroach
DB,
compared
to
what
we
have
made
here,
you
know
ty
KB,
our
key
value
part
is
completely
separate
and
therefore
can
be.
You
know,
moved
around
and
pluggable
from
different
systems
on
top
right,
which
is
kind
of
this
larger
point
that
we
hopefully
we
made
during
this
presentation.
B
B
They
are
still
more
strictly
a
allowed
GP
scalable
solution
to
my
knowledge,
so
you
know
I
think
we
both
drew
our
inspiration
from
the
Google
spanner
paper
back
in
the
days,
but
we
definitely
have
a
lot
of
differences
between
us
and
you
know:
there's
like
this
more
superficial
compatibility,
you
know
we
are
compatible
with
my
sequel
and
cockroach
is
compatible
with
post
quest.
So
you
know
different
people
kind
of
use
us
in
different
ways.
I
don't
know
at
if
you
have
anything
to
add
to
that
yeah.
D
A
A
A
All
right,
everybody
so
onto
the
the
next
topic,
so
we've
got
cube
con
coming
up.
Pretty
quick.
We've
got
our
two
confirmed
sessions
for
cube
con,
so
the
first
is
a
introduction
session
from
the
swg
and
then
the
second
is
a
more
advanced
session
in
previous
meetings
that
we've
had.
We
decided
on
topics
for
each
of
the
sessions.
A
So
that's
the
the
intro
session
that
we're
trying
to
plan
and
then
there's
an
advanced
session
where
we've
been
we've
asked
the
the
TOC
to
give
us
feedback
on.
You
know
what
they
want
from
us,
so
what's
what's
a
definite,
what's
essentially
the
definition
of
cloud
native
and
how
to
storage
fit
into
that,
and
was
the
TSC
me
from
the
swt
you
know:
what's
that
chart
are
gonna
be
for
us
for
the
next
six
months
or
or
whatever
they
really
want
from
us
and
so
to
two
sessions.
A
The
first
session
is
what
I'd
like
to
get
people
together
to
to
plan
in
in
previous
meetings.
We
had
five
folks
that
signed
up
to
be
a
part
of
those
planning
sessions.
So
aside,
I
think
you
stepped
up,
or
it's
been
Hindman
and
Steve
from
Red
Hat.
Is
there
anybody
else
on
this
call
that
would
like
to
help
us
out
in
creating
the
set
the
intro
session
for
cute
gonna
be
a
part
of
that
Quintin
I.
Think
I
saw
your
your
name
pop
up
on
the
intro.
Are
you?
F
D
A
Okay,
so
lots
of
folks,
it
looks
like
we
got
seven
or
eight,
so
I
think
that
the
next
step
here
is
to
send
out
a
an
invite
but
say
a
poll
to
get
some
available
times
for
everybody
over
the
next
two
weeks.
I
feel
like
we've
got
probably
a
couple
to
a
few
sessions
that
will
try
to
organize
between
us
and
it'll,
make
some
progress
on
putting
together.
You
know
a
handful
of
slides
for
that
intro
and
then
figure
out
who
can
be
on
that
panel.
A
Okay,
cool
all
right,
so
intercession
looks
like
we're
covered
I.
Think
that
the
next
swg
meeting
that
we
have
in
two
weeks,
hopefully
we
can
have
made
enough
progress
there
to
discuss
it
with
the
team.
Sorry
about
the
background,
noise
and
and
then
we'll
go
from
there
and
then
following
that
is
gonna,
be
the
cue
card
for
the
session.
I.
A
A
Have
not
gotten
the
TOC
folks
to
to
confirm
so
I'm
gonna
reach
out
to
them
again
in
the
background
what's
been
happening,
is
that
Camille's
been
talking
with
folks
on
the
swg
about
what
they
are
interested
in
and
or
you
know,
fact
how
that
has
to
be?
Is
everybody
is
doing
and
what
they
want
to
see
it
turn
into
etc,
and
so
we're
hoping
to
get
feedback
from
Camille
on
that
so
I
think
at
a
minimum.
We'd
have
Camille
talking
to
us
and
hopefully
other
folks
in
the
TOC.