►
From YouTube: This Week in Cassandra 02/22/2016
Description
Link to blog: http://bit.ly/1PK8ncs
A
A
A
General
so.
C
A
C
A
We
go
hair
and
beard
coming
at
you
all
right.
Let's
take
a
look
at
what
we
have
today.
A
B
You
that's
like
Achilles
is
one
of
those
things
like
I
I,
really
liked
it.
It
stays
up-to-date.
The
driver
has
some
excellent
stuff
like
codec
support
and
provides
features
like
an
interceptor
API
and
like
built-in
JSON
serialization
to
civilization,
like
I've,
almost
wanted
to
go
back
and
open
the
issue
and
say
that
I
do
we
highs
too
humble,
but
I
guess
some
problem.
B
D
A
C
E
C
You
can
do
that
and
that's
not
unheard
of
in
the
Cassandra
community,
so
Avinash
she's,
one
of
the
original
people
who
came
up
with
Cassandra.
You
probably
remember
this
on
the
mailing
list
was
there
was
a
guy
on
the
mailing
list
that
always
had
the
right
answer,
but
it
wasn't
Avinash.
It
was
Bill
yeah,
Hastings.
E
C
A
A
B
D
E
Watched
the
video
yeah-
it
is
quite
interesting,
so
there
was
the
a
live
demo
with
Cassandra
and
dippin
soldiers.
They
use
the
zip
in
Cassandra
interpreter,
and
so
what
is
nice
is
that
they
they
showed
you
the
query
execution
so
that
every
five
second,
it
refreshes.
So
every
five
seconds
you
can
see
your
timeline
updated
from
new
data
coming
in
fine,
but
then
it
just
hangs
like
the
application.
The
Zeppelin
thing
just
stopped
working.
Why?
Because
they
were
running
on
them
on
one
vm.
E
A
A
B
What
I
like
about
his
post,
the
most
is
it
you
have.
You
know
to
like
nine
line
snippets
on
the
bottom,
and
it's
there
that
Braavos,
because
it's
well
formatted
it
will
foreigner
of
readability.
Basically,
this
is
how
you
copy
the
contents
at
the
table
from
one
cluster
to
another
and
even
like
tada,
you
know,
and
I
should
mention.
C
It
she
just
a
shout
out
to
rest
he's
one
of
the
ease,
the
top
contributors
to
to
stack
overflow,
spark
and
Cassandra
questions,
and
his
answers
are
encyclopedic.
It's
like
the
Wikipedia
of
knowledge
when
stuff,
because
he
doesn't
just
say
yeah.
Do
it
this
way,
he's
like?
Yes
do
it
this
way.
Here's
three
examples
of
illustrate.
You
know
these
wonderful
ways
of
how
this
works,
and
it's
just
so
complete.
Well,.
B
B
A
Okay,
can't.
B
The
before,
like
meet
up
tonight,
is
doing
spark
and
Cassandra
stuff
also
for
people
in
Portland
period.
No
next.
A
Yes,
this
is
this
is
actually
it's
not
even
like
it's
not
just
like
an
app
that
runs
on
top
of
it.
It's
it's
like
way
more
intense
than
that,
like
they're
they're,
building
like
they're
like
yeah,
we're
gonna
build
like
a
real
like
geo
database
using
Sandra's
as
a
building
block,
which
is
pretty
intense,
scary.
A
C
E
C
A
E
C
E
A
C
B
Air,
it's
a
really
good
author
on
the
stuff.
He
goes
into
like
extreme
detail,
and
it's
just
one
thing
after
another.
You
can
follow
along
at
home
with
the
tutorial
with
the
example
style
of
it.
He
links
to
all
the
code.
There's
links
to
like
documentation
and
blog
posts
were
it's
relevant,
so
it's
one
of
cigs
like
when
we
do
this
stuff.
B
We
try
to
do
it
like
as
thorough
as
possible
and
like
this
is
a
good
sort
of
walk
through
of
my
first
experience
with
the
storage
engine,
doing
something
concrete
like
dropping
columns
and
like
lo
and
behold,
we
produced
a
bug
so
which
is
good
Oh
11,
0,
5
0,
which
aab
is
in
the
process
of
getting
worked
out
right
now,
there's
a
workaround
for
it,
but
not
to
be
a
little
bit
more
complex
with
the
way
that
calm,
the
schema
changes
to
our
hashes
and
the
system
tables.
So
that
that's
read
the
issue.
B
A
I,
you
know,
I
have
two
things:
I'm
gonna
stay
here.
One
I
agree
with
what
you
were
saying:
I'm.
The
first
time
I
saw
Aaron
speak
was
at
the
Cassandra
summit
like
two
or
three
summits
ago,
and
he
was
just
talking
about
the
engine.
Internals
and
I
was
like
yes
like
love
me
smear
and
Morton,
like
his
attention
to
detail
is
ridiculous.
I
love
it
he's
just
like
okay
and
now
we're
gonna
dive
into
this
java
class,
and
this
is
what
it
does,
and
this
is
how
it
works.
A
Here's
what
you
need
to
know
and
I
was
like
like
thank
you
for
not
like
you
know,
just
giving
me
like
this
cursory
overview
is
just
super
in-depth
storage
engine
yeah.
B
D
Other
thing
this
meant
this
blog
post
mentioned
was
SS
table
tools,
which
this
is
the
first
time
I'd
ever
seen
a
link
to
that
tool
before
I
didn't
realize
it
existed.
You
know,
because
that's
this
table
to
JSON
was
was
deprecated,
I
think
with
the
30
release.
So,
like
that's,
that's
kind
of
cool
like
to
know
that
tool
exists.
Something
for
you
know,
kind
of
fill
the
gap
know
about
them.
B
This
is
it
does
a
good
job
of
like
actually
showing
the
new
format
in
the
sort
of
native
way,
and
if
you,
if
you're
curious
about
what,
like
code
to
take
that
apart,
it's
pretty
clear
usage
of
like
the
SS
table
reader
infrastructure.
Now
that
pants
down
to
everything,
so
you
can
actually
follow
trace.
That
along
is
a
good
way
to
get
your
and
aromas
yeah.
Mr.
C
E
E
E
B
This
doesn't
sort
this
sorely
a
matter
of
like
formatting,
errands,
copious
and
notes.
Let's
taste
like
wrote
for
this
one,
please
needs
him:
hey.
B
C
B
It
I
I
think
it's
fun.
I
I,
like
getting
information
out
there,
I've
had
since
I
started
working
in
the
community.
So
it's
just
it's
something
like
to
do.
Yeah.
A
Blogging
is
a
good
time
you
gotta
I
kind
of
similar,
a
similar
vein
to
what
you
know.
We
were
saying
about
Aaron's
posts
like
you're
you're,
walking
down
patchy,
Cassandra
logging.
You
know
we
talked
a
little
bit
about
that
in
the
last
this
weekend.
Cassandra
huge
I
love
it
like
you
hear
people
talk
about,
oh
just
whatever.
B
Like
that,
the
writing,
for
me
helps
me,
tie
this
back
together,
two
ways
in
which
I
would
present
this
to
users
who
might
be
new
to
Cassandra
and
and
technical
topics
in
particular,
because
you
know
this
it
it's
practice
for
me
in
our
professional
work
and
what
we
do.
But
on
the
other
hand
it
just.
It
keeps
me
close
to
looking
at
it
from
somebody
else's
perspective.
C
B
B
C
C
Well,
you're
you're
gonna
find
edge
cases
and
you're
gonna
like
why's
that
and
then,
but
then
you're
also
going
to
come
up
with
that.
A
nice
clean
way
of
describing
it
because
you
could
just
talk
about
something
random
ways
and
if
you're
having
a
conversation,
it
works.
But
if
you're
trying
to
present
it
to
anybody
any
number
of
greater
than
two.
Then
it's
in
trouble.
Yep,
yeah,
yeah.
A
What's
there
are
alternatives
to
datastax
ops
center,
so
I
think
it's
with
the
newest
version
of
ops
center.
That's
going
to
be
DSE
only
there's
a
month
of
some
features
that
that
require
it
or
it's
just
going
to
be
part
of
the
data
sex
enterprise
package.
So
this
is
really
just
about
furin
operations.
How
to
monitor
your
cluster
I
think
this
is
definitely
worth
a
good
read,
I'm
interested
to
see
what
comes
out
of
the
community,
for
you
know,
tools
specific
to
Cassandra.
B
Yeah,
we
were
big
fans,
agree,
fauna
and
we've
tried
a
couple
different
backends
with
it's
just
too
the
thing
that
we
know
that
works,
but
is
the
graphite
back
end
and
it
does
a
pain
to
scale,
because
it
is
the
storage
architecture.
Once
you
want
to
go
big
for
even
the
modders
moderate
sized
cluster,
you
have
to
federate
graphite,
which
is
non-trivial
and
as
you
can't
it's
really
difficult
to
do
like
running
container
cuz
graphite
pre
allocates
files,
so
you
try
to
run
gratifying
container
and
maxes
out
like
that.
B
The
24
gig
limit,
whatever
is
a
docker
cuz
of
all
the
Cassandra
metrics.
It's
just
like
I
know,
pre-allocate,
like
all
the
stuff
and
Dockers
like
it's
like.
Never
thought,
sup
are
you
so
you
can
fix
it,
but
it's
just
one
of
those
things
like
if
you're
new
to
docker,
then
you
don't
know
that
you
need
to
set
your
runtime
with,
like
the
extended
storage
flags,
all
that
so.
E
B
So
many
not
something
you
do
in
container
per
se,
but
when
you
launch
your
container
watcher
thing,
so
it's
this
one's
a
little
things,
but
there's
a
a
couple
things
that
we've
tried
that
cry:
kyros
DB.
This
is
now
supported
as
a
back-end
for
her
fauna.
There's
a
couple
other
ones
out
there.
The
open
TS
is
DB.
This
supports
a
Kraken
back
and
figure
fauna
now
and
we've
used
in
flux.
Before
on
a
single
node.
It
works
great
on
a
single
node,
but
man.
Nothing
else
is
use.
B
When
you
try
to
cluster
it.
That's
not
quite
there
yet,
but
the
influx
queries
is
are
probably
fastest.
When
you
want
to
do
like
you
know,
metric
name
star
for
like
a
host,
then
you
know
other
metric
name.
It
does
sort
of
embedded
star
queries,
probably
the
fastest
and
what
anything
else
we've
seen
so
far.
Cool.
B
E
C
B
A
Right
next
up
on
the
list,
time
series
data
modeling
for
medical
devices,
I
Danny,
traphagen,
datastax,.
B
A
And
this
is
definitely
going
to
be
a
field
that
just
is
not
getting
smaller
over
time
like
we're,
probably
not
going
to
see
people
be
like
whoa
I,
don't
know.
If,
like
we
need
to
collect
information
to
make
decisions
about
like
how
much
insulin,
you're
given
and
and
things
like
that,
so
there's
debts
it's
you
know
it
can
only
be
a
growing
market.
It's.
D
Good
to
see
like
a
concrete
example,
too,
of
internet
of
things,
like
it
kind
of,
gets
mocked
a
lot
online
like
internet
with
you,
this
whole
concept
of
the
Internet
of
Things.
So
it's
good
to
see
like
something
that's
kind
of
applicable,
and
you
know
that's
actually
like
a
maybe
a
real
world.
You
case
like.
D
B
Not
always
be
recorded,
so
I
does
kind
of
funny.
I
was
like
wait.
What
the
video,
I
think
so
that's
a
that's.
A
alpha,
/
beta,
is
like
the
next
version
of
my
Cassandra
presentation
that
I've
been
doing
for
a
little
while
it's
got
some
new
content
in
their
arm
and
it's
a
lot
better
formatted
for
like
actually
presenting
that
put
the
existing
slides
are
up
or
previous
slides
are
up
with
a
bunch
of
detail
on
these
were
more
visual,
so
it's
actually
I
think
trying
to
scare
people
less
and
communicate
better.
B
So,
if
your
be
in
Atlanta,
if
folks
in
Elena,
I'm
headed
there
for
casino
days
and
march
and
I'll
be
doing
the
latest
version
of
this,
it's
gonna
cover
everything
from
like
all
stuff
epithelia
lately,
like
logging,
multiplication,
shin
from
disk
and
wire
level
encryption
as
well,
and
also
the
two
changes
coming
out
with
three
out
four
there's
commit
log
file,
encryption
and
the
hints
file
encryption,
because
since
you're,
still
in
a
file
now
and
Jason
Brown
from
Apple
did
both
those
things
and
they're
hitting
the
three
tough
work.
A
This
is
a
it's
a
group
setting
I'm
I,
don't
have
to
horn
yeah,
so
yeah,
that's,
but
it
just
goes
to
show
you
how
its
cool
feature
good
cept,
that
you're
thinking
about
it.
Well,.
B
Because
we
we've
done
that
point,
you
have
to
get
to
tell
people
use
encrypted
fest
or
something
like
a
filesystem
encryption
like
you
could
do
or
you
do
like
a
separate
mounted
encrypted
file
system
completely
for
that.
But
you,
corrupt
FS,
is
just
a
single
like
file
on
krypter.
So
we've
done
that
before
for
people
they
want
it
like
encrypt,
just
commit
log,
or
you
know,
have
something
else
to
an
Amazon
EBS
encryption
for
data
river,
so
yeah.
So.
C
D
A
B
There's
a
series
of
questions
that
somebody
did
a
couple
weeks
ago
that
had
to
do
around
the
the
I
forget
the
name
of
the
standard
but
they're
doing
like
the
mil
spec
government
compliances
did
yeah
yeah
that
yeah
and
that
that's
interesting
so
like,
if
somebody's
actually
doing
a
full-scale
audit
of
that
and
they're
putting
into
the
sort
of
the
database
they
have
their.
That
stuff
is
gonna,
make
a
lot
easier
for
you
know,
three-letter
agencies
to
adopt
this
stuff,
listen,
the
US
and
their
equivalents
abroad.
A
C
C
C
A
A
All
right,
I
think
will
shut
this
party
down
then
we're
at
about
25
minutes
so
more
stuff.
Coming
later,
there
always
definitely
check
out
the
blog
post
that
accompanies
this
video.
If
you
want
to
take
a
look
at
any
of
the
links
that
we've
talked
about,
there's
some
job
opportunities
and
upcoming
meetups.
So
there's
extra
information
on
the
blog
post
that
we
didn't
talk
about
here
so
definitely
check
that
out.
Alright,
let's
shut
this
party
down,
they
said.