►
From YouTube: 2020 05 05 Memory Team Weekly Meeting
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
All
right
so
I
send
a
note
last
night
and
hopefully
people
get
a
chance
to
take
a
look.
There
are
some
well
there's,
no
an
opportunity
that
we
need
to
fill
from
the
enablement
section,
specifically
from
your
memory
or
search
the
telemetry
and
fulfillment
teams.
They
need
to
augment
their
teams
to
work
out
some
some
high-priority
ICV
goals
for
the
coming
quarter.
I'll,
let
folks
read
the
details
there
what's
needed
on.
A
Our
end
is
I
need
to
hear
feedback
by
the
end
of
tomorrow
on
interest
level
for
folks,
and
it
could
be
anywhere
from
not
interested
at
all.
Yeah
can
I
start
tomorrow
right
and
give
me
some
honest
feedback.
Let
me
know
if
you
have
any
questions
for
those
that
are
watching
the
recording
later
on
just
find
some
time
to
chat
with
me.
I
need
to
hear
from
you
tomorrow,
sorry
for
the
short
notice.
A
This
was
expedited
to
us
as
well,
but
I
do
need
to
hear
from
everybody,
and
if
you
have
questions,
that's
fair,
please
reach
out
at
any
time.
Ask
me
questions.
If
I'm
not
available,
Sean
will
always
be
available.
Josh
will
have
information.
Any
one
of
us
can
answer
questions
about
the
opportunity,
but
just
let
me
know
which
way:
you're,
leaning
and
like
I
said,
and
it's
in
the
details
in
the
doc.
We
need
to
fill
one
position.
A
It
needs
to
come
from
enablement
and
if
there's
multiple
people
that
are
interested,
then
we'll
go
back
and
kind
of
figure
out
who's
the
best
fit.
Does
it
make
sense
for
them
to
move?
Is
it
the
right
opportunity
and
then
we'll
come
back
and
figure
out
who's
going
to
move
if
nobody's
interested
it'll
be
a
harder
conversation,
because
then
we'll
have
to
go
back
and
say
you
know,
we
do
need
someone.
A
D
A
One
one
person,
if
they're
interested
and
if
nobody's
interested
across
the
to
team,
so
there's
ten
people
total
across
memory
and
search,
and
we
need
one
person
who's
interested
in
switching
over
to
either
fulfillment
to
look
for
telemetry
and
they
can
express
an
interest
on
either
one
but
ultimately
I'm
not
sure
he
makes
that
decision
where
they're
gonna
land,
but
they
can
express
their
interest.
So
no,
it
could
come
from
the
search
team.
A
There
could
be
someone
on
search
team,
that's
incredibly
excited
and
wants
to
join
one
of
those
teams
to
go
work
on
what
they're
working
on
but
yeah.
It
could
come
from
the
memory
team
as
well.
Okay,
thanks,
but
there's
just
one
position:
one
person
who
needs
to
move
from
the
two
teams
to
go
help
out
with
this
opportunity
and.
E
Soumitra
team
and
they're
focused
on
improving
things
like
the
users
paying
to
make
sure
it
can
scale
for
for
Caleb
calm
and
to
make
sure
that
we
can
instrument
parts
of
our
applications.
That
right
now
are
challenging
the
industry
with
usage
pay.
But
so
that's
that's.
Some
of
the
reasons
there
is
we're
trying
to
get
more
data-driven
on
my
temperature
side
and
we're
trying
to
operate
get
live
calm
a
little
more
as
a
kind
of
a
healthy
salon.
Business
than
the
current
means.
F
A
Alright,
so
we
still
have
12
days
of
miles
in
the
milestone,
so
we
can
catch
up
next
week
on
what
we
should
add
to
the
release.
Clothes
there's
some
good
things
that
should
go
into
this
release
post
in
DJ's
on
atomic
processing,
we'll
get
more
we'll
get
some
issues
out
there
to
make
sure
that
we
summarize
those
for
the
release
post
em
our
totals
for
April
great
second
highest,
ever
44.
Oh.
A
Yeah,
okay,
our
thanks
for
the
feedback
we've
gotten
so
far,
I
haven't
had
time
to
actually
go
back
and
review
all
the
feedback.
There's
a
lot
going
on
right
now,
so
I
will
work
on
that
this
week
to
make
sure
that
we
incorporate
feedback.
I
like
I,
didn't
read
what
Mattias
said
about
getting
more
specific
on
the
proactive
development
piece
and
getting
some
actual
actionable
items
in
there.
A
So
I
will
revise
that
incorporate
the
feedback
takes
Mattias
see
the
12:10
retro
need
feedback
from
people
that
aren't
actually
in
this
meeting
right
now
so
Alexei
he
won't
get
his
end
in
time.
It'll
still
be
good
to
get
his
feedback
and
I
will
ping
the
coracle
indirectly
on
that
dimples,
like
the
new
medium
board,.
C
D
Didn't
get
reach
out
about
the
Patriot
documentation,
I'm
gonna
reach
out
tomorrow.
It
just
got
back
today,
so
I
only
responded
to
mass
of
the
conversations
I
want.
The
current
stuff
I
actually
did
push
yeah
another
iteration
of
ultras,
because
I
huge
nothing's
changes
so
change
it
so
I'm
hoping
that
maybe
image
that
I
am
still
like.
Looking
for
feedback
from
night
I
see
I
mean
its
cost
factors.
That's
on
production,
loving
the
change
from
that.
D
D
D
Mmm-Hmm
I'm
gonna
write
a
message
once
like
I
think
I
think
is
good
enough
for
now
the
pretty
like
well-thought
points
from
our
last
discussion,
so
like
maybe
getting
the
feedback
before
the
fuse.
All
Friday
was
being
actually
useful,
okay
to
not
cross
the
time
because,
like
we
don't
really
have
a
lot
of
time.
We
have
like
two
weeks
left
mm-hmm.
C
C
B
Yes,
so
that
that's
like
it's
in
review
Kamel.
Thank
you
again
for
looking
and
it
I
I
addressed
your
feedback,
so
fingers
crossed
it
should
be,
should
be
ready
to
go
in
now.
Test
of
this
as
well
looks
fine
I
think
we
can
I
think
we're
very
close
to
him
to
shipping
us
I
actually
spend
more
time
on
13.1
items
already,
because
there
wasn't
like
too
much
left
in
the
13.0
milestone,
so
yeah
I'm
also
spending
some
time
on
looking
to
this
yeah
into
why
exports
cuts.
Oh,
apparently
got
so
much
slower
in
this
pipeline.
B
B
D
I
may
be
gonna
add
to
the
dog,
because
I've
had
some
random
idea
and
when
we
started
like
report
u.s.
S&P
SS
into
Prometheus,
it
would
be
pretty
useful
to
push
as
much
data
about
my
moral
usage
application
configuration
into
our
usage
pings
understand
exactly
how
customer
instances
are
configured
and
what
is
their
memory
usage
to
understand
better
how
people
to
run.
D
A
D
B
So
I
only
really
touched
the
brief.
You
touch
the
surface
of
this,
because
I
look
at
an
MRI
that
was
related
to
it,
so
it
looked
like
so
so
we
I
think
we
shop
data
into
Redis
and
then
a
jump
runs
I
think
the
thing
I
looked
at
it
was
like
counting
I,
don't
remember
what
it
was
counting,
but
it
was
like
some
content
Redis,
but
we're
like
what
happens
to
this
data
and
like
where,
where
do
we
end
up
actually
collecting
it?
I.
D
I
think
it's
like
key
value,
storage
or
something
that,
like
you,
can
use
an
arbitrary
set
of
data
and
then
create
a
beautiful
graphs
like
I,
don't
know
version
by
dimension
or
something
like
that.
I
don't
know
like
the
specific.
It's,
like
my
house
like
how
it
wouldn't
be
working.
I
know
it
doesn't
look
like
we
push
that
into
periscope
is.
B
D
D
D
Because,
like
I'm,
actually
thinking
that
this
could
be
our
okay,
our
like
increase
gathering
data
I
mean
key
result,
gather
data
about
architecture
or,
like
my
more
usage
of
our
on-premise
customer
into
into
periscope,
because,
like
there's
gonna,
give
us,
maybe
some
hints
about
how
people
different
Iran.
What
are
the
conflicts
and
maybe
cannot
give
us
markings
about
the
direction
in
which
we
should
optimize?
D
B
E
So
I
can
answer
a
bit
there.
We
did
we're
almost
back
usage
thing
was
not
part
of
that
usage
thing.
It's
been
the
product
for
a
long
time.
People
can
turn
it
off
if
they
want
to
and
we're
working
to
improve
it.
So
our
larger
customers
don't
have
a
problem
with
it,
but
what
they
do.
The
problem
with
it
is
that
they
need
to
have
their
compliance
team
review.
All
the
data
attributes
that
go
out
so
they're
like
look
I,
don't
have
a
problem.
I
mean
collect
a
chance
to
review
it,
and
so
like
one.
E
My
problems
are
something
right
now.
Is
that
we're
trying
to
hope
in
the
future
rather
I,
don't
think
it's
being
tackled
right
now
is
that
we
need
to
have
a
way
to
opt
in
to
the
new
data
collection
attributes
right.
So
if
we
add
memory,
they
have
been
need
a
way
to
like
look
at
it
and
say:
okay
now
get
live,
wants
to
start
collecting
memory,
stats,
I'm,
okay
with
that
and
the
compliance
you
can
sign
off
and
not
just
the
current
model
of
you
know
you
upgrade
and
all
of
a
sudden,
a
whole
bunch.
E
More
data
starts
leaving
your
gillip
instance
and
and
so
that
somebody
don't
trust
us
is
that
it's
that
they
need
to
have
their
compliance
to
look
at
these
things
and
make
sure
they're
okay
with
it
before,
for
the
data
starts
to
leave,
and
so
we
took
the
folks
and
he's
just
paying
again
cuz.
It's
not
considered
a
problem
for
many
and
the
main
blocker
to
having
it
more
time
are
widely
his
features
like
this.
They
have
so
much
of
control
over
it,
working
without
being
just
purely
on
or
off
so.
E
E
If
we
can
get
this
relatively
easily,
that
that
would
be
amazing,
so
I'm
not
sure
how
much
we
can
gather
and,
for
example,
like
there
at
the
Ruby
code,
which
is
where
the
users
ping
currently
runs,
but
if
we
can
gather
it
there,
that
should
make
it
relatively
easy
to
pick
up
without
having
to
go
to
Prometheus
or
something
like
that,
which
would
be
a
lot
of
a
larger
lift.
Because,
right
now
we
wait
to
figure
out
whether
to
meet.
D
Because,
like
Prometheus
would
be,
if
you
would
have
from
it
is,
it
would
be
super
straightforward,
like
to
run
different
queries
against
these
primitives,
but
maybe
like
we
really
advise
people
to
run
from
meteors.
You
just
do
we
know
exactly
how
many
of
our
accusers
come
from.
It
is
to
monitor
it
up.
E
I
think
a
lot
of
them
do.
Our
support
team
has
been
recommending
it
to
get
it
running.
The
question
I
have
is
I'm,
not
sure
how
many
people
run
the
embedded,
Prometheus
server,
that
we
have
and
I'm
not
sure
the
good
level
vacation.
The
Gila
application
always
knows
where
it
is
so
I
think
a
lot
of
folks,
probably
almost
nearly
all
of
our
customers
hit
the
Prometheus.
E
E
D
Like
one
is
one
one
complex
aspect,
we
gonna
be
in
Ruby
figuring,
combat
topology,
which
is
like
pretty
challenging
to
do
it
and
second
task
we
need
are
getting
compliance
that
we
can
actually
gather
this
later
and
get
maybe
to
some
extent
user
approval.
If
we
consider
this
data
to
be
sensitive
to
users
or
not,
which
is
more
like
the
like
a
product
side
of
the
story,
right,
Josh,
I'm,.
E
Not
worried
about
I'm
not
worried
about
how
sensitive
it
is.
We
gather
data
like
this
all
the
time
and
we
don't
tell
anyone
basically,
so
what
we
can
do
is
we
can
go
to
the
telemetry
team.
There's
a
approval
process
for
for
adding
new
things
usage
thing
and
they
can
they'll
take
a
look
at
it
and
in
general,
they'll
tell
us
if
it's
a
good
idea
or
not
to
go
collect
but
in
general
I
think
getting
things
like
memory
usage
would
not
be
controversial
at
all,
but
that
would
be
fine.
A
So
I
can
create
an
update.
The
okay
are
to
reflect
this.
This
path,
which
is
great
thanks
for
your
suggestion.
You
know
no
need
to
apologize.
That's
actually
a
good
idea.
I
can
create
the
epic
and
the
issues
they'll
be
fairly
stubbed
out,
but
if
there's
anybody
else
that
wants
to
tackle
it,
I
will
certainly
abdicate
that
responsibility.
A
Right
we
had
a
planning
meeting
scheduled
for
tomorrow.
I
might
end
up
canceling
that
so
I
can
make
some
time
to
chat
with
everybody
about
the
telemetry
meeting
and
quite
honestly,
I'm
not
prepared
for
a
planning
meeting
tomorrow.
I
don't
have
a
buncha
tissues
to
break
down.
So
then
we
probably
just
reschedule
out
for
next
week.