►
From YouTube: 2021-04-29 Kubernetes SIG Scalability Meeting
Description
Agenda and meeting notes - https://docs.google.com/document/d/1h...
A
All
right
all
right
welcome
everyone
to
our
sixth
grade
public
meeting.
A
We
don't
have
a
lot
of
agenda
for
today,
just
a
few
announcements
and
organizational
informations,
so
I
uploaded
as
promised,
like
the
last
four
meetings
to
youtube
from
the
period
I
was
away.
So
if
you
want
to
take
a
look
here
is
the
link
to
our
playlist.
A
As
I
said,
I
was
hoping
that
apple
will
be
here
today.
I
wanted
to
discuss
the
issue
we
touched
two
weeks
ago
about
the
performance
benchmarks
he
he
did
for
priority
and
fairness.
We
were
hoping-
and
he
was
positive
about
this
when
we
were
talking
and
that
he
would
share
them
with
us
and
basically
open
source
them,
so
we
can
start
running
them
continuously.
A
So,
let's
leave
it
in
the
next
item.
Until
above
is
here.
A
Three
more
things
I
wanted
to
share
with
you.
One
is
that
we
have
this
issue
with
pear
dash.
We
are
not
able
to
upgrade
it
beyond
2.22
version
and
because
the
next
version
consumes
more
memory
than
the
previous
version
anywhere
out
of
resources
in
our
a
cluster.
I
open
an
issue
for
this.
I
I
mark
as
hell
wanted.
I
hope
someone
will
be
able
to
take
a
look
and
help
us
with
that.
A
There
is
pr
actually
opened
by
magic,
who
is
here
today
to
change
the
aa
cluster,
to
use
to
use
the
an
app
right,
no
auto
provisioning,
which
would
solve
the
issues
we
have
and
actually
ship
the
discussion
that
we
have
with
the
aaa
and
management
team
to
talk
about
the
actual
resources
used
by
the
by
the
jobs
we
run
done
rather
than
talking
about
whether
it's
okay
to
add
bigger
notes
or
not,
but
yeah
anyway.
A
I
hope
that
they
will
get
unblocked
like
it
has
to
get
unlocked,
because
we
we
have
some
updates
to
partners
who
would
like
to
deploy,
for
example,
the
things
that
voitek
mentioned
two
weeks
ago
about
our
new
network
tests.
So
we
still
don't
have
this
data
in
purpose
because
we
weren't
able
to
upgrade
further.
B
A
Yeah,
like
that's
another
way,
to
actually
fix
that
so
to
find
the
change
that
increase
the
memory
usage.
I
even
I
believe
it's
my
change,
because
I
I
added
like
so
what
I
did
in
perth
dash
is
that
we
had
some
of
these
tests
that
were
neither
load,
neither
density
test
and
they
basically
weren't
displayed
in
perth,
because
paradise
has
this
very
complex
logic
for
matching
for,
like
looking
for
the
files
on
gcs,
so
long
story
short
it
they
did.
A
The
file
is
looking
for
has
to
have
this
prefix
or
suffix
with
the
load
or
density
in
them.
So
I
added
like
this
like
third
group
of
parsers
that
are
looking
for
like
any
test,
so
I
believe
this
might
be
the
culprit
because
they
are
not
looking
for
density,
although
they
are
just
looking
for
the
name
of
the
measurement.
A
So
it's
likely
like
that
now
for
each
test.
We
are
actually
like
displaying
sorry.
B
A
And
displaying
the
same
data
twice,
because
one
is
much
for
the
density
or
or
or
load
parser
and
the
other
is
parts
for
this
catch
all
parser.
So
that's
the
reason
we
can
roll
this
back.
That's
an
option,
but
I
think
it's
not
sustainable
long
long
term
that
we
need
to.
You
know,
fight
for
every,
like
gigabyte
or
something
yeah,
even.
A
Like
your
arguments
that
our
sky
bts
they
use
like
terabytes
of
memory
or
even
more
so
fighting
about
one
gigabyte,
imperf
dash,
it
actually
doesn't
make
sense.
If
you
really
want
to
save
memory,
then
we
should
optimize
our
scale
test
right.
Yeah.
B
A
A
A
A
A
A
The
work
is
striking
this
issue
and
one
more
that
I
had
an
ai
last
week
to
go
through
the
hell
bonded
issues
and
to
check
whether
there
are
any
available
and
also
like
basically
check
because
we
had
we
had
some
issues
where
they
were
assigned
to
someone,
but
like
nothing,
was
really
going
on
there.
So
I
I
I
went
through
like
few
issues
on
top
of
the
list
and
basically
pink
them.
A
D
Sorry,
hey
there's
one
small
thing
from
my
end,
so
I
just
shared
an
issue
link
with
you,
so
th.
This
recently
came
out
in
like
one
of
the
work
that
we're
doing
here.
So
what
we
saw
was
for
hcd
when
there
is
a
lot
of
heavy
read
requests
right.
There
is
like
memory
memory
allocation.
There
are
some
unnecessary
allocations
happening
around
so
yeah.
It's
it's
a
little
bit
funny,
but
you
can
see
what
the
change
was.
D
So
as
part
of
the
request
at
cd
was
logging
at
one
place,
the
response
size
and
previously
it
was
calling
range
response,
dot
size
function,
sorry,
proto,
dot,
size
function,
which
is,
I
think,
calling
creating
a
duplicate
of
that
object,
which
was
unnecessarily
doubling
the
memory,
and
this
this
was
actually
a
significant
impact
on
some
of
the
cluster.
So
okay
ciao
has
joined
the
call,
so
he
was
he
was
investigating
into
this
issue
and
he
improved
that
fix.
Ciao.
You
wanna
talk
about.
E
Yeah,
so
can
you
guys
hear
me
yes.
A
E
Yep,
so
basically,
I'm
like
founding
this
issue,
because
one
of
our
cloud
cluster,
the
scd,
is
out
of
memory,
so
market
and
technically
we
don't
want
like
database
to
crash.
E
So
we
did
some
like
customer
case
study
and
learned
that
customer
is
issuing
some
big
list
pause
costs
across
all
namespaces
without
pagination,
and
then
we
like
reproduce
it
in
our
dev
cluster
and
find
out
like
and
also
do
some
like
a
profiling
into
the
lcd
and
find
out,
there's
like
unnecessary
photo
buffer
like
to
just
to
like
compute
the
size
using
the
portal
size
to
show
like
the
whole
response.
E
So
we
think
it's
unnecessary
and
after
we
change
it
to
like
a
range
response,
size
which
doesn't
take
the
whole
like
memory
like
a
usage
like
the
allocation,
so
we
do
observe
a
big
memory
used
percent
jobs,
so
yeah.
So
that's
the
context.
D
So
I
think
this
is
after
the
fix,
like
the
the
memory
usage
went
down
by
almost
50
percent,
with
like
a
lot
of
range,
heavy
requests.
E
Yeah,
like
another
finding
is
actually
the
lcd
3.3
and
a
street
of
four,
but
in
the
release
branch
they
are
built
by
like
between
go
112
and
the
goal
15
so
like
starting
from
go
12.
They
are
using
some
like
a
linux,
mli
system
call
which
doesn't
like
it.
So
basically
it
doesn't
like
a
release
memory.
E
E
It's
not
like
subtracting
the
lazy
free
process
usage
which
in
gold
and
like
garbage
collector.
E
E
So
in
one
like
a
goal
116,
they
are
like
turn
off
this
feature,
and
then
we
use
previous
using
the
previous,
like
linux
system
call.
So
the
monitoring
can
correctly
calculate
the
memory
use
percent.
A
D
D
Yeah
yeah,
so
I
I
think
in
in
summary,
what
is
happening
is
so
even
though
gc
kicks
in
and
memory
is
freed,
it
it
still
shows
under
the
process.
Memory
like
the
memory
is
still
with
the
process,
and
even
though
it
is
free,
it
is
only
apparently
reclaimed
back
when
the
os
is
actually
under
memory
pressure.
So
when
the
os
actually
needs
some
more
memory
which
it
is
when
it
comes
and
takes
it,
and
then
we
see
a
dip,
so
just
because
memory
is
high,
it
doesn't
mean
that
it
is
actually
using
that
much.
D
So
basically,
this
go
debug
that
environment
variable
setting
that
will
change
this
behavior.
So.
A
Does
it
mean
that
if
I
understand
correctly
that,
given
this
this,
I
don't
want
to
call
it
back,
but
the
way
the
golang
works
with
with
not
releasing
the
memory
immediately
to
the
operating
system?
It
will
get
even
better
like
this
graph
or
is
my
understanding
wrong.
D
Yeah
yeah
yeah,
instead
of
being
at
a
flat
line,
even
though
gc
is
happening
it'll.
Basically,
it
should
be
like
drop
as
soon
as
it
does
gc.
So
that's
what
charles
finding
was.
A
A
I
see
cool
so.
A
Had
like
similar
issues
with
inside
kubernetes,
I
assume
like
same
story
right
when
we
were
serializing
some
problems.
C
Nothing
like
that.
I
can't
remember
exactly
it
was
like
long
time
ago,
but
but
yes,
I
remember
that,
like
the
size
of
the
protobuf
were
were
causing
us
problems.
That's.