►
From YouTube: PEEPanEIP#75: Nimbus & TheMerge with Zahary
Description
Topics covered
-----------------------------
00:44 - Client introduction
1:08 - Guest’s introduction
2:51 - Presentation
3:05 - How Nimbus fits into the mission of Status
5:05 - What makes Nimbus unique?
6:20 - DIY enthusiasts and supported by web3 pools
7:21 - Independent studies (Better ROI from lower resource costs)
7:45 - Effectiveness that you can count on
10:13 - Audit - Built to minimize external dependencies.
11:05 - Nimbus team
11:22 - Q&A
11:41 - What is the status of Nimbus execution client? Is it available to the public?
13:46 - What is there to improve Nimbus user experience after the merge?
15:23 - What are the changes to beacon chain client post merge?
16:43 - Testnet experience and interop performance
21:25 - Favorite client combination
23:15 - Client diversity
23:42 - Best EL CL client combination
25:32 - Specific testnet incidence that changed the path of progress
25:53 - Ropsten testnet bug
27:12 - Sepolia testnet readiness
28:42 - Goerli & Prater testnet merge
30:07 - The journey for the Merge
31:20 - Panda ASCII art
32:25 - What is a client dev’s expectation from the community?
Resources:
-----------------
Slides: https://docs.google.com/presentation/d/1Jjp06-modPyS2KIulung35A1ce6EcSFJ/edit#slide=id.g11290a97025_0_110https://docs.google.com/presentation/d/1kPyUoo7cSYZbZHRKGJ7GYXMxPS5WrLsY/edit#slide=id.p6https://docs.google.com/presentation/d/1kPyUoo7cSYZbZHRKGJ7GYXMxPS5WrLsY/edit#slide=id.p6
Portal Network - https://youtu.be/YzL2GmxA6Wk
Shadow Forking - https://youtu.be/3AkDPL0RWtk
KnowYourClient - https://www.youtube.com/playlist?list=PL4cwHXAawZxoruie1hmYWJfiakpYivXft
PEEPanEIP playlist - https://www.youtube.com/playlist?list=PL4cwHXAawZxqu0PKKyMzG_3BJV_xZTi1F
TheMerge playlist - https://www.youtube.com/playlist?list=PL4cwHXAawZxqoLxXqZqT4hcYhoHoP6w12
Check out upcoming EIPs in Peep an EIP series at https://github.com/ethereum-cat-herders/PM/projects/2
Follow at Twitter
-----------------------------------------------
Nimbus @ethnimbus | Pooja Ranjan @poojaranjan19
Contact Ethereum Cat Herders
---------------------------------------------------
Discord: https://discord.io/EthCatHerders
Twitter: https://twitter.com/EthCatHerders
Medium: https://medium.com/ethereum-cat-herders
GitHub: https://github.com/ethereum-cat-herders
Website: https://www.ethereumcatherders.com/
A
The
ethereum
community
is
moving
towards
the
merge,
making
it
important
for
all
validators
to
consider
running
both
execution,
client
and
consensus.
Client
node,
which
is
the
best
client
combination
suiting
your
needs,
is
a
question
that
we
would
like
to
help
you
in
this
know.
Your
client
series
welcome
to
p
paneer
episode.
75,
I'm
your
host,
puchar,
ranjan
and
with
me
today
is
zahri
an
ethereum
developer,
working
on
client
members.
Thanks
for
joining
us,
zahri.
A
A
brief
introduction
about
the
client
as
per
the
official
website
nimbus,
is
an
ethereum
client
implementation
that
strives
to
be
as
lightweight
as
possible
in
terms
of
resource
used,
we'll
get
to
learn
more
about
nimbus
in
this
episode
with
zari.
But
first
may
we
have
a
brief
introduction
of
the
guest
and
his
way
to
the
ethereum
community.
B
All
right,
so
I've
been
involved
with
ethereum
for
the
past
four
years.
I
had
to
start
from
almost
zero
knowledge
when
I
joined
the
nimbus
team,
but
I've
been
on
quite
the
right
to
learn
quite
deeply
about
everything
about
ethereum.
Before
joining
the
nimbus
team,
I
was
one
of
the
contributors
of
the
nem
programming
language
and
actually
that's
how
I
ended
up
on
the
team
when
jared
hope,
the
founder
of
status
announced
that
status
will
build,
building
an
ethereum
client
in
him.
That
was
not
only.
B
That
was
probably
one
of
the
first
high
profile
project
in
the
new
programming
language.
So
for
me
it
was
quite
exciting
to
join
the
team.
I
was
actually
the
first
hire
and
helped
the
company
recruit
some
of
the
best
kind
of
more
more
act,
most
active
developers
from
the
nim
community
and
I'm
really
proud
of
the
team
that
we
have
gathered
to
work
on
the
project.
A
Awesome
this
is
really
nice
to
know
that
you
were
one
of
the
first
hire
for
the
team.
Well,
thank
you
so
much
for
the
introduction.
It
is
always
nice
to
know
people
who
have
been
working
so
hard
to
keep
running
etm.
Well,
I
suppose
we
have
got
a
lot
to
cover
on
this
call.
So
without
further
ado,
let's
peep
in.
B
So
welcome
to
this
brief
nimbus
introduction.
I
have
a
short
presentation
here
that
I'll
focus
only
on
the
most
important
sites
in
the
interest
of
keeping
this
podcast
open
for
questions
later.
So
let
me
start
by
introducing
nimbus.
I
think.
First
of
all,
we
need
to
understand
how
nimbus
fits
into
the
wider
mission
of
status.
B
We
develop
the
nimbus
gland
with
the
focus
on
efficiency,
trying
to
minimize
the
resources
that
it
consumes
and
making
sure
that
this
vision
for
very
light
light
clients
would
be
possible
in
the
future
right
now.
Nimbus
is
a
fully
featured
beacon
out
in
validator
plant,
but
I
envision
a
future
where
millions
of
devices
will
be
using
our
software
components
to
process
and
accept
ethereum
payments
in
mobile
environments.
This
could
be
your
local
merchant,
as
I
mentioned
you
could
buy.
B
B
Basically,
this
is
another
way
to
get
a
network
of
much
more,
lighter
nodes
that
doesn't
suffer
from
the
problem
of
clients
requiring
the
availability
of
all
of
server
nodes.
Basically,
the
portal
network
is
consist
of
homogeneous
nodes
which
act
both
as
client
servers.
At
the
same
time,
all
right,
something
that
we
are
quite
excited
is
just
how
far
we
were
able
to
push
this
efficiency.
B
For
example,
there
are
many
users
right
now
in
the
rocket
pool
community
which
run
nimbus
and
an
execution
cloud
like
get
on
a
single
raspberry,
pi
device
device.
So
you
know
raspberry
pi
4,
with
its
8
gigabytes
of
memory
are
totally
enough
to
run
this
combination
of
software.
You
have
to
attach
an
external
hard
drive,
an
ssd
hard
drive
to
make
it
happen,
but
the
whole
setup
is
still
pretty
cheap,
runs
entirely
fanless
and
consumes
very
little
energy.
B
You
know
whatever
client
you
install,
you
can
be
certain
that
you
get
getting
the
maximum
validated
rewards
possible
pretty
much.
Basically,
the
clan.
The
different
clans
differ
very
very
slightly
in
terms
of
the
rewards
you're
getting
here.
You
are
seeing
the
results
of
one
study
performed
by
the
consensus
team.
The
authors
of
deco,
which
demonstrated
tekku,
has
a
very
slight
lead
over
nimbo
switches
in
the
second
place,
but
I
should
point
out
that
the
nemesis
results
here
are
from
a
very
large
fleet
of
diverse
nimbus
installations.
B
The
entire
list
of
nimbus
dependencies.
The
third
party
dependency
that
our
project
is
relying
on
is
visible
on
the
screen
here.
This
means
that
we
have
tight
control
over
the
develop
software.
We
have
the
freedom
to
optimize
each
and
every
component
to
make
sure
that
together
they
deliver
this
efficiency
that
I
talked
about
earlier.
B
A
A
B
And
the
nemo's
execution
client
is
still
in
development.
It
is
able
to
currently
pass
the
entire
suit
of
hive
tests,
so
you
can
argue
that
it
is
feature
complete.
Although
right
now
it's
not
optimized
enough
to
be
proactive,
go
for
running
on
mainnet,
basically,
the
components
that
we
haven't
focused
on
yet
is
the
storage
engine.
B
We
are
still
running
with
a
storage
engine
that
we
developed
the
kind
of
very
first
steps
in
the
projects
four
years
ago,
which
is
correct,
but
very
much
not
optimal.
So
running
a
main
eighth
note.
We
can
consume
numer
several
terabytes
of
data
right
now,
so
for
all
practical
purposes,
we
don't
recommend
trying
out
the
execution
cloud
layer.
Yet
this
is
a
focus
of
our
development
right
now
and
we
hope
to
have
perhaps
a
better
release
by
the
time
of
the
march.
A
B
A
B
Yes,
we
really
hope
to
deliver
a
very
simplified
and
streamlined
installation
process
and
we
are
we'll
be
building
an
optional
integration,
with
the
status
app
so
basically
you'll
be
able
to
install
nimbus
on
a
server
on
your
home,
computer
or
home
network
and
pair
your
mobile
device,
where
the
status
of
the
install
and
this
will
provide
a
management
interface
and
a
monitoring
dashboard
for
your
validators
and
that
you
can
access
on
the
go.
Whenever
you
are.
A
A
B
Obviously,
we
want
to
achieve
a
mode
of
operation
which
has
a
kind
of
steady
state
storage
requirements
by
pruning
the
data
once
it
becomes
unnecessary
for
the
network
after
the
weeks
of,
for
example,
past
the
weeks
of
which
the
subjectivity
horizon-
and
we
are
always
working
on
various
functionalities
and
features
which
a
name
are
intended
for
kind
of
power.
Users
such
as
taking
posts
and
so
on.
A
A
A
B
Well,
I
think,
as
client
teams,
we
have
probably
learned
a
lot
during
the
development
of
both
ethereum
one
and
the
development
of
the
beacon
chain.
So
we
did
go
through
a
very
gradual
process
of
rising
the
bar
of
what
we
hope
to
achieve.
Basically
when
we
get
got
together
during
the
interrupt
event
in
greece.
B
And
then,
when
we
were
certain
that
this
works,
fine
and
all
the
clients
can
communicate
to
each
other
without
problems
which
write
more
significant
challenges
such
as
organizing
a
test
net
where
all
the
nodes
would
join
and
produce
logs
and
so
on,
and
during
the
interop.
Even
during
the
interop
event,
we
had
a
network
able
to
produce
blocks.
B
So
over
the
next
weeks
we
continues
to
rising
the
bar
by
recreating
the
same
conditions
in
a
little
bit
more
long,
cliff
networks
in
a
fashion
where
the
coordination
was
happening
fully
online
online.
So
it's
a
bit
more
difficult
to
travel,
shoot
problems,
but
I
think
all
we
know
the
process
has
been
quite
smooth.
Most
of
the
time
the
clients
were
working
well
together
and
where
issues
were
detected,
they
were
relatively
quickly
addressed.
B
I
should
make
a
huge
shout
out
to
pari
from
the
ethereum
foundation,
who
is
managing
this
testnet
in
very
efficient
way,
he's
making
sure
that
each
and
every
client
combination
is
well
represented
and
we,
the
client
teams,
get
a
lot
of
useful
data
out
of
the
nodes
that
he's
running
and
now
the
combination
of
basically,
for
some
time
we
were
launching
this
new
custom
network
that
starts
from
scratch
without
a
lot
of
history
right.
So
these
are
much
smaller
blockchains
that
may
net
or
even
any
of
the
test
nets
the
long
cliff
test
nets.
B
So
we
gradually
continued
raising
the
bar
by
creating
something
called
the
shadow
fork,
which
is
you
can
think
of
about
it
as
a
network
which
shares
all
the
history
of
mainnet,
but
at
some
point
in
time,
diverges
in
a
different
direction.
So
we
exit
pretend
to
execute
the
merge
on
a
network
which
share
the
history
with
mainnet
but
diverges
in
this
different
direction,
where
the
merge
has
already
happened.
B
A
Right,
I
would
agree,
I
mean
paris
has
done
a
great
job
here,
bringing
all
client
themes
together
and
continuously
sharing
data
with
the
client
combination.
Those
are
working,
fine
wondering
if
there
is
any
favorite
client
combination
of
yours-
or
maybe
you
think,
like
numbers-
can
be
best
paired
with
any
particular
execution.
Client.
B
Well,
I
would
encourage
people
to
run
minority
class.
This
is
important
for
the
network
because,
for
example,
imagine
that
some
kind
of
issue
is
discovered
in
a
majority
plant.
There
is
even
the
risk
that
the
network
might
reach
a
consensus
where
this
issue
has
affected
the
outcome
of
the
blockchain.
So
if
we
have
a
majority
client
that
controls
over
50
of
the
network,
this
is
a
significant
risk.
B
But
if
everybody
is
running
a
different
client,
then
any
client
having
issues
will
just
experience
a
temporary
voltage,
which
would
be
easy
to
kind
of
resolve,
I
would
say,
and
the
network
as
a
whole,
it
will
be
in
a
much.
We
will
have
much
lower
risk
systemic
for
this
kind
of
problem
and
obviously
it's
also
good
for
mitigating
the
potential
for
security
threats,
vulnerabilities
discovered
in
the
clients
and
so
on.
So
I
would
recommend
for
anyone
to
kind
of
try
to
gouge
what
is
the
current
cloud
distribution
and
then
go
for
the
minority
client.
A
That
is
right.
Client
diversity
is
like
one
of
the
hottest
topic
of
discussion
these
days.
So
I
wonder
like
what
are
your
thoughts
you
just
shared
about,
like
you
would
recommend
going
for
minority
clients?
The
other
day
we
were
talking
to
bizu
bazu
is
also
considered
to
be
one
of
the
minority
client.
However,
I
think
that
these
are
all
colors.
Those
are
adding
to
ethereum
networks,
so
more
users,
we
can
add
to
all
these
individual
client
providers
would
be
good
too
good
for
the
new
joiners,
as
well
as
the
ethereum
ecosystem.
B
A
B
A
B
We
had
a
little
bit
of
rough
start
indeed,
because
the
rockstar
network
was
a
little
bit
of
a
first
time
for
us.
For
basically
like
any
other
organization,
we
have
a
devops
department
which
has
prepared
some
scripts
for
the
deploying
ethereum
nodes,
and
basically
we
have
this
brand
new
code.
That
was
never
tried
before
and
I
was
supposed
to
run
for
the
first
time
for
the
robston
lounge
and
it
did
had
suffice
to
say
it
had
some
small
issues.
A
B
So
it's
very
important
for
us
to
create
a
diverse
set
of
basically
configurations
that
go
through
the
merge.
Maybe
you
should
try
a
client
which
is
not
synced,
for
example
during
the
march.
Maybe
you
want
your
plans
slightly
later,
maybe
your
weekend
on
this
thing,
but
your
execution
node
is
not
synced
stuff,
like
that.
We
are
supposed
to
try
all
these
different
scenarios
and
make
sure
that
they
work
well
and
the
test
nets
offer
us
the
ability
to
kind
of
try
this
few
times
and
gain
more
confidence
that
the
corners
cases
have
been
handled.
A
That's
right,
that's
exactly
the
purpose
of
test
nets,
and
that
is
one
of
the
reasons
why
we
are
having
so
many
test
nets
like
being
tested
before
we
go
for
the
main
net
match
we're
talking
about
the
future
test
net.
I
understand.
After
sepolia's
test
net
merge
there
would
be
godly
and
prata
testnet,
so
prater,
I'm
assuming
it's
a
beacon
chain
test
net,
so
you
may
have
I
mean
the
team
may
have
been
familiar
with
the
proto.
I
wonder
what
is
the
general
experience
working
with
the
product
as
nut.
B
Well,
all
the
clan
teams
have
been
running
the
majority
of
validators
on
the
broader
test
net.
It
is
a
long,
cliff
theorem
network
that
I
I'm
sure
many
of
your
listeners
are
familiar
with.
So
what
would
be
challenging?
I
guess
is
that
it
is
a
network
where
we
are
merging
both
a
beacon
chain
with
long
history
and
an
ethereum
one
network
with
long
history.
So
in
terms
of
size,
it's
not
really
approaching
mainnet,
but
it's
quite
a
bit
closer
to
this
scale
that
we
are,
after
with
the
main
main
transition.
A
B
B
One
of
the
primary
motivations
for
me
was
actually
addressing
the
climate
change
concerns
of
cryptocurrencies
right.
So
the
proof
for
stake
eliminates
this
enormous
waste
produced
by
the
kind
of
proof
of
work
algorithms.
So
the
sooner
we
deliver
the
merge,
obviously
the
better,
but
at
the
same
time
we
need
to
be
cautious.
A
B
A
A
B
Well,
obviously,
one
important
question
is:
how
do
we
ensure
the
sustainability
of
clan
development
and,
obviously,
right
now
we
have
different
player
players
in
the
ecosystem
who
support
our
work.
Basically,
our
work
is
important
both
for
status
and
the
ethereum
foundation,
so
they
provide
great
support
for
us
and
we're
truly
grateful,
but
I
think
in
the
long
term
we
will
need
a
plan
for
how
the
client
teams
will
continue
to
be
sustained
in
the
future.
A
A
I
really
hope
that
information
shared
today
in
this
episode
will
be
useful
for
new
users.
New
validators,
as
well
as
the
existing
validator
may
be
considering
running
alternate
nodes
if
they
are
not
running
nimbus
or
any
other
minority
client
consider
running
one
of
those
nodes.
Thank
you
zari.
We
really
appreciate
you
taking
out
time
and
talking
to
us.