►
Description
Don’t miss out! Join us at our next event: KubeCon + CloudNativeCon Europe 2022 in Valencia, Spain from May 17-20. 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.
解决服务网格应用者的困境 | Solving the Service Mesh Adopter’s Dilemma - Anita Ihuman, Layer5
我该使用哪种服务网格，如何开始？有哪些不同的服务网格，它们的对比情况如何？了解不同服务网格的功能性，并直观地操作网格配置。本次闪电式演讲介绍了一种开源多服务网格管理平面，Meshery，其提供（五种并且未来还会持续增长）不同的服务网格，还介绍了其示例应用程序，并对服务网格部署的性能进行基准测试。Meshery 促使对 Istio 的各种配置场景进行基准测试，比较网格内外以及不同网格之间的服务（应用程序）的性能。其根据部署最佳实践检查网格和服务配置。部分服务网格项目使用 Meshery 作为每个版本的性能基准工具。
Which service mesh should I use and how do I get started? What are the different service meshes, and how do they contrast? Learn about the functionality of different service meshes and visually manipulate mesh configuration. This lightning talk introduces Meshery, an open source, multi-service mesh management plane that provisions (five and counting) different service meshes, their sample applications and benchmarks the performance of service mesh deployments. Meshery facilitates benchmarking various configuration scenarios of Istio, comparison of performance of services (applications) on and off the mesh and across different meshes. It vets mesh and services configuration against deployment best practices. Some of the service mesh projects use Meshery as their performance benchmark tool for each release.
A
A
Let's
get
to
the
topic
of
today
right.
What
establishment
is
the
first
question.
Every
single
adulterer
will
want
to
ask,
but
savage
mentions
in
the
layman's
terms,
are
simply
a
way
to
control
how
different
parts
of
an
application
or
micro
services
communicate
or
share
data
with
one
another
they're
otherwise
considered
as
a
micro
service
platform,
which
is
true.
A
A
It
is
very
possible
to
put
the
logic
down
and
governs
the
communication
between
each
of
these
services
without
the
use
of
establishment
layer.
But
in
a
situation
where
the
communication
starts
becoming
more
complex
over
over
the
day
or
as
time
goes
by,
then
the
need
for
the
surface
mesh
starts
becoming
more
valuable.
A
People
adopt
service
mentions
for
countless
number
of
reasons,
but
it's
often
as
a
result
of
the
use
cases
they're
looking
to
benefit
from
and
some
of
the
capabilities
that
people
frequently
look
out
for
in
a
service
meshes
or
people
deploy
establishing
matches
for
are
one
the
observability
using
services
you
can
using
service
meshes.
You
can
generate
all
kinds
of
traces,
logs
and
metrics.
You
can
also
ingest
this
into
your
monitoring
system
of
choice
to
get
a
value
without
insur
instrumenting
your
application.
A
This
is
one
of
the
very,
very
important
capabilities
that
adopters
prepare
and
look
out
for
because
you
can
handle
the
consistent
metrics
across
the
fleet.
You
can
trace
the
flow
of
requests
across
levels
across
the
services
and
you
can
also
deal
with
the
metrics
without
instrumenting
your
applications.
Another
important
capability
is
security,
of
course,
attention.
A
Of
course,
you
will
be
able
to
do
so
much
with
them.
So
much
of
chaos.
Engineering.
When
you
use
of
selfish
messages,
you
can
configure
the
mesh
to
provide
a
bunch
of
resonance
to
your
tablet
and
you
can
even
add
retries
for
your
field
request
and
with
the
use
of
establishment
sheets.
You
can
handle
the
time
out
the
systemic,
systematic
fault
injections.
You
can
control
the
connection
full
size
as
well
as
the
request
loads.
A
A
A
To
help
with
modernization
with
the
use
of
the
service
measures,
you
can
modernize
your
id
inventory.
Without
writing
your
application.
You
can
adopt
certain
micro
services
level
and
also
use
your
regular
services,
and
that
is
perfectly
fine.
Like
I
said
earlier,
you
can
actually
use
this
mic.
Service
meshes
returns
running
your
application
on
the
microservices,
and
you
can
also
adopt
new
frameworks
that
are
arising,
while
you
view
the
application
and
finally,
you
also
get
to
move
to
the
cloud.
A
Of
course,
the
introduction
of
the
plan
has
made
the
and
made
the
development
process
a
lot
easier
and
mid
speed
and
aspect
of
everything
when
it
comes
to
the
production.
So
all
of
this
saves
the
developer
so
much
time
and
improves
the
developer
speed,
which
is
another
very
important
reason
why
people
want
or
why
you
want
to
adopt
establishments.
A
This
is
where
all
of
the
service
proxies
are
logically
grouped,
and
they
are
responsible
for
a
lot
of
purposes
such
as
executing
of
traffic
control,
health
checking,
routing
load,
balancing
auto
scaling,
auto
authorization,
observability
and
so
on.
Now,
moving
on
to
the
second
layer,
which
is
the
control
plane,
this
is
where
an
operation
interface
with
a
service
mesh.
It
deals
with
speaking
of
speaking
to
the
proxies
and
updating
the
configuration
for
a
given
service
match.
A
So
in
the
control
plane,
you
should
be
able
to
see
that
your
establishment
should
provide
policies,
configuration
and
as
well
as
platform
integration.
It
takes
a
set
of
isolated,
left
side
class
proxies
and
turns
them
into
a
service
mesh
and
also
it
does
not
touch
any
packets
requests
into
the
data
pack
and
finally,
the
last
layer,
which
is
called
the
management
thing.
This
is
the
layer
at
the
top
which
helps
to
federate
the
service
mesh
deployments.
A
This
the
management
plane
helps
to
perform
a
lot
of
things
that
you
would
most
likely
not
expect
from
the
control
plane,
and
these
include
providing
the
federation,
the
backend
system,
integration,
expanding
policies
and
governance,
continuous
delivery,
integration,
workflow
chaos,
engineering
application
performance,
and
so
many
more
and
mostly
sent
to
this
we're
going
to
look
at
the
service.
Mesh
management
thing
called
measuring,
which
enables
operators,
developers
and
services
owners
to
realize
the
full
potential
of
the
federal
sms.
A
Most
of
these
projects,
like
the
service,
mesh
interface
and
sectional
performance,
are
also
match
a
project
that
satisfies
the
mystery
project
in
the
media
five
community.
Now
you
can
see
a
quick
view
of
what
this
service
mesh
performance
looks
like.
If
you
want
to
know
more
about
the
search
performance,
I'll
suggest
you
check
out
the
links
provided
on
this
slide,
which
I'll
provide
the
link
too
much
later,
and
you
will
understand
all
about
service
mesh
performers.
A
All
right
now
moving
forward
we're
going
to
be
looking
at
the
adopters
dilemma
as
an
adopter
or
as
someone
who
is
new
to
service
machines,
you
must
have
tons
of
questions
which,
of
course
I
did
when
I
got
introduced
or
when
I
I
knew
they
heard
of
savage
nurses,
and
I
always
asked
which
service
mesh
is
actually
the
best
for
an
organization,
because
since
they're,
like
numerous
service
questions
out
there,
how
does
one
get
started
with
the
services?
And
then
what
is
the
cash
using
the
publisher
and
all
of
those
questions?
A
And
there
are
over
20
different
service
meshes
and
within
the
layer
5
project
within
the
f5
community.
We
have
a
project
that
creates
the
landscape
of
these
different
service
measures
and
their
related
technologies,
and
so
on.
This
landscape
project
you're,
going
to
see
that
it
allows
you
to
compare
which
service
mesh
is
a
better
option
for
you
or
organization
and
which
has
more
capabilities
to
help
you
in
one
way
or
another.
Why?
A
A
It's
a
community
that
offers
cloud
native
management
softwares
that
exploits
the
unique
positions
service
merchants
have
in
changing
how
developers
write
applications,
how
operators
run
modern
infrastructures
and
how
service
mesh
owners
are
able
to
manage
their
own
sas
operation
and
then
with
tools
from
cloud
native
infrastructure
and
the
application
layer.
5
has
empowered
a
lot
of
developers,
operators
and
service
members
owners,
layer,
five
communities
towards
three
cloud
native
foundation,
projects,
chairs
network
and
service
mesh
groups
within
the
cncf,
and
it's
one
of
them,
one
of
my
favorite
open
source
community.
A
Of
course,
one
of
the
popular
projects
in
layer,
5
community,
is
the
measuring
which
is
considered
the
service
mesh
management
plane
measuring
is
the
largest
open
source
project
that
is,
between
the
near
five
community
and
within
the
immersion
project.
There
are
a
few
projects
that
also
satellites
this
project,
and
some
of
them
are
extensions,
whereas
others
are
actually
stand
alone
project.
Some
of
these
projects
are
already
donated
to
the
cncf.
As
you
can
see,
or
like
I
mentioned
earlier,
we
have
the
service
mesh
patterns,
servicemen
performance
and
service
interface
and
so
on.
A
Measuring
has
participated
in
quite
a
number
of
internship
programs
and
some
of
which
are
the
season
google
season
of
doc,
the
google
summer
of
code,
the
lineus
foundation,
mentorship
program.
We
call
infrastructure
initiative
and
as
well,
it's
a
project
under
the
cncf
which
has
sub-satellites
projects
such
as
the
service,
mesh
performance
and
establishment
interface.
A
Measure
is
simply
the
multi-mesh
management
platform
that
handles
the
life
cycles,
the
workloads,
the
performance,
the
configurations,
the
patterns,
practices,
skills
and
futures
within
your
application,
and
it
has
the
product.
This
project
already
has
quite
a
number
of
doctors
and
supporters,
most
of
which
are
looking
to
incorporate
measuring
into
their
release
process
in
order
to
measure
the
adherence
to
service
mesh
standards.
A
This
is
a
picture
description
of
what
the
messagery
architecture
looks
like.
You
can
see
that
it
actually
does
so
much
for
your
application
and
handles
all
of
the
service
mesh
management
that
goes
on
from
the
left
circle
down
to
the
workloads
and
the
configurations
and
deals
with
the
patterns,
and
so
many
others.
So
all
of
this,
our
goals
are
within
the
mystery
project.
A
If
you
want
to
join
the
mystery
community,
you
are
so
much
welcome
to
actually
jump
on
this.
It
is
a
warm
welcoming
community
and
this
project's
immersion
project,
along
with
all
its
sub
projects,
are
built
within
the
five
community
and
the
layer.
5
community
has
already
engineered
past
engineers
from
different
organizations
of
that
intel,
red
hat
rackspace,
hashgraph
maintenance,
and
so
many
others
from
different
open
source
organizations
and
their
five
communities,
particularly
measuring
runs
as
number
one
most
popular
liners
foundation
mentorship
project
lay.
A
We
have
a
project
called
the
measurement
program
which
helps
which
highlights
members
of
the
organization
of
the
community
to
help
mentor
and
onboard
others
that
are
coming
into
the
project
and,
of
course,
we
are
happy
highlighted
as
the
number
one
most
popular
project
on
the
linux
foundation.
Mentorship
program,
there's
actually
so
much
going
on
within
this
community.