►
From YouTube: Next Great Data Developer: Lyuben Todorov — CasTor
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
A
Castor
was
created
because
a
new
law
that
might
be
passed
in
the
UK,
the
new
legislation
would
force
Internet
service
providers
to
monitor
all
of
the
traffic
generated
by
its
users,
with
a
heightened
interest
into
traffic
involved
with
communication
programs,
including
emails
and
messaging
programs.
Yes,
there
is
a
proof
of
concept
that
the
communication
channel
can
be
secured
from
your
isp
by
using
Cassandra
and
tour.
A
The
anonymity
network
Cassandra
was
used
in
castor
for
variety
of
reasons,
the
two
most
important
reasons
being
the
Cassandra
sport
usage
couldn't
be
linked
to
emails
in
any
way,
and
the
Cassandra
allows
you
to
build
a
highly
distributed
system.
If
a
company
ran
a
central
email
servers
and
clients
would
be
relying
on
that
company
maintaining
server.
If,
however,
the
email
service
was
based
on
each
client's
machine,
then
clients
would
only
be
dependent
on
people
actually
using
the
system.
A
A
Because
it's
the
first
one
of
the
application,
there
are
some
terms
of
service
and
some
instructions.
What
you
need
to
do
to
get
the
note
started
as
you
can
see
the
application
created
a
new
cassandra
folder
on
the
desktop.
This
will
be
our
server
that
will
be
connecting
as
a
new
note
or
cluster
once
you
click
ok
here
the
new
node
will
be
bootstrapped
into
the
cluster
and
it
will
be
ready
for
use.
A
A
This
one
is
now
part
of
the
node
part
of
the
cluster
and
it's
ready
for
use
by
our
application.
If
we
rerun
the
client,
because
it
won't
be
on
the
first
one
anymore,
it'll
connect
through
tor
and
display
the
interface
that
we
use
for
doing
messages.
Now
we
can
simply
log
in
you
can
pick
any
user
name
and
you'll
be
ready
to
get
started.
You're
displayed
with
a
really
simple
interface
to
create
new
messages.
A
Cassandra
was
the
only
database
used
for
their
application.
In
this
slide,
I'll
provide
an
overview
of
all.
The
data
was
modeled
in
the
cast
or
application
since
I
wanted
to
avoid
simulating
joins
in
the
client-side
application.
The
data
ended
up
being
fairly
unnormalized
to
key
spaces
were
used
with
the
column
family.
A
This
allows
the
system
to
be
distributed
across
users
rather
than
depending
on
a
central
server.
Each
time
cast
our
bootstraps
a
new
users,
local
cassandra
server
to
the
cluster.
The
replication
factor
of
the
message
key
space
is
incremented,
meaning
that
each
node
still
contains
all
the
data
stored
relating
to
messages.
This
allows
the
system
to
be
running
without
any
problems,
even
when
users
switch
their
machines
off.
A
How
was
my
experience
of
using
cassandra
for
my
project?
I
would
say
a
really
pleasant
experience
and
I
do
think.
Cassandra
is
the
way
to
go.
If
you
need
a
distributed
system,
I'm
used
to
spending
lots
of
time
trying
to
work
out
how
to
get
anything
started
with,
especially
with
servers,
but
I
had
Cassandra
running
within
minutes,
the
very
first
time
I
downloaded
it
and
used
it.
A
So
it
was
very
easy
to
set
up
I've
had
bad
experiences
in
the
past
when
it
comes
to
clustering
databases,
but
Sandra
made
it
very,
very
simple
to
get
stuff
done
really
quickly,
and
it's
really
easy
to
expand
clusters
and
build
very,
very
big
clusters
with
a
lot
of
machines
in
them.
I
found
client-side
SSL
to
be
a
bit
difficult
to
set
up.
The
other
problem.
A
I
had
was
the
authentication
key
space
are
used
because
I
couldn't
afford
to
distribute
users
passwords
to
everybody,
even
if
they
are
hashed
and
encrypted
it
wasn't
safe
and
because
it
didn't
have
control
over
writing
the
only
one
node
I
ended
up
using
a
separate
server
for
that.
The
final
problem
I
had
with
Cassandra
was
a
bootstrapping
between
different
operating
systems,
nine
out
of
ten
times
what,
but
occasionally
would
crash,
and
it
wouldn't
be
the
same
mirror
every
time,
so
I
decided
to
just
branch
the
applications
at
the
different
systems.