Sheriff. DDoS detection system

1.

SHERIFF
DDoS detection system

2.

MISSION
Today it is difficult to imagine a
successful organization that does
not have a web sites. Moreover
most companies directly depend
on their performance. But not
every service guarantees
trouble-free operation when
faced with a DDoS attack. The
goal of the project in brief is to
preserve the autonomy of the
resource, by providing protection
and minimizing the threat from
DDoS attacks.

3.

NOT ON MY SHIFT

4.

ML Algorithm
DDoS Overview
DoS and DDoS, its classifications
and how to defend
Grafana monitoring
System of monitoring and
visualization of traffic
01
03
02
04
Self learning algorithm which
analyzes traffic
Alerts and notification
Customization alerts for
anomalies

5.

01
DDoS
overview

6.

DoS and DDoS
DOS attack an external attack on an information
system that leads to a denial of service. As a rule,
the goal is to prevent users from gaining access
to such a system, or to make it very difficult for
them to obtain such access
DDoS using not only single computer to fill the
complete transfer speed of the server, so a
disseminated assault from different machines is
more often than not used-a DDoS assault.
Tainted machines from which demands are sent
are too called zombies.

7.

Structure of DDoS
zombie / agent
handler
attacker
victim
handler
zombie / agent

8.

DDOS AND PREVENTION
All devices connected to the
Internet can potentially become
part of an attacker’s
infrastructure and be used in
DDoS attacks.
You should build DDoS
protection at all levels. You
can pass traffic through the
cleaning network, organize
site protection at the
transport and network levels

9.

02
ML
algorithm

10.

ML algorithm
A major theme of forecasting at scale is
that analysts with a variety of
backgrounds must make more forecasts
than they can do manually forecasting
time series data based on an preservative
model where non-linear trends are fit with
yearly, weekly, and daily seasonality, plus
holiday effects.
95%
Accuracy

11.

03
GRAFANA
MONITORING

12.

TRAFFIC MONITORING
alert
prediction
real traffic
Data and predict
We visualized traffic and
predicted traffic to control
any anomalies and if your
traffic exceed the alert line
you will get the push

13.

04
CONFIGURE
ALERTS

14.

ALERT!
Grafana has different notification
variations. From a primitive message
to a detailed description and reason
for the notification, notifications can
also be configured for a group of
users, including messengers, using
their
API This presentation template was created
CREDITS:
by Slidesgo, including icons by Flaticon, and
infographics & images by Freepik.

15.

OUR TEAM
Aisin Anuar
Designer
Zhalgaskhan Abilkassym
ML Developer
Meraryslan Meraliyev
Supervisor