Technical Report

ATTACKING SOFTWARE-DEFINED NETWORKS:
A FIRST FEASIBILITY STUDY

TECHNICAL REPORT
Fingerprinting a SDN network
If a client sends packets to a SDN network, this client will observe different response times,
because the flow setup time can be added in the case of New-Flow (i.e., no flow rule for
handling packets in the data plane) compared with the case of Existing-Flow (i.e., there is a flow
rule for handling packets in the data plane). To describe this more clearly, we simply formalize
the response time that is observed at a client side. First, we define the response time for the
Existing-Flow case as α, and the additional flow setup time as β. In addition, for brevity, we
define the response time for the case of New-Flow and Existing-Flow as T1 and T2 respectively,
and they can be represented as follows. T1 (w/o flow rule in the data plane) = α + β T2 (w/ flow
rule in the data plane) = α.
In this case, if an attacker can clearly differentiate T1 from T2, he/she can fingerprint a SDN
network. However, an attacker will still face two challenges: (i) how to collect T1 and T2 values,
and (ii) how to know whether T1 values are different from T2 when considering random noises.
SDN Scanner: The first challenge can be addressed by our new network scanning method, header
field change scanning, which scans networks as changing network header fields. When SDN
scanner collects T1 and T2 values, it follows the following steps.
First, it sends two (or more) specifically crafted packets to a target network and records the
response time of each packet. At this time, SDN scanner considers the response time for the first
packet could represent T1, and the time for the second packet shows T2. And, SDN scanner
repeat this operation by changing a field of the packet header. Finally, SDN scanner collects T1
and T2 for each different header field. This operation is presented in Figure 1.
Page | 1
Figure 1: Simplified function diagram of SDN scanner
Statistical Testing for Two Sample Sets: Once an attacker collects samples of T1 and T2 using
SDN scanner, he/she now faces the second challenge, which can be solved by employing
statistical testing methods, such as ttest [2]. This method simply tests whether two sample sets
are significantly different from each other or not with a high confidence. This test just requires
the mean and standard deviation values of each sample that can be easily obtained, and the test
method is pretty simple. Of course, an attacker can easily use more advanced statistics or
machine learning techniques to improve the accuracy.
Launching DoS attacks to a SDN network
If an attacker runs SDN scanner and collects network information, he/she can investigate
whether a target network is using SDN or not through a simple statistical testing method. If the
test results show that a target network is likely to use SDN, the attacker will further conduct the
resource consumption attack. Since the attacker already knows the condition of the flow rule for
the target network (with the help of SDN scanner), now he/she just needs to send network
packets to consume SDN resources of the target network.
FEASIBILITY STUDY OF THE POTENTIAL USE

OF DRILL CUTTINGS IN CONCRETE
TECHNICAL REPORT
Page | 2
Drill Cuttings
Figure 2 shows the drill cuttings aggregate size distribution. The maximum grain size is less than
6 mm, and coefficient of uniformity (Cu) and curvature (Cc) are 8.63 and 1.22, respectively. As
can be seen in Figure 2, the drill cutting sample is classified as fine aggregates in accordance
with ASTM-C330.
Figure 2. Drill cuttings aggregate size distribution 3.2. Supplementary Cementitious Materials
Flyash and silica fume are two supplementary cementitous materials which are widely used in
concrete production. To investigate the effect of flyash and silica fume on the compressive
strength of concrete containing drill cuttings, samples including specified percentages of
aforementioned materials were made.
Flyash is a byproduct of coal combustion which is directly emitted to air and cause air pollution.
Currently, because of the air pollution concerns, different strategies are applied to collect flyash
from gas streams and use it for other applications. During past decades, flyash is extensively
used as an additive in concrete. Using flyash in concrete lowers the hydration temperature and
increases the compressive strength because of pozzolanic effects (22, 23). Similarly, silica fume
with small particle size which has pozzolanic behavior can fill the space between aggregates and
cement, consequently increase the strength and durability of concrete (24, 25).
Sample Preparation
The test specimens were cast from 5 separate batches of concrete: one control, 4 mixes
containing different percentages of drill cuttings, fly ash, silica fume, and a mixture of silica
fume and fly ash. Drill cuttings obtained from drilling site (Figure 3a), dried in oven at
100~105ºC for 24 hours first, then passed through Sieve No. 100. Concrete mixture was
designed according to ACI recommendations. Portland cement type I was used in this study as
cementitious material. Cement and drill cuttings were thoroughly mixed to obtain a uniform
colour. The coarse aggregate was a crushed limestone with 100% passing the sieve No.3/8-in.
Page | 3
(9.5-mm) and 20% passing the sieve No. 4 (4.75-mm) and with none passing the sieve No. 8
(2.36-mm). The coarse aggregate had absorption of 1.8% and relative density of 2.63, whereas
the fine aggregate absorption was 0.5% with relative density of 2.61. The cylindrical specimen
20 cm (8 in) × 10 cm (4 in) were cast as shown in Figure 3b, and were tested in triplicate The
reported strength values represented the average strength of three specimens. After being
stripped from molds, the specimens were submerged in water for 7 days at room temperature.
The mixture proportions associated are shown in Table 1. To compensate the loss in compressive
strength, different percentages of silica fume and fly ash was added to the mixture as an additive.
Page | 4

Technical Report

Uploaded by

Copyright:

Available Formats

Technical Report

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Technical Report

Uploaded by

Copyright:

Available Formats

ATTACKING SOFTWARE-DEFINED NETWORKS:

A FIRST FEASIBILITY STUDY

Fingerprinting a SDN network

FEASIBILITY STUDY OF THE POTENTIAL USE

You might also like