CHAPTER TWO: THE GLOBAL PICTURE
Indeed, when it comes to estimates of individual botnet size, the common technique of
reporting unique zombie IP addresses measured over long periods of time is likely to significantly
overestimate the number of infected devices.50 While botnet size measurements remain
controversial, evidence suggests that ‘successful’ bot herders typically control groups of infected
computers of the order of tens or hundreds of thousands of devices – rather than the frequently reported
‘millions’ of devices.51 On this basis, the total number of large criminal ‘commercial’ botnets globally
is likely comparatively small. In addition,
however, a much higher number of small,
Example of information harvested by botnets:
‘amateur’ botnets, consisting of low zombie
‘Torpig’
populations, may also exist.52
C&C domains controlled by academic researchers for a
period of 10 days
The harm – Such malicious networks
183,000 zombie devices identified during 10 days.
are nonetheless capable of significant harm.
Average ‘Live’ zombie population at any one time of
During one single 10 day period, a botnet of
49,000. Most zombies likely in Northern Europe and
North America
around 183,000 zombie devices was found
Victim credentials for 8,300 accounts at 400 different
to harvest almost 310,000 items of victim
financial institutions sent to C&C server
bank account, credit card, and webmail and
Details of 1,700 credit cards sent to C&C server
social networking credentials.53 As discussed
298,000 victim usernames and passwords for webmail and
in the section in this Chapter on ‘cybercrime
social networking sites sent to C&C server
perpetrators,’ the potential of botnets to
Sufficient aggregate bandwidth of zombies to launch a
massive DDoS attack
harvest such information has been
instrumental in the development of
Source: Stone-Gross et al.
cybercrime criminal ‘markets’ based largely
on botnet sale and rental.54 As noted in the
2010 UNODC Transnational Organized
Crime Threat Assessment, the market in personal information harvested through botnets can be
largely divided – with different individuals focused on collecting volumes of financial and identifying
information, selling it on, and ‘cashing it out.’55
Content offences
Overview – One third to one half of the most common cybercrime acts concern contentrelated offences.56 Content may be regulated by criminal law for a range of reasons, including where
contrary to national security, public safety, public order, health or morals, or the rights and freedoms
of others.
Globally, information from over 4,600 requests by national authorities for removal of
content from Google services shows that a wide range of material is perceived by governments to
impinge upon these areas.57 Not all such material may strictly engage the criminal law. Nonetheless,
50
51
52
53
54
55
56
57
It is likely that IP address based measurements only correspond well to the number of infected devices when reported over short
time scales, such as one hour. Unique IP addresses measured over longer periods significantly overestimate the number of devices
due to DHCP churn. In one botnet study, 1.25 million unique zombie IP addresses identified over 10 days corresponded only to
183,000 bots according to unique bot-ID (Stone-Gross, B., et al. 2009. Your Botnet is My Botnet: Analysis of a Botnet Takeover.
In: 16th Annual ACM Conference on Computer and Communications Security (CCS), 9-13 November 2009). In addition, zombie counts are
affected by the ‘no-see’ time before a device or IP address is considered to no longer be a member of the botnet (see
http://www.shadowserver.org/wiki/pmwiki.php/Stats/BotCounts).
See, for example, http://www.secureworks.com/cyber-threat-intelligence/threats/waledac_kelihos_botnet_takeover/;
http://www.secureworks.com/cyber-threat-intelligence/threats/The_Lifecycle_of_Peer_to_Peer_Gameover_ZeuS/; StoneGross, B., et al. 2009. Your Botnet is My Botnet: Analysis of a Botnet Takeover. CCS ‘09.
See, for example, http://www.symantec.com/connect/blogs/botnets-masses
Stone-Gross, B., et al., 2009. Your Botnet is My Botnet: Analysis of a Botnet Takeover. CCS ‘09.
See, for example, Panda Security, 2010. The Cybercrime Black Market: Uncovered.
UNODC, 2010. The Globalization of Crime. A Transnational Organized Crime Threat Assessment.
See above, Section 2.2 The global cybercrime picture, Distribution of cybercrime acts.
Data from www.google.com/transparencyreport
35
Select target paragraph3
Connect to a paragraph
Connect to an entity
Disable highlights
Add to table of contents