Top Threats To Cloud Computing: The Egregious 11
Top Threats To Cloud Computing: The Egregious 11
Top Threats To Cloud Computing: The Egregious 11
Cloud Computing
The Egregious 11
© 2019 Cloud Security Alliance – All Rights Reserved. You may download, store, display on your
computer, view, print, and link to the Cloud Security Alliance at https://cloudsecurityalliance.org
subject to the following: (a) the draft may be used solely for your personal, informational, non-
commercial use; (b) the draft may not be modified or altered in any way; (c) the draft may not be
redistributed; and (d) the trademark, copyright or other notices may not be removed. You may quote
portions of the draft as permitted by the Fair Use provisions of the United States Copyright Act,
provided that you attribute the portions to the Cloud Security Alliance.
Contributors
Jon-Michael Brook
Alexander Getsin
Greg Jensen
Laurie Jameson
Michael Roza
Neha Thethi
Ashish Kurmi
Shachaf Levy
Shira Shamban
Vic Hargrave
Victor Chin
Zoran Lalic
Randall Brooks
The latest report highlights the Egregious Eleven (ranked in order of significance per survey results
with applicable previous rankings):
1. Data Breaches
2. Misconfiguration and Inadequate Change Control
3. Lack of Cloud Security Architecture and Strategy
4. Insufficient Identity, Credential, Access and Key Management
5. Account Hijacking
6. Insider Threat
7. Insecure Interfaces and APIs
8. Weak Control Plane
9. Metastructure and Applistructure Failures
10. Limited Cloud Usage Visibility
11. Abuse and Nefarious Use of Cloud Services
New, highly rated items in the survey are more nuanced and suggest a maturation of the consumer’s
understanding of the cloud. These issues are inherently specific to the cloud and thus indicate a
technology landscape where consumers are actively considering cloud migration. Such topics refer
to potential control plane weaknesses, metastructure and applistructure failures, and limited cloud
visibility. This new emphasis is markedly different from more generic threats, risks and vulnerabilities
(i.e. data loss, denial of service) that featured more strongly in previous Top Threats reports.
We hope this document raises organizational awareness of the top security issues and their mitigations
—and ensures that they are taken into consideration when budgeting for cloud migration and security.
The report provides controls recommendations and reference examples that are meant to be of use
to compliance, risk, and technology staff. Executive management will also benefit from exposition
technology trends and overviews in the report.
There are cases of data breaches being undetected until months after the compromise. In such
incidents, the implications might not be immediately apparent (e.g., IP theft). For example, the
United States Office of Personnel Management (OPM) and Sony Pictures breach both had a dwell
time of approximately one year1.
Key Takeaways
1. Data is becoming the main target of cyber attacks. Defining the business value of data and
the impact of its loss is essential important for organizations that own or process data.
2. Protecting data is evolving into a question of who has access to it
3. Data is accessible via the internet is the most vulnerable asset to mis-configuration or
exploitation.
4. Encryption techniques can help protect data, but negatively impacts system performance
while making applications less user-friendly
5. A robust and well tested incident response plan that considers the CSP and data privacy
laws will help data breach victims recover.
1
Improving Cyber Resiliency https://cloudsecurityalliance.org/artifacts/improving-metrics-in-cy-
ber-resiliency/
CCM Controls
AIS Application and Interface Security EKM Encryption and Key Management
AIS-01: Application Security EKM-01: Entitlement
AIS-02: Customer Access Requirements EKM-02: Key Generation
AIS-03: Data Integrity EKM-03: Sensitive Data Protection
AIS-04: Data Security / Integrity EKM-04: Storage and Access
CCC Change Control and Configuration GRM Governance and Risk Management
Management GRM-02: Data Focus Risk Assessments
CCC-05: Production Changes GRM-06: Policy
GRM-10: Risk Assessments
DSI Data Security and Information Lifecycle
Management IAM Identity and Access Management
DSI-01: Classification IAM-01: Audit Tools Access
DSI-02: Data Inventory / Flows IAM-04: Policies and Procedures
DSI-03: Ecommerce Transactions
DSI-04: Handling / Labeling / Security Policy
DSI-05: Non-Production Data
DSI-07: Secure Disposal
© Copyright 2019, Cloud Security Alliance. All rights reserved. 7
THREAT ANALYSIS LINKS AND REFERENCES
Spoofing Identity 1. Timehop Security Incident, July 4, 2018: https://www.timehop.com/
Tampering with Data security/
Repudiation 2. Uber Discloses Year-Old AWS Data Breach, Exposing Millions of Users:
Information Disclosure https://awsinsider.net/articles/2017/11/21/uber-aws-data-breach.aspx
Denial of Service 3. Amazon hit with major data breach days before Black Friday: https://www.
Elevation of Privilege theguardian.com/technology/2018/nov/21/amazon-hit-with-major-data-
breach-days-before-black-friday
4. VOIPO database exposed millions of call and SMS logs, system data:
https://www.zdnet.com/article/voipo-database-exposed-millions-of-call-
and-sms-logs-system-data/
Business Impact
The business impact of a misconfigured item can be severe depending on the nature of the
misconfiguration and how quickly it is detected and mitigated. The most commonly reported effect
is the exposure of data stored in cloud repositories.
Key Takeaways
1. Cloud-based resources are highly complex and dynamic, making themit challenging to
configure.
2. Traditional controls and change management approaches are not effective in the cloud.
3. Companies should embrace automation and employ technologies that scan continuously
for misconfigured resources and remediate problems in real time.
1. A misconfigured AWS Simple Storage Service (S3) cloud storage bucket exposed detailed
and private data of 123 million American households. The data set belonged to Experian, a
credit bureau, which sold the data to an online marketing and data analytics company called
Alteryx. It was Alteryx that exposed the file.
2. An unsecured Elasticsearch database owned by Exactis resulted in another massive breach
containing highly personal data of 230 million U.S. consumers. The database server was
configured to be publicly accessible.
3. Level One Robotics, an engineering company specializing in automation process and
assembly, exposed highly sensitive proprietary information belonging to more than 100
manufacturing companies, including Volkswagen, Chrysler, Ford, Toyota, General Motors,
Tesla and ThyssenKrupp. In this case, the misconfigured asset was an rsync (backup) server
that allowed unauthenticated data transfer to any rsync client.
CCM Controls
AIS Application and Interface Security EKM Encryption and Key Management
AIS-01: Application Security EKM-03: Sensitive Data Protection
AIS-04: Data Security / Integrity EKM-04: Storage and Access
CCC Change Control and Configuration GRM Governance and Risk Management
Management GRM-01: Baseline Requirements
CCC-02: Outsourced Development GRM-02: Data Focus Risk Assessments
CCC-03: Quality Testing
CCC-05: Production Changes HRS Human Resources
HRS-09: Training / Awareness
DSI Data Security and Information Lifecycle
Management IAM Identity and Access Management
DSI-01: Classification IAM-02: Credential Lifecycle / Provision
DSI-04: Handling / Labeling / Security Policy Management
IAM-05: Segregation of Duties
Business Impact
No matter how big or small the enterprise, proper security architecture and strategy are required
elements for securely moving, deploying, and operating in the cloud. Successful cyberattacks
can have a severe impact on businesses, including financial loss, reputational damage, legal
repercussions, and fines.
Key Takeaways
1. Ensure security architecture aligns with business goals and objectives.
2. Develop and implement a security architecture framework
3. Ensure the threat model is continuously kept up to date.
4. Bring continuous visibility into the actual security posture
• Technology and cloud giant Accenture recently confirmed it inadvertently left a massive
store of private data across four unsecured cloud servers, exposing highly sensitive
passwords and secret decryption keys that could have inflicted considerable damage on
the company and its customers. The servers, hosted on Amazon’s S3 storage service,
contained hundreds of gigabytes of data for the company’s enterprise cloud offering, which
the company said provides support to the majority of the Fortune 100 companies. The data
could be downloaded without a password by anyone who knew the servers’ web addresses.
• Researchers at Kromtech Security Center discovered a trove of data belonging to the Honda
Connect App, which was exposed online. The data was stored on two unsecured, publicly
accessible and unprotected Amazon AWS S3 Buckets.
CCM Controls
IVS Infrastructure and Virtualization Security
AIS Application and Interface Security
AIS-04: Data Security / Integrity IVS-06: Network Security
IVS-08: Production / Non-Production
GRM Governance and Risk Management Environments
GRM-01: Baseline Requirements IVS-09: Segmentation
GRM-02: Data Focused Risk Assessments IVS-13: Network Architecture
GRM-05: Management Support/Involvement
GRM-08: Management Policy STA Supply Chain Management, Transparency,
and Accountability
IAM Identity and Access Management STA-03: Network / Infrastructure Services
IAM-02: Credential Lifecycle / Provision STA-05: Supply Chain Agreements
Management
Credentials and cryptographic keys must not be embedded in source code or distributed in public-
facing repositories (such as GitHub) because there is a high risk of discovery and misuse. Keys need to
be appropriately secured, and a well-secured public key infrastructure (PKI) is required to ensure key-
management activities are carried out.
Identity management systems must scale to handle lifecycle management for millions of users as
well as CSPs. Identity management systems must support immediate de-provisioning of access
to resources with personnel changes, such as job termination or role transitions. Such identity
management lifecycle processes should be integrated and automated within cloud environments and
accomplished a timely manner.
Identity systems are becoming increasingly interconnected, and federating identity with a cloud
provider (e.g., Security Assertion Markup Language (SAML)) is becoming more prevalent to ease
the burden of user maintenance. Organizations planning to federate identity with a cloud provider
must understand the security around the cloud provider’s identity solution, including processes,
infrastructure, and segmentation between customers (in the case of a shared identity solution).
In cases where legacy systems require the use of passwords alone, the authentication system must
support policy enforcement, such as the verification of strong passwords and organizational-defined
rotation period policies.
The management of cryptographic keys used to protect data at rest must occur throughout their
lifecycles, including creation, distribution, storage, replacement, and deletion. Doing so helps
address attacks that feature unauthorized access to keys. Stolen cryptographic keys—coupled with a
lack of key rotation policy–may dramatically increase effective elapsed breach time and scope.
Any centralized storage mechanism containing data secrets (e.g., passwords, private keys, or confidential
customer contact databases) is an extremely high-value target for attackers. Choosing to centralize
passwords and keys is a compromise that an organization must consider carefully: the convenience
of centralized key management against the threat of grouping these keys. As with any high-value
asset, monitoring, and protection of identity and key management systems should be a high priority.
Business Impact
Malicious actors masquerading as legitimate users, operators or developers can read/exfiltrate,
modify and delete data; issue control plane and management functions; snoop on data in transit; or
release malicious software that appears to originate from a legitimate source. As a result, insufficient
identity, credential, or key management can enable unauthorized access to data and potentially
catastrophic damage to organizations or end-users.
Key Takeaways
1. Secure accounts, inclusive to two-factor authentication and limited use of root accounts.
2. Practice the strictest identity and access controls for cloud users and identities.
3. Segregate and segment accounts, virtual private clouds (VPCs) and identity groups based
on business needs and the principle of least privilege.
4. Rotate keys, remove unused credentials and privileges, employ central and programmatic
key management.
• In December 2018, a German student hacked data protected by weak passwords and
shared the information using a cloud platform. The 20-year-old utilized passwords such
as “Iloveyou” and “1234” to hack into online accounts of hundreds of lawmakers and
personalities whose political stances he disliked. German cybersecurity officials revealed
CCM Controls
EKM Encryption and Key Management HRS Human Resources
EKM-01: Entitlement HRS-01: Asset Returns
EKM-02: Key Generation HRS-03: Employment Agreements
EKM-03: Sensitive Data Protection HRS-04: Employment Termination
EKM-04: Storage and Access HRS-08: Technology Acceptable Use
HRS-09: Training / Awareness
HRS-10: User Responsibility
Business Impact
Account and service hijacking implies full compromise: control of the account, its services, and data
within. In such a scenario, business logic, function, data, and applications reliant on the account
services are at-risk.
The fallout from such compromises has been severe at times. In recent breach cases, there were
significant operational and business disruptions—including examples of the complete elimination of
organization assets, data, and capabilities.
Account hijacking consequences include data leaks that lead to reputational damage, brand value
degradation, legal liability exposure, and sensitive personal and business information disclosures.
Key Takeaways
1. Account Hijacking is a threat that must be taken seriously
2. Defence-in-depth and IAM controls are key in mitigating account hijacking
• In June 2014, the AWS account of Code Spaces—a former code-hosting service company—
was compromised when it failed to protect its administrative console with multi-factor
authentication. The business was forced to close after the destruction of its assets.
• In 2018, consumer cloud services hijacked and sold data in darknet marketplaces at
commercial scale.
• 2017 marked the rise of cloud account-targeted campaigns, in particular for Microsoft
Office 365.
• In April 2010, an Amazon cross-site scripting (XSS) bug enabled credentials theft, and in
2009, numerous Amazon systems were hijacked to run Zeus botnet nodes.
CCM Controls
BCR Business Continuity Management and IVS Infrastructure and Virtualization Security
Operational Resilience IVS-01: Audit Logging/Intrusion Detection
BCR-01: Business Continuity Planning IVS-08: Production / Non-Production
Environments
IAM Identity and Access Management
IAM-02: Credential Lifecycle / Provision SEF Security Incident Management, E-Discovery,
Management and Cloud Forensics
IAM-05: Segregation of Duties SEF-01: CSP points of contact for IR escalation
IAM-08: Trusted Sources and support
IAM-10: User Access Reviews
IAM-11: User Access Revocation
Insider threats are more prevalent than you may think. The CLOUD SERVICE MODEL
Netwrix 2018 Cloud Security Report indicates that 58 percent
Software as a Service (SaaS)
of companies attribute security breaches to insiders. Insider Platform as a Service (PaaS)
negligence is the cause of most security incidences. Infrastructure as a Service (IaaS)
Business Impact
Insider threats can result in the loss of proprietary information and intellectual property. System
downtime associated with attacks can negatively impact company productivity. Additionally, data
loss or other customer harm can reduce confidence in company services.
Dealing with insider security incidents involves containment, remediation, incident response,
investigation, post-incidence analysis, escalation, monitoring, and surveillance. These activities can
add significantly to a company’s workload and security budget. The Ponemon Institute reported
that—of the companies interviewed—the average cost of insider incidents in 2017 (per company) was
more than $8.7 million, with the maximum cost running as high as $26.5 million.
• In June 2018, Tesla CEO Elon Musk sent an e-mail to Tesla employees alleging there was a
saboteur within his company’s ranks. The saboteur, a disgruntled employee, allegedly used
false usernames to make changes to the code used in the Tesla Manufacturing Operation
System. The employee also exported “large amounts of highly sensitive Tesla data to
unknown third parties.”
• Also in 2018, an employee at the Punjab National Bank in India gained unauthorized access
to a sensitive password in the SWIFT interbank transaction system to release funds in a
fraudulent transactional chain. A diamond merchant created the scheme to buy rough
stones from suppliers. The total price tag for the bank: $1.8 billion.
• According to the IBM X-Force Threat Intelligence Index 2018: “Misconfigured cloud servers,
networked backup incidents, and other improperly configured systems were responsible
for the exposure of more than 2 billion records, or nearly 70 percent of the total number of
compromised records tracked by X-Force in 2017.”
DSI Data Security and Information Lifecycle IAM Identity and Access Management
Management IAM-01: Audit Tools Access
DSI-04: Handling / Labeling / Security Policy IAM-05: Segregation of Duties
DSI-06: Ownership / Stewardship IAM-08: Trusted Sources
IAM-09: User Access Authorization
EKM Encryption and Key Management IAM-10: User Access Reviews
EKM-02: Key Generation IAM-11: User Access Revocation
EKM-03: Sensitive Data Protection
IVS Infrastructure and Virtualization Security
GRM Governance and Risk Management IVS-09: Segmentation
GRM-03: Management Oversight
GRM-04: Management Program STA Supply Chain Management, Transparency
GRM-06: Policy and Accountability
GRM-07: Policy Enforcement STA-09: Third Party Audits
GRM-10: Risk Assessments
Business Impact
While most providers strive to ensure that security is well-integrated into their service models, it is
critical for consumers of those services to understand the security implications associated with the use,
management, orchestration, and monitoring of cloud services. Reliance on a weak set of interfaces and
APIs exposes organizations to a variety of security issues related to confidentiality, integrity, availability,
and accountability. Additionally, regulatory and financial impacts could be very significant.
Key Takeaways
1. Practice good API hygiene. Good practice includes diligent oversight of items such as
inventory, testing, auditing, and abnormal activity protections.
2. Ensure proper protection of API keys and avoid reuse.
3. Consider using standard and open API frameworks (e.g., Open Cloud Computing Interface
(OCCI) and Cloud Infrastructure Management Interface (CIMI)).
• Facebook announced a significant data breach affecting more than 50 million accounts
on Sept. 28, 2018. Reportedly, credential theft vulnerability was introduced into Facebook
code in July of 2017, more than a year earlier. The company admitted it didn’t know what
information was stolen, nor how many other user accounts were compromised as a result of
the breach.
CCM Controls
AIS Application and Interface Security
AIS-01: Application Security
AIS-03: Data Integrity
AIS-04: Data Security / Integrity
Business Impact
A weak control plane could result in data loss, either by theft or corruption. This could lead to a
massive business impact, particularly if data loss includes private user data. Regulatory punishment
for data loss may be incurred as well. For example, under General Data Protection Regulation (GDPR)
regulations, incurred penalties can reach €20M— or four percent— of global revenue.
With a weak control plane, users may also be unable to protect their cloud-based business data and
applications, which can lead to frustration and a loss of confidence in the service or the product
provided. Ultimately, this may translate to a revenue decrease.
Key Takeaways
1. Adequate security controls provided through a CSP are necessary so that cloud customers
can fulfill their legal and statutory obligations.
2. The cloud customer should perform due diligence and determine if the cloud service they
intend to use possesses an adequate control plane.
• The management plane of a cloud service is very critical and needs to be adequately
protected by identity and access controls. Two-factor authentication should be part of
the standard suite of controls provided to a cloud customer by the CSP. Unfortunately,
many CSPs only make two-factor authentication available to their customers as a premium
service. Such practices weaken the security posture of cloud customers—particularly those
who do not or cannot utilize this premium service.
CCM Controls
AIS Application and Interface Security IVS Infrastructure and Virtualization Security
AIS-03: Data Integrity IVS-01: Audit Logging / Intrusion Detection
AIS-04: Data Security / Integrity IVS-04: Information System Documentation
IVS-06: Network Security
AAC Audit Assurance and Compliance IVS-09: Segmentation
AAC-03: Information System Regulatory Mapping IVS-13: Network Architecture
Above the waterline, cloud consumers must understand how to properly implement cloud
applications to fully utilize the cloud platform. For example, applications that are not designed
for cloud environments will not be able to fully interact and use available cloud resources and
capabilities. Merely taking a “lift-and-shift” approach does not suffice when migrating business
operations and applications to the cloud.
Business Impact
Metastructure and applistructure are critical components of a cloud service. Failures involving
these features at the CSP level can severely impact all service consumers. At the same time,
misconfigurations by the tenant could disrupt the user financially and operationally.
Key Takeaways
1. Cloud service providers must offer visibility and expose mitigations to counteract the
cloud’s inherent lack of transparency for tenants.
2. Cloud tenants should implement appropriate features and controls in cloud native designs.
3. All CSPs should conduct penetration testing and provide findings to customers.
EKM Encryption and Key Management STA Supply Chain Management, Transparency and
EKM-02: Key Generation Accountability
EKM-03 - Sensitive Data Protection STA-03: Network / Infrastructure Services
Un-sanctioned app use: This occurs when employees are using ARCHITECTURE
cloud applications and resources without the specific permission
Infra
and support of corporate IT and security. This scenario results
Meta
in a self-support model called Shadow IT. When insecure cloud Info
services activity does not meet corporate guidelines, this Appli
behavior is risky— especially when paired with sensitive corporate
data. Gartner predicts that by 2020, one-third of all successful CLOUD SERVICE MODEL
security attacks on companies will come through shadow IT Software as a Service (SaaS)
systems and resources. Platform as a Service (PaaS)
Infrastructure as a Service (IaaS)
Sanctioned app misuse: Organizations are often unable to
analyze how their approved applications are being leveraged by
insiders who use a sanctioned app. Frequently, this use occurs without the explicit permission of the
company, or by external threat actors who target the service using methods such as credential theft,
Structured Query Language (SQL) injection, Domain Name System (DNS) attacks and more.
In most cases, it comes down to discerning valid and invalid users from one another; if their
behaviors are out of the norm; and if they abide by corporate policies.
Business Impact
The risks are widespread but can be summed up with the following points:
• Lack of governance: When employees are unfamiliar with proper access and governance
controls, it is common to see sensitive corporate data placed in public access locations vs.
private access locations.
• Lack of awareness: When data and services are in use without the knowledge of the company,
they are, in essence, unable to control their IP. The employee has the data, not the company.
• Lack of security: When an employee incorrectly sets up a cloud service, it can become
exploitable not only for the data that resides on it but for future data.Malware, botnets,
cryptocurrency mining malware, and more can compromise cloud containers—which puts
organizational data, services, and finances at risk.
When asked about the impact of unsanctioned cloud use in their respective environments, 50 percent
of the respondents cited in the Oracle and KPMG Cloud Threat Report 2019 indicated this unsanctioned
use has led to “unauthorized access to data,” and another 48 percent cited the “introduction of
malware” as a result.
© Copyright 2019, Cloud Security Alliance. All rights reserved. 34
Key Takeaways
1. Mitigating these risks starts with the development of a complete cloud visibility effort from
the top down. This process usually originates with tasking an organization’s cloud security
architect with the creation of a comprehensive solution that ties into people, process, and
technology. Actions outlined below can help jumpstart this process.
2. Mandate companywide training on accepted cloud usage policies and enforcement thereof.
3. All non-approved cloud services must be reviewed and approved by the cloud security
architect or third-party risk management.
4. Invest in solutions like cloud access security brokers (CASB) or software defined gateway
(SDG) to analyze outbound activities and help discover cloud usage, at-risk users, and to
follow behavior usage of credentialed employees to identify anomalies.
5. Invest in a web application firewall (WAF) to analyze all inbound connections to your cloud
services for suspicious trends, malware, distributed denial-of-service (DdoS), and Botnet risks.
6. Select solutions that are specifically designed to monitor and control all of your key enterprise
cloud applications (enterprise resource planning, human capital management, commerce
experience, and supply chain management) and ensure suspicious behaviors can be mitigated.
7. Implement a zero-trust model across your organization.
• According to 2018 research conducted by cloud security firm Lacework: “More than 22,000
container orchestration and API management systems are unprotected or publicly available
on the internet – highlighting the reality of the risks of operating workloads in the cloud.”
• The “Skyhigh Networks Cloud Adoption & Risk Report Q2 2015 reported that “the average
enterprise now uses 1,083 cloud services. That astounding figure is almost 50 percent
higher than this time last year, and up to 100 percent from two years ago.”
• Of those 1,000-plus cloud services in use today, the Skyhigh Networks Cloud Adoption &
Risk Report Q2 2015 stated that many might fall into the category of Shadow IT. Simply put:
the IT department had no role in helping to select and deploy the services of these Shadow
IT services, and might not even know they are being used.”
Business Impact
If an attacker has compromised the management plane of a customer’s cloud infrastructure, the
attacker can use the cloud service for illicit purposes while the customer foots the bill. The bill could
be substantial if the attacker consumed substantial resources, such as mining cryptocurrency.
Alternatively, attackers can also use the cloud to store and propagate malware. Enterprises must be
aware and have controls in place to deal with these new attack vectors. This may mean procuring
security technology that can monitor cloud infrastructure or API calls from and to the cloud service.
Key Takeaways
• Enterprises should monitor their employees in the cloud, as traditional mechanisms are
unable to mitigate the risks posed by cloud service usage.
• Employ cloud data loss prevention (DLP) technologies to monitor and stop any unauthorized
data exfiltration.
• The Zepto variant of the Locky ransomware spreads via cloud services such as Microsoft
OneDrive, Google Drive, and Box by sharing a malicious file with potential victims.
• The CloudSquirrel attack arrives via an email phishing attack. This attack email attempts to
trick its victim into opening its message using an important-sounding link (such as a “tax
invoice”). Once opened, CloudSquirrel infects users by downloading additional malicious
encrypted payloads via a Java Archive (JAR) file. The malware then establishes a connection
with its command and control hosted in Dropbox. Its commands masquerade as plain text
files with fake extensions such as.mp4, .wmv, .png, .dat, and .wma.
CCM Controls
AIS Application and Interface Security HRS Human Resources
AIS-02: Customer Access Requirements HRS-05: Mobile Device Management
HRS-08: Technology Acceptable Use
BCR Business Continuity Management HRS-09: Training / Awareness
and Operational Resilience
BCR-09: Impact Analysis IAM Identity and Access Management
IAM-02: Credential Lifecycle / Provision
CCC Change Control and Configuration Management
Management IAM-04: Policies and Procedures
CCC-02: Outsourced Development IAM-05: Segregation of Duties
IAM-09: User Access Authorization
DSI Data Security and Information Lifecycle IAM-10: User Access Reviews
Management IAM-11: User Access Revocation
DSI-01: Classification IAM-12: User ID Credentials
DSI-02: Data Inventory / Flows
DSI-04: Handling / Labeling / Security Policy IVS Infrastructure and Virtualization Security
IVS-01: Audit Logging / Intrusion Detection
EKM Encryption and Key Management IVS-02: Change Detection
EKM-03: Sensitive Data Protection IVS-06: Network Security
IVS-13: Network Architecture
GRM Governance and Risk Management
GRM-01: Baseline Requirements
Considering that user interfaces and APIs are the modern way to consume services, it is concerning
that there are still significant challenges when it comes to securing these features.
The cloud—with its complexity—is also the perfect place for attackers to hide. It is also, unfortunately,
an ideal launchpad for attacks. Last but not least, insider threats make it more challenging to protect
organizations from data loss.
This Top Threats in Cloud Computing report suggests an interesting and somewhat new perspective
on cloud security. This new outlook focuses on configuration and authentication, and shifts away
from the traditional focus on information security (e.g., vulnerabilities and malware). Regardless,
these security issues are a call to action for developing and enhancing cloud security awareness,
configuration, and identity management.
In the first stage of research, the group’s goal was to create a shortlist of cloud security concerns.
The group started with a list of 26 security concerns (updating the previous report’s 12 and adding
14 new issues). The group discussed the 26 points in a series of meetings, asking working group
members to indicate the importance of each matter in relation to their respective organization.
This stage of the research also provided the opportunity for working group members to suggest
additional concerns not included in the list of 26. After considering all the survey results and other
information, the working group identified the top 19 most salient cloud security concerns.
In the second stage of the research, the group’s main goal was to rank—via importance—this
condensed list of 19. The group wanted the study to capture what security professionals thought
were the most relevant cloud security concerns, so a 10-point sliding scale was chosen as the
research instrument. Respondents were instructed to rate cloud security issues from “1 to 10,” with
“1” being “very insignificant” and “10” being “very significant”. The points for each category were
averaged, and the security concerns were then ranked according to their mean. The working group
then arrived at the top 11 by excluding all security issues with a mean of less than seven.
Finally, the working group also analyzed the security concerns using the STRIDE threat model,
which was developed by Microsoft to evaluate information security threats. Specifically, the security
concerns discussed in this paper are evaluated to determine whether they fall into any of the
following threat categories: