Traditional security problems in another context

Data Security

• Confidentiality, Integrity, Authenticity

Virtual machine

• Integrity, Authenticity

Hypervisor

• Integrity, Authenticity

• Availability - Consider the case where the Google is not available
• Data loss – Any personal information can be used against you
• Transfer Bottlenecks
• Bugs
• Reputation – Breaches can reduce trust

Confidentiality

• Will the sensitive data stored on a cloud remain confidential? Will cloud compromises leak confidential client data (i.e., fear of loss of control over data)
• Will the cloud provider itself be honest and won’t peek into the data?

Privacy issues raised via massive data mining

• Cloud now stores data from a lot of clients, and can run data mining algorithms to get large amounts of information on clients

Integrity

• How do I know that the cloud provider is doing the computations correctly?
• How do I ensure that the cloud provider really stored my data without tampering with it?

Availability

• Will critical systems go down at the client, if the provider is attacked in a Denial of Service attack?
• What happens if cloud provider goes out of business?

• We can map the cloud infrastructure

• We can install a new VM

• Try to get information from another VM

Create the cloud map

Use existing tools such as:

• nmap

• Hping

• Wget

Since the cloud infrastructure has a pair of external – internal addresses, we can rely on DNS to find this relation

• requires an internal user

• e.g. the EC2 gives a cloud based DNS service

Emerging technology renovation comes along with emerging threats and security issues. 

In order to facilitate analysis of the security risks faced in Cloud Systems, it is necessary to classify the identified threats into distinct categories.

The following sections present our proposal for such a classification, utilising three main categories:

• infrastructure and host related threats that affect the infrastructure of the cloud
• service provider related risks that may affect the clients, who seek a service in the cloud and
• generic threats that may affect both the infrastructure and the service providers/clients. 

• Natural disasters
• Unauthorised physical access to facilities or equipment
• Deficient training/negligence of employees
• Dumpster diving: Dumpster diving
• Password guessing
• Unauthorised access to data or information technology systems

• Compromisation of operational security logs
• Network breaks
• Privilege escalation
• Insecure or ineffective data deletion
• Malicious scanning or observation
• Insecure or obsolete cryptography
• Economic Denial of Service (EDoS) and exhaustion of resources
• Isolation malfunction
• Billing fraud

• Insufficient logging and monitoring
• Cloud Service failure or termination
• Failure of third party suppliers
• Lock in
• Compliance problems

• Cloud data provenance, metadata management and jurisdiction. This includes:

• Cloud Process Provenance
• Cloud Data Provenance
• Cloud Workflow Provenance
• System (or Environment) Provenance

• Infrastructure’s modifications
• Data processing
• Administrative and ownership changes
• DoS to co-tenants due to misjudgement or misallocation of resources
• Subpoena and e-discovery

• Replay Attacks
• Data interception:
• Man in the middle: The attacker can impersonate the victim by changing the public key/user association.
• Eavesdropping: Gain information or to create a foundation for a later attack. (Data scavenging, traffic or trend analysis, social engineering, sniffing, etc.)
• Side channel: Attackers are able to infiltrate into sensitive data, across virtual machines of the cloud infrastructure

• Browser security
• XML signature element wrapping
• Injection vulnerabilities
• Customer’s negligence and Cloud Security
• Management interface exposure
• Loss of governance

• Social engineering attacks
• Distributed Denial of Service (DDoS)
• Encryption keys exposure or loss
• Service Engine exposure
• Malware and Trojan horses
• Malicious Insider of Cloud Provider