No abstract

No abstract

The paper presents a survey and analysis of the current security measures implemented in cloud computing and the hypervisors that support it. The viability of an efficient virtualization layer has led to an explosive growth in the cloud computing industry, exemplified by Amazon's Elastic Cloud, Apple's iCloud, and Google's Cloud Platform. However, the growth of any sector in computing often leads to increased security risks. This paper explores these risks and the evolution of mitigation techniques in open source cloud computing. Unlike uniprocessor security, the use of a large number of nearly identical processors acts as a vulnerability amplifier: a single vulnerability being replicated thousands of times throughout the computing infrastructure. Currently, the community is employing a diverse set of techniques in response to the perceived risk. These include malware prevention and detection, secure virtual machine managers, and cloud resilience. Unfortunately, this approach results in a disjoint response based more on detection of known threats rather than mitigation of new or zero-day threats, which are often left undetected. An alternative way forward is to address this issue by leveraging the strengths from each technique in combination with a focus on increasing attacker workload. This approach would make malicious operation time consuming and deny persistence on mission time-scales. It could be accomplished by incorporating migration, non-determinism, and resilience into the fabric of virtualization.

The x86 multicore processor architecture and the concepts associated with symmetric multiprocessing have become the linchpin of modern operating systems and cloud computing. A solid understanding of these technologies has become critical to any developer entering the field. Unfortunately, the complexity associated with discovering, enabling, using, and virtualizing multiple cores has created a paucity in the available documentation, transferable knowledge, and readable code exemplars. This paper describes our experience in overcoming these hurdles in the design of a from scratch, multi-core operating system--Bear--for secure and resilient cloud computing. In particular, we trace the intricacies involved in the development of a multi-core microkernel with an integrated multi-core hypervisor. By exploring the implementation details, from bootstrapping through core virtualization to process scheduling, this paper aims to fill the knowledge gaps, highlight potential pitfalls, and introduce multicore development in a concise start-to-finish exemplar.

In the world of computer security, attackers are constantly looking for new exploits to gain data from or control over a computer system. One category of exploit that can prove quite effective at accessing privileged data is side channel exploits. These exploits attempt to take advantage of vulnerabilities that are inherent in the design of a system rather than vulnerabilities in the code that has been written for and is running on said system. In other words, they exploit side effects of computation. Examples of this include measuring the power consumption of a system’s processor over time and analysing that power usage to leak system secrets or reading secrets from a system by analysing the electromagnetic radiation the system leaks as it processes data. Another type of side channel attack is a cache-based side channel attack, which exploits the timings of cache and memory accesses to determine data from the target system. We discuss some of the more common types of side channel attacks used to interpret data values from the microarchitectural changes created by transient executions. In particular, we will focus on attacks that are capable of recovering data that is processed through transient execution in some way and then wrongly accessed using a side channel, such as the Spectre and Meltdown classes of attack. We also discuss other attacks of a similar type and survey some popular mitigations for these attacks. We provide a survey of all available Spectre proof-of-concept repositories on GitHub, evaluating whether they work on different platforms. Finally, we review our experiences with these types of attacks on modern systems and comment on the attacks’ practicality, reliability, and portability. We conclude that these types of attacks are interesting, but there are some practicality and reliability concerns that make other attacks easier much of the time.