Untitled

Cafaro, Massimo; Epicoco, Italo; Mirto, Maria; Lezzi, Daniele; Aloisio, Giovanni

doi:10.1002/cpe.v20:15

Cited by 11 publications

(1 citation statement)

References 96 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other SAs taking into account the workflow inside the cloud have been published by Liu et al who created three heuristic based algorithms. One of these SAs is for example "throughput maximization" [8] which tries to utilize global load balancing using the Overall Average Load (OAL) algorithm and local optimization of tasks by the extended min-min algorithm (EMM). A combination between global and local scheduling also exists.…”

Section: Introductionmentioning

confidence: 99%

C<sup>3</sup>-Sched — A cache covert channel robust cloud computing scheduler

Betz

Westhoff

2014

The 9th International Conference for Internet Technology and Secured Transactions (ICITST-2014)

View full text Add to dashboard Cite

Several cloud schedulers have been proposed in the literature with different optimization goals such as reducing power consumption, reducing the overall operational costs or decreasing response times. A less common goal is to enhance the system security by applying specific scheduling decisions. The security risk of covert channels is known for quite some time, but is now back in the focus of research because of the multitenant nature of cloud computing and the co-residency of several pertenant virtual machines on the same physical machine. Especially several cache covert channels have been identified that aim to bypass a cloud infrastructure's sandboxing mechanism. For instance, cache covert channels like the one proposed by Xu et. al. use the idealistic scenario with two alternately running colluding processes in different VMs accessing the cache to transfer bits by measuring cache access time. Therefore, in this paper we present a cascaded cloud scheduler coined C 3 -Sched aiming at mitigating the threat of a leakage of customers data via cache covert channels by preventing processes to access cache lines alternately. At the same time we aim at maintaining the cloud performance and minimizing the global scheduling overhead.

show abstract

Section: Introductionmentioning

confidence: 99%