Testing Performance-Isolation in Multi-core Systems

Danielsson, Jakob; Seceleanu, Tiberiu; Jägemar, Marcus; Behnam, Moris; Sjödin, Mikael

doi:10.1109/compsac.2019.00092

Cited by 18 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand the response time of the proposed classifier slightly increases then stabilizes[29]. Indicating that the performance of the proposed classifier is not adversely affected by the number of rules[2].…”

mentioning

confidence: 78%

“…In addition these results are consistent with what we would expect and also with research done in [32] where experiments were performed to compare the performance of a IPtabels which is a list based firewall versus the optimized list based firewall. The results showed that the performance of the optimized list based firewall is not drastically affected by the increase in the number of rules unlike that of list based firewall [2]. With the list based classifier, the accuracy is 56 percent as compared to that of optimized solution of 89%.…”

Section: Performance Evaluationmentioning

confidence: 95%

“…The hits were to determine the positioning of rules for optimal performance [35,42,43].After the establishment of hits distribution then the sorting Algorithm 1 was applied.Algorithm 1 takes an array of rules and priorities as input then returns a sorted array or rules based on priority. For optimal performance rules with the greatest hits are placed at the top and those with fewer hits follow [2,47,48].This is expected to reduce the time taken for searching since the most commonly hit rules will be at the top. Therefore, most hits will be Protocol Storage Area Networks Copyright © 2021 MECS I.J.…”

Section: A Rules Priority Estimation and Sortingmentioning

confidence: 99%

“…Consolidation of storage into storage area network (SAN) rather than using separate physical infrastructure is a common approach meant to save on hardware and operational costs. However, this results into resource contention among the network users as they only get a fraction of the resource shared [1,2] such as disk cache and bandwidth. Contrastingly, the traditional free for all option is neither a solution as it may end up in significant efficiency degradation when the network is shared by multiple users [3,1,4].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

An Enhanced List Based Packet Classifier for Performance Isolation in Internet Protocol Storage Area Networks

Kithinji¹,

Mutua²,

wathi³

2021

IJITCS

View full text Add to dashboard Cite

Consolidation of storage into IP SANs (Internet protocol storage area network) has led to a combination of multiple workloads of varying demands and importance. To ensure that users get their Service level objective (SLO) a technique for isolating workloads is required. Solutions that exist include cache partitioning and throttling of workloads. However, all these techniques require workloads to be classified in order to be isolated. Previous works on performance isolation overlooked the classification process as a source of overhead in implementing performance isolation. However, it’s known that linear search based classifiers search linearly for rules that match packets in order to classify flows which results in delays among other problems especially when rules are many. This paper looks at the various limitation of list based classifiers. In addition, the paper proposes a technique that includes rule sorting, rule partitioning and building a tree rule firewall to reduce the cost of matching packets to rules during classification. Experiments were used to evaluate the proposed solution against the existing solutions and proved that the linear search based classification process could result in performance degradation if not optimized. The results of the experiments showed that the proposed solution when implemented would considerably reduce the time required for matching packets to their classes during classification as evident in the throughput and latency experienced.

show abstract

mentioning

confidence: 78%

Section: Performance Evaluationmentioning

confidence: 95%

Section: A Rules Priority Estimation and Sortingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Enhanced List Based Packet Classifier for Performance Isolation in Internet Protocol Storage Area Networks

Kithinji¹,

Mutua²,

wathi³

2021

IJITCS

View full text Add to dashboard Cite

show abstract

“…Previous performance evaluations involving hypervisors on the Banana Pi were done by Toumassian et al [10] and Danielsson et al [11]. Toumassian et al evaluated the performance overhead of hypervisors compared to a native Linux setup.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of a Hypervisor-Based Smart Controller for Industry 4.0 Functions in Manufacturing

Schade

Barton

Fleischer

et al. 2021

2021 IEEE 7th World Forum on Internet of Things (WF-IoT)

View full text Add to dashboard Cite

Niyama: Node scheduling for cloud workloads with resource isolation

Thiyyakat

Kalambur

Sitaram³

2022

Concurrency and Computation

View full text Add to dashboard Cite

Cloud providers place tasks from multiple applications on the same resource pool to improve the resource utilization of the infrastructure. The consequent resource contention has an undesirable effect on latency-sensitive tasks. In this article, we present Niyama-a resource isolation approach that uses a modified version of deadline scheduling to protect latency-sensitive tasks from CPU bandwidth contention.Conventionally, deadline scheduling has been used to schedule real-time tasks with well-defined deadlines. Therefore, it cannot be used directly when the deadlines are unspecified. In Niyama, we estimate deadlines in intervals and secure bandwidth required for the interval, thereby ensuring optimal job response times. We compare our approach with cgroups: Linux's default resource isolation mechanism used in containers today. Our experiments show that Niyama reduces the average delay in tasks by 3×-20× when compared to cgroups. Since Linux's deadline scheduling policy is work-conserving in nature, there is a small drop in the server-level CPU utilization when Niyama is used naively. We demonstrate how the use of core reservation and oversubscription in the inter-node scheduler can be used to offset this drop; our experiments show a 1.3×-2.24× decrease in delay in job response time over cgroups while achieving high CPU utilization.

show abstract

Testing Performance-Isolation in Multi-core Systems

Cited by 18 publications

References 9 publications

An Enhanced List Based Packet Classifier for Performance Isolation in Internet Protocol Storage Area Networks

An Enhanced List Based Packet Classifier for Performance Isolation in Internet Protocol Storage Area Networks

Evaluation of a Hypervisor-Based Smart Controller for Industry 4.0 Functions in Manufacturing

Niyama: Node scheduling for cloud workloads with resource isolation

Contact Info

Product

Resources

About