Security aware dynamic scheduling algorithm (SADSA) for real-time applications on grid

Security is a major concern of modern real-time applications, besides requiring stringent latency bound. However, encryption algorithms are computation intensive task which impacts the timeliness of the real-time applications. Therefore, there exists a trade-off between the desired level of security...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Cluster computing Ročník 23; číslo 2; s. 989 - 1005
Hlavní autoři: Singh, Surendra, Batabyal, Suvadip, Tripathi, Sachin
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York Springer US 01.06.2020
Springer Nature B.V
Témata:
Algorithms
Completion time
Computer Communication Networks
Computer Science
Deadlines
Embedded systems
High performance computing
Operating Systems
Optimization
Processor Architectures
Quality of service
Real time
Satellite communications
Scheduling
Security
Security services
Tradeoffs
Grid computing
Real-time applications
Security
Scheduling algorithm
ISSN:1386-7857, 1573-7543
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Security is a major concern of modern real-time applications, besides requiring stringent latency bound. However, encryption algorithms are computation intensive task which impacts the timeliness of the real-time applications. Therefore, there exists a trade-off between the desired level of security and the service guarantee. In this paper, we propose a security-aware dynamic scheduling algorithm (SADSA) using a grid of computational elements (CEs) which performs this trade-off and tries to maximize the instantaneous average security level of the packets besides providing a guaranteed service. As packets arrive, we first assign them to the CEs based on the utilization value of a CE, which is the ratio of completion time and a deadline of the last packet in a CE. The security level of all the packets is then dynamically adjusted to meet the minimum required security level while maximizing the average security level of all the packets in that CE. We first show that the proposed assignment algorithm is NP-hard, is 2-competitive to the optimal solution, and that the proposed algorithm provides a sub-optimal solution. Further, using extensive simulation, we show that the proposed SADSA algorithm performs better in terms of guarantee ratio, average security level and overall performance compared to the existing algorithms.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-019-02970-w