Network Function Virtualization (NFV) and Service Function Chaining (SFC) offer flexible end-to-end services that deploy virtual network functions in clouds of competing providers. Orchestration of virtual network fun...
详细信息
ISBN:
(纸本)9781538680889
Network Function Virtualization (NFV) and Service Function Chaining (SFC) offer flexible end-to-end services that deploy virtual network functions in clouds of competing providers. Orchestration of virtual network functions occurs in a distributed and trustless environment that must tolerate byzantine failures and collusion attacks. This paper proposes BSec-NFVO, a blockchain-based system that secures orchestration operations in virtualized networks, ensuring auditability, non-repudiation and integrity. We propose an NFV-tailored blockchain and a transaction model. BSec-NFVO provides a modular architecture to secure orchestration in a simple and agile way. We develop a prototype of BSec-NFVO for the Open Platform for Network Function Virtualization (OPNFV) with an adaptation of the normal-case of a collusion-resistant consensus protocol. The results show BSec-NFVO incurs low overhead to the cloud orchestrator and presents stable performance as the number of consensus participants increases.
Drone networks offer rapid network deployment to areas that can pose access difficulty. This paper investigates the deployment of multi-hop drone-based unmanned aerial vehicles networks with a focus on the self-organi...
详细信息
ISBN:
(纸本)9781538680889
Drone networks offer rapid network deployment to areas that can pose access difficulty. This paper investigates the deployment of multi-hop drone-based unmanned aerial vehicles networks with a focus on the self-organization aspect. When rescue drones carry out their rescue operations which may fly faraway from the gateway, relay drones are autonomously deployed to maintain connectivity. We study the multiple dedicated connections where each rescue drone is connected to the gateway via dedicated relay drones. We show that this approach lacks sharing of relay drones and thus requires more relay drones. We then propose a centralized greedy algorithm and a distributed solution to significantly reduce the number of relay drones. We show that while the distributed self-organized drones (DSOD) solution requires a slightly higher number of relay drones than the greedy algorithm, it eliminates the need for global message exchange which makes it attractive for practical use.
We present a new greedy b-MATCHING algorithm suitable for running on a GPU. Our algorithm differs from previous efforts at designing parallel algorithms for this problem in that it does not use software locks and that...
详细信息
ISBN:
(纸本)9781538655559
We present a new greedy b-MATCHING algorithm suitable for running on a GPU. Our algorithm differs from previous efforts at designing parallel algorithms for this problem in that it does not use software locks and that it also exploits substantially more of the available concurrency. We achieve this by allowing the same vertex to concurrently match with several other vertices and also by letting multiple vertices simultaneously match with the same target vertex. We have compared our algorithm using a Pascal P100 GPU with the previous best shared memory algorithm for this problem both when running on a 16 core Xeon E5 and on a Xeon Phi. On average our algorithm outperforms the Xeon E5 by a factor of 4.6 and the Xeon Phi by a factor of 2.3. We also show that our algorithm using an NVIDIA DGX-1 multi-GPU system is highly competitive compared to a distributed memory implementation running on one of the top ten computers from the current TOP500 list.
暂无评论