Eylem Seç
Toward Efficient Traffic Control and Monitoring in Software-defined Datacenter Networks
Başlık:
Toward Efficient Traffic Control and Monitoring in Software-defined Datacenter Networks
Yazar:
Aljaedi, Amer Omar, author.
ISBN:
9780438140370
Yazar Ek Girişi:
Fiziksel Tanımlama:
1 electronic resource (201 pages)
Genel Not:
Source: Dissertation Abstracts International, Volume: 79-11(E), Section: B.
Advisors: C. Edward Chow Committee members: Francisco Torres-Reyes; Jonathan Ventura; Chuan Yue; Yanyan Zhuang.
Özet:
The multi-tenant datacenter is one of the most challenging networking environments due to its scale and dynamic network conditions. The increasing demand for computing resources under the pay-as-you-go business model has driven the growth of large-scale datacenters, where a tremendous number of tenants subscribe to cloud services on a daily basis. Thus, the underlying datacenter network has to handle a large number of configuration updates. Also, network-wide state monitoring has become indispensable for efficient resource management in datacenters, as tenants' applications are often deployed over distributed computing nodes to serve a sheer amount of user requests simultaneously. Therefore, datacenter operators monitor the network state constantly for various reasons, such as traffic engineering, dynamic rate limiting, VM bandwidth profiling, and network debugging. Such large-scale shared networks require cost-effective traffic control and monitoring techniques to maintain efficient isolation of the tenants' traffic while serving their demands.
In this dissertation, we present the design and implementation of a scalable network virtualization technique that allows tenants to effectively migrate their workloads, including configurations, to the cloud services. Our proposed network virtualization technique outperforms the L2-in-L3 tunneling protocols that are widely used in production datacenters. As our research work utilizes Software-Defined Networking (SDN), which is increasingly adopted in datacenters, we provide an innovative technique, named PathSeer, for tracing packet trajectories on SDN. PathSeer allows SDN developers and operators to observe the network forwarding behavior and localize failures without interfering with the operations of other network protocols. PathSeer can trace complicated packet trajectories in datacenter networks that cannot be traced by the other state-of-the-art techniques. Moreover, for tackling traffic monitoring challenges in large-scale SDN-enabled datacenters, we present a scalable SDN-based monitoring system called OTMEN for datacenters. The OTMEN system offloads monitoring tasks from the SDN controller and distributes the monitoring workload to edge nodes of the network. It significantly reduces the overhead on the controller and demonstrates improvements over existing monitoring solutions. APIs are provided to enable various SDN applications in datacenters to monitor network traffic.
Notlar:
School code: 0892
Konu Başlığı:
Mevcut:*
Yer Numarası | Demirbaş Numarası | Shelf Location | Lokasyon / Statüsü / İade Tarihi |
---|---|---|---|
XX(695962.1) | 695962-1001 | Proquest E-Tez Koleksiyonu | Arıyor... |
On Order
Liste seç
Bunu varsayılan liste yap.
Öğeler başarıyla eklendi
Öğeler eklenirken hata oldu. Lütfen tekrar deneyiniz.
:
Select An Item
Data usage warning: You will receive one text message for each title you selected.
Standard text messaging rates apply.