The baseline VM placement algorithms only consider serv er hardware resources

The baseline vm placement algorithms only consider

This preview shows page 3 - 5 out of 6 pages.

The baseline VM placement algorithms only consider serv- er hardware resources, such as Least Used Host placement algorithm and Least Busy Host placement algorithm [7]. The rapid growth of distributed analytics (e.g., MapReduce) and time-sensitive applications raise a big concern about network performances. Hence, network-aware placement algorithms are proposed. E2 [22], a VM-based NFV framework, relies on them to minimize intra-server traffic when mapping each VNF instance to a particular server, because inter-server communication incurs less latency and bandwidth between servers. E2 [22] targets Central Offices (COs), where VNFs are clustered and interconnected by a Layer-2 network. COs are very common in the backhaul of mobile networks to provide a wide variety of services. The initial NF placement involves four steps. The first step is to merge individual service chains into a single policy graph (pGraph), in order to identify the relationship between all NFs. The second step is sizing. E2 determines the number of instances for each NF in the pGraph given the estimate of the load on a NF and the per-instance capacity. The workload of a NF will be evenly distributed among the its VNF instances. The third step is converting the pGraph to an iGraph, or the ”instance graph”. In the iGraph, each node represents an instance of NF and the edge weight captures the traffic demand between two VNF instances. The final step is the actual instance placement with the objective to minimize inter-server traffic. The optimization problem is modeled as graph partition, which E2 solves by an iterative local search algorithm based Kernighan-Lin heuristic. Statos [10] is another network-aware NFV platform being able to correctly forward traffic in face of mangling NFs (we have discussed them in Sec. III-A1). Statos exhaustively
Image of page 3

Subscribe to view the full document.

enumerates the possible downstream pathes for each mangling NF in the policy graph. The mangling NF are cloned as many times as the number of downstream pathes, with each clone being responsible for one downstream path. By simply cloning the mangling NFs when there is ambiguity, Statos ensures the correctness of service chaining even though this transformation potentially increases the number of needed VNF instances. Finally, Statos employs network-aware VM placement algo- rithms (such as TMVPP [20]) to avoid congestion. 2) Path Tightly Controlled Placement: In existing VM placement algorithms, it is assumed that the routing path between each VM-pair is determined merely by the physical locations of VMs. Even though sometimes traffic engineering is allowed, the network operators still cannot explicitly con- trol the routing path. With the tremendous convenience and flexibility bought by SDN, some NFV frameworks control the routing pathes between NF pairs, enabling joint placement and routing optimization for better performance.
Image of page 4
Image of page 5
  • Spring '16

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern

Ask Expert Tutors You can ask 0 bonus questions You can ask 0 questions (0 expire soon) You can ask 0 questions (will expire )
Answers in as fast as 15 minutes