Internet Systems Lab (ISL)
Abstractions for Enterprise Network ManagementThe use of abstractions to simplify network design and configuration has been a long cherished vision in the networking community. We posit that achieving more fundamental break-throughs requires abstractions that go beyond merely modeling the underlying protocols and mechanisms. We investigate a new class of abstractions that are: (i) task-driven, i.e., capture the intended performance, security, manageability, or resilience of a network design; and (ii) network-wide, i.e., capture the requirements of the network as a whole rather than of individual devices. Our focus is on the management of enterprise networks. Despite their critical importance, and their striking differences and diversity compared to carrier networks, enterprise networks have been largely unexplored by networking researchers. We envision a three-pronged research process that involves: A distinguishing feature of this research is its "white-box'' methodology to studying network designs. Rather than infer network characteristics with limited support from network operators as is common practice today, we will capitalize on our extensive ties with real network operators, and conduct studies using data such as router configuration files obtained with their support, and iterative interactions with them. We are currently designing abstractions in two areas that are critically important, and widely prevalent in enterprises. (i) use of virtualization, in particular VLANs, to simplify management goals; and (ii) network evolution through planned maintenance. 1. Scientific Study of Network Evolution Configuring network devices (e.g. routers and switches) today is difficult due to increases in network scale, complexity of network protocol designs, and the diversity of devices from different vendors (e.g. Cisco and Juniper). One key challenge network operators face is changing configurations of an operational network to accommodate new needs. As a result, to increase the manageability and reliability of configuration changes, there is a strong need to discover and design high-level abstractions network operators could leverage to easily and accurately specify intended changes of various complexity for the network. Publications
Presentations
Collaborators
Students
Funding
2. Top-Down Enterprise Network Design Despite their critical importance, and their striking differences compared to carrier networks, enterprise networks have been largely unexplored by the research community. In this project, we formulate key tasks facing operators of today's enterprises through abstractions that are task-driven and network-wide. We model the objectives of a task as a set of optimization problems, and derive configurations from solutions to these problems. This approach ensures configurations conform to the operator intent, yet it still allows customization of the configurations by modeling design constraints and strategies as inputs for selecting particular optimization problems to solve. To demonstrate our ideas, we consider two critical areas in enterprise network management that have received little attention to date: VLAN design and reachability control. Our initial success both establishes the feasibility of top-down design in the two areas we studied and gives us hope that our approach will apply to a broad class of problems. More specifically, we aim to develop systems that can generate box level configuration from high-level design requirements, and a pre-selected set of protocols and mechanisms, and it investigates "semantic auditing" techniques for formally verifying that a network's existing box-level configuration produces the intended network behavior. Publications
Presentations
Collaborators
Students
FundingNSF Award Numbers: CNS-0721488, CNS-0520210, and CNS-0721574 |
©Copyright 2007 - Internet Systems Lab - Purdue University