Ravi Manghirmalani

StEERING: A software-defined networking for inline service chaining

Network operators are faced with the challenge of deploying and managing middleboxes (also called inline services) such as firewalls within their broadband access, datacenter or enterprise networks. Due to the lack of available protocols to route traffic through middleboxes, operators still rely on error-prone and complex low-level configurations to coerce traffic through the desired set of middleboxes. Built upon the recent software-defined networking (SDN) architecture and OpenFlow protocol, this paper proposes StEERING, short for SDN inlinE sERvices and forwardlNG. It is a scalable framework for dynamically routing traffic through any sequence of middleboxes. With simple centralized configuration, StEERING can explicitly steer different types of flows through the desired set of middleboxes, scaling at the level of per-subscriber and per-application policies. With its capability to support flexible routing, we further propose an algorithm to select the best locations for placing services, such that the performance is optimized. Overall, StEERING allows network operators to monetize their middlebox deployment in new ways by allowing subscribers flexibly to select available network services.

NetRevert: rollback recovery in SDN

Recognizing the inevitability of human error and hardware failures, which can significantly hamper the computer system performance, the capability of checkpointing and rollback recovery has been proposed and widely used in servers and distributed systems. These facilities allow fast recovery from failures, software bugs, and misconfigurations. Though promising, they are rarely, if ever, provided for networking systems, thus the network operators still rely on ad-hoc and error-prone processes to recover from errors. In this work, we propose a checkpoint and rollback-recovery system for Software-Define Networking (SDN) systems, exploiting SDNs simple abstraction, network-wide view, and direct control properties. We propose an approach to efficiently identify the network-wide consistent states for rollback while preserving correctness.

A compressive method for maintaining forwarding states in SDN controller

Many controller applications require querying existing flow entries in the switches for improving resilience and optimizing resource utilization. These applications can benefit if the controller maintains a copy of the forwarding tables in its memory. However, a naive approach of simply keeping all the tables may encounter scalability challenges. In this work, we identify the redundancy across the flow tables of all switches in the network. We then propose a model-based compression method to efficiently store the network-wide forwarding states. Our evaluation results on a variety of topologies show up to 98% reduction in size.

Elastic network monitoring with virtual probes

The next generation monitoring solutions should evolve as fast as the applications they will support both in terms of functionality (e.g. new protocols, monitoring function, etc.) as well as in terms of scale and performance. We believe the field of network monitoring is yet to experience the same revolutionary transformation as software-defined networking (SDN) and network function visualization (NFV) brought to the telecom world. To this end we present a flexible network monitoring application we use to demonstrate the potential of next generation network monitoring with advanced probe functionality and virtualization. Our demonstration shows a new OVS-based (Open vSwitch) vProbe capable of adding and processing monitoring metadata in IETFs recently proposed network service header (NSH) for service chaining [1], [2]. The monitoring metadata in this work is used for troubleshooting performance issues (i.e. excessive latency) along service function paths (SFPs). Benefits of our approach are the flexibility it offers to create new monitoring solutions as well as the virtualization of monitoring resources in order to benefit from the same economies of scale and elasticity features of NFV. This demonstration is built as part of our proposed monitoring framework presented in [3] which consists of a monitoring platform which, on the northbound, offers a full control as well as a more limiting intent based interface. It can spin new vProbes when and where needed. The data plane used in this demonstration consists of OpenFlow switches (OVS) extended with the NSH header [4]. This work can be described as a troubleshooting use case customized for service chaining. We demonstrate a video traffic performance degradation scenario and pin down where the issue is for a given flow on a service chain.

A monitoring framework at layer4–7 granularity using network service headers

Traffic measurement and network monitoring needs have always daunted operators and recent trends of virtualization and programmable networks put a higher stress on this topic. The reality is that todays monitoring toolsets are viable for legacy network operations and traffic. With recent industry trends such as network function virtualization (NFV), software-defined networking (SDN) and everything as a service (XaaS), will follow higher network dynamicity and traffic change patterns. Network monitoring and measurement schemes need to evolve to accommodate these upcoming trends. The research community has been very active in past years in proposing new or improved mechanisms for various monitoring tasks. In this work however we leverage another recent development in the industry for network function chaining and propose to use the newly defined tunneling headers by IETFs service function chaining (SFC) working group to achieve specific and novel monitoring tasks not possible with existing methods. This papers contributions are two folds: first it proposes a framework and an architecture able to perform unified monitoring across network, cloud, physical and virtual boundaries. Second, based on layer4-7 information it proposes novel monitoring mechanisms applicable to NFV traffic. The latter distinguishes itself with respect to existing proposals by its flexibility and potential for widespread deployment as it piggybacks on IETFs SFC standardization work. This papers contributions are the system architecture, the monitoring metadata encodings with their respective use cases, as well performance results for monitoring overhead measured on a testbed implementation of this proposal.