Sarvesh Bidkar

Living on the edge: Monitoring network flows at the edge in cloud data centers

Scalable network wide flow monitoring has remained a distant dream because of the strain it puts on network router resources. Recent proposals have advocated the use of coordinated sampling or host based flow monitoring to enable a scalable network wide monitoring service. As most hosts in data centers get virtualized with the emergence of the cloud, the hypervisor on a virtualized host adds another network layer in the form of a vSwitch (virtual switch). The vSwitch now forms the new edge of the network. In this paper, we explore the implications of enabling network wide flow monitoring inside virtual switches in the hosts. Monitoring of network flows inside the server vSwitch can enable scalability due to its distributed nature. However, assumptions that held true for flow monitoring inside a physical switch need to be revisited since vSwitches are usually not limited by the same level of resource constraints that exist for physical switches and routers. On the other hand, vSwitches do not implement flow monitoring in hardware, as it is done in some physical switches. We present the design and implementation of EMC2 - a scalable network wide monitoring service for cloud data centers. We also conduct an extensive evaluation of various switch based flow monitoring techniques and share our findings. Our results indicate that while layer-3 flow monitoring protocols such as NetFlow can give a very good network coverage without using too many resources, protocols that sample packet headers (such as sFlow) need to be carefully configured. A badly configured sFlow vSwitch can degrade application network throughput by up to 17% and can also choke the management network by generating monitoring data at a very high rate.

Scalable segment routing—a new paradigm for efficient service provider networking using carrier ethernet advances

Segment routing has recently been proposed in the IETF toward making IP/MPLS networks service oriented while simplifying network operations. Segment routing computes paths at the source node using node identifiers and adjacency identifiers conjoined together to create a source-routed path. We propose a scalable transport paradigm as an enabler toward implementing segment routing in provider networks. We propose omnipresent Ethernet, our modification of carrier Ethernet (which is based on source-routed, binary-routed labels embedded in an Ethernet frame), to implement segment routing. We identify some of the scalability issues of the segment routing proposal in the context of source-routing overhead. To absolve scalability issues of segment routing, two routing schemes that implement multidomain source-routing techniques are proposed. The hierarchical segment-routing (H-SR) scheme is proposed, which deploys a limited number of special nodes called swap nodes that are capable of label swapping to implement routing. The swap-node placement problem is formulated as an integer linear program to minimize the total routing distance within a network. Three heuristic techniques for swap-node selection based on centrality paradigms are presented. The H-SR scheme is further improved by a proposed multisegment-routing (M-SR) scheme that assumes all nodes in the network to be capable of label swapping. The H-SR and M-SR schemes are shown to significantly enhance the scalability of segment routing. A test-bed is built using carrier Ethernet hardware to support segment routing that validates the implementation of the proposed schemes. A simulation model to evaluate the schemes from a scalability perspective and their associated trade-offs is presented.

A scalable framework for segment routing in service provider networks: The Omnipresent Ethernet approach

Segment routing has recently been proposed in the IETF towards making IP/MPLS networks service-oriented and efficient. Segment routing involves identifying paths at the source node using adjacency identifiers conjoined together to create a source-routed path. To this end, we propose a transport paradigm that will act as an enabler towards implementing segment routing in service provider networks. Specifically, we use Carrier Ethernet that is gaining acceptance as an IP/MPLS carrier technology. We propose the use of our modification of Carrier Ethernet called Omnipresent Ethernet (based on source routed binary labels embedded in an Ethernet frame) towards implementing segment routing. We evaluate segment routing for large service provider networks and understand scalability limitations of source routing. To absolve the scalability issues, a hierarchical segment routing (H-SR) scheme that uses special nodes called Swap-Nodes is proposed. Three techniques for swap-node selection based on centrality paradigms that facilitate and enhance scalability of segment routing are put forth. A test-bed is built to validate segment routing along with a simulation model to evaluate the proposed hierarchical segment routing scheme.

Field trial of a software defined network (SDN) using carrier Ethernet and segment routing in a tier-1 provider

Software Defined Networking (SDN) has brought a paradigmatic shift in the networking industry and has led to significant benefits in the data-center and enterprise network domains. The service provider networks that form the largest segment of networking industry, are now evaluating SDN technologies for adoption. In this paper, we present a SDN framework for service provider networks and report the first field trial of SDN in a tier-1 service provider domain. The proposed SDN framework is built using Carrier Ethernet and augmented with recently proposed Segment Routing paradigm manifested through Software Defined-Carrier Ethernet Switch Routers (SD-CESRs). Carrier Ethernet on account of its distinct, programmable control plane and Segment Routing through its source routing capabilities facilitates SDN implementation. The SD-CESRs are deployed in a tier-1 service provider network in the metropolis of Mumbai. The SDN framework is extended through specific APIs to enhance revenue bearing services portfolio of the service provider and performance results from the field are shown to validate the benefits of SDN adoption.

On the Design, Implementation, Analysis, and Prototyping of a 1- μs, Energy-Efficient, Carrier-Class Optical-Ethernet Switch Router

The integration of layer-2 carrier-class packet technologies with optical transport network is termed as packet-optical integration and is being deployed by service providers for migration from legacy SONET/SDH systems. We present a state-of-the-art carrier-class switch router that facilitates packet-optical integration, thereby achieving best of both the optical and packet worlds. The premise of this switch router is the use of carrier ethernet technology as a packet enabler for achieving statistical multiplexing at fine granularities, while maintaining rich operations, administration, maintenance, and provisioning features. To this end, we proposed the omnipresent ethernet concept that uses binary routing and source routing to support: 1) layer 2 and layer 3 switching and routing, 2) low latency of the order of 1-5 μs even for layer 3 processing, and 3) low-energy consumption. The omnipresent ethernet framework leads to a software defined networking solution, whereby a centralized controller admits services and configures nodes based on homogenous networking parameters. In this paper, we report a commercial implementation of a packet-optical network demonstrated by our designed, fabricated, developed carrier ethernet switch router (CESR). We discuss the architectural considerations, design, and implementation of both the hardware and the control software. A switch architecture achieving 1-μs port-to-port delay across layers 1, 2, and 2.5 based on an opportunistic principle of virtual output queuing is showcased. The novelty is a scalable 96-Gbps cross connect fabric implemented in a field programmable gate array using a two-stage buffer system. A multistate software defined control plane is also reported. An exact analysis of the switch architecture using a combinatorial G/G/1 model is developed. An energy audit of the CESR is showcased. A test bed in the lab replicating a field deployment is presented. An exhaustive set of test cases are developed to test our designed CESR. Results are shown for latency, throughput, service support, frame-size stability, power, and bit error rate, thus validating our design.

Circuit performance in a packet network: Demonstrating integrated Carrier Ethernet Switch Router (CESR) + Optical Transport Network (OTN)

Circuit switched systems in the metropolitan and core networks are being steadily replaced by the packet optical integration brought about by the steady advances in Carrier Ethernet, Optical Transport Network (OTN) and Reconfigurable Optical Add-Drop Multiplexer (ROADM). We demonstrate a state-of-the-art Carrier Ethernet Switch Router (CESR) with integrated OTN function as a mechanism to better facilitate packet-optical integration leading to service and layer 2.5 routing support. The design, development of this CESR+OTN platform is described. The test-bed showcases an integrated high-speed optical network configured as an inter-connected ring topology of 16 nodes, multi-degree cross-connects and demonstrates next generation application services used for cloud computing and mobile backhaul. Performance results indicate that the packet optical technology proposed by us and showcased in a live test-bed compare similar to (or in some cases better than) a circuit switched network.

Design of a shared memory Carrier Ethernet switch compliant to Provider Backbone Bridging-Traffic Engineering (IEEE802.1Qay)

Carrier Ethernet is emerging as a new transport paradigm across metropolitan and core networks. Provider Backbone Bridging-Traffic Engineering or PBB-TE was standardized in the IEEE as 802.1Qay as a mechanism to provide a dedicated transport service at the Ethernet layer. This paper discusses implementation of the PBB-TE standard using shared memory switch architecture, though the same architecture argument can be extended to implement MPLS-TP (the other manifestation of Carrier Ethernet). While shared memory switch architectures have been well investigated, we provide to the best of our knowledge the first carrier-class aggregation switch implemented in a single Field Programmable Gate Array (FPGA). This low-cost implementation paves the way for advances in Carrier Ethernet technologies to be made available to the access part of the network using rapid prototyping and commercial off the shelf components. The switch architecture supports multiple QoS levels and implements circuit emulation to transport traditional circuit services over a packet backbone. A rigorous simulation study validates our effort.

Demonstrating OpenFlow over a Carrier Ethernet Switch Router (CESR) - a services perspective

The OpenFlow concept was recently proposed primarily to facilitate policy routing in large enterprise/service provider networks. Carrier Ethernet technologies such as MPLS-TP and PBB-TE are currently considered to be the forerunners in such transport networks. Both these technologies specify the need for a managed control plane. Carrier Ethernet Switch Routers (CESRs) are being proposed for transport functionality in service provider networks. The future of the transport network is to provide for cloud-like services whereby policy routing would be a key enabler technology. The OpenFlow specification is quite in accordance with the implementation of a centralized, unified control plane that is mandatory for Carrier Ethernet. In this paper, we demonstrate the OpenFlow paradigm implementation over CESRs to support next generation cloud applications and services. We showcase architecture, modifications and protocol performance with specific focus on OpenFlow in operator networks. The main novelty of this work is in transforming the OpenFlow controller to co-exist with the network management system of a CESR thus facilitating the CESR to control and interface with the OpenFlow protocol. We adapt the OpenFlow paradigm to an implemented CESR. Hardware results for service performance, network utilization, latency behavior across the network and protection switching are shown. We conclude for the first time, the feasibility of adapting the OpenFlow standard to service provider equipment.

Using MPLS-TP for Data-Center Interconnection

Data center interconnectivity is particularly very important and essential for emerging applications such as cloud computing and financial trading. Current data center architectures are built using Ethernet switches or IP routers - both with significant cost and performance deficiencies. We propose for the first time, extending MPLS-TP into the data-center. To this end, a new look-up protocol for MPLS-TP is proposed. Autonomic communication within the data center is possible using our look-up protocol that enables fast creation and deletion of LSPs. The MPLS-TP based data-center that is architected in this paper leads to performance betterments over both IP and Ethernet. To this end, a comprehensive simulations model is also presented. Operations within the data-center using MPLS-TP are also extended to inter-data-center operations using LSP setup across a core network. OAM and performance issues are investigated.

Demonstrating a software defined network (SDN) using Carrier Ethernet Switch Routers in a provider network

We demonstrate SDN in a service provider network using adapted Carrier Ethernet Switch Routers developed based on principles of segment routing. SDN based service composition and delivery is showcased and validated.