Chin Guok

Intra and Interdomain Circuit Provisioning Using the OSCARS Reservation System

With the advent of service sensitive applications such as remote controlled experiments, time constrained massive data transfers, and video-conferencing, it has become apparent that there is a need for the setup of dynamically provisioned, quality of service enabled virtual circuits. The ESnet on-demand secure circuits and advance reservation system (OSCARS) is a prototype service enabling advance reservation of guaranteed bandwidth secure virtual circuits. OSCARS operates within the energy sciences network (ESnet), and has provisions for interoperation with other network domains. ESnet is a high-speed network serving thousands of Department of Energy scientists and collaborators worldwide. OSCARS utilizes the Web services model and standards to implement communication with the system and between domains, and for authentication, authorization, and auditing (AAA). The management and operation of end-to-end virtual circuits within the network is done at the layer 3 network level. Multi-protocol label switching (MPLS) and the resource reservation protocol (RSVP) are used to create the virtual circuits or label switched paths (LSPs). quality of service (QoS) is used to provide bandwidth guarantees. This paper describes our experience in implementing OSCARS, collaborations with other bandwidth-reservation projects (including interdomain testing) and future work to be done.

A User Driven Dynamic Circuit Network Implementation

The requirements for network predictability are becoming increasingly critical to the DOE science community where resources are widely distributed and collaborations are world-wide. To accommodate these emerging requirements, the energy sciences network has established a science data network to provide user driven guaranteed bandwidth allocations. In this paper we outline the design, implementation, and secure coordinated use of such a network, as well as some lessons learned.

Software-Defined Networking for Big-Data Science - Architectural Models from Campus to the WAN

University campuses, Supercomputer centers and R&E networks are challenged to architect, build and support IT infrastructure to deal effectively with the data deluge facing most science disciplines. Hybrid network architecture, multi-domain bandwidth reservations, performance monitoring and GLIF Open Lightpath Exchanges (GOLE) are examples of network architectures that have been proposed, championed and implemented successfully to meet the needs of science. Most recently, Science DMZ, a campus design pattern that bypasses traditional performance hotspots in typical campus network implementation, has been gaining momentum. In this paper and corresponding demonstration, we build upon the SC11 SCinet Research Sandbox demonstrator with Software-Defined networking to explore new architectural approaches. A virtual switch network abstraction is explored, that when combined with software-defined networking concepts provides the science users a simple, adaptable network framework to meet their upcoming application requirements.

Advance reservation frameworks in hybrid IP-WDM networks

New e-Science and grid applications require the coordination of geographically distributed scientific instruments along with data and computing resources. Due to the quality of service requirements of these applications, these distributed resources can be connected by a wavelength-routed optical network, allowing each application to get dedicated bandwidth. These networks are referred to as LambdaGrids. One important service provided in these networks is advance reservation. Applications need to coordinate the use of both grid resources and the network. Advance reservation allows these applications to reserve bandwidth in advance to guarantee availability. In this article, we discuss different networks and frameworks that support advance reservation of bandwidth. We discuss the general architecture of each network and the type of advance reservation services supported.

Hybrid networks: lessons learned and future challenges based on ESnet4 experience

ESnet, the Energy Sciences Network, has the mission of providing the network infrastructure to the U.S. Department of Energys Office of Science programs and facilities, which depend on large collaborations and large-scale data sharing, enabling them to accomplish their science. ESnet4 - a hybrid IP and dynamic circuit network designed in 2006 and completed in 2008 - has managed to effectively satisfy the networking needs of the science community, easily handling dramatic growth in traffic requirements: around 80 percent growth year over year and 300 percent growth with the Large Hadron Collider (LHC) coming online. In this article, we examine the benefits and limitations of the current hybrid architecture based on actual production experience; discuss open research problems; and predict factors that will drive the evolution of hybrid networks, including advances in network technology, new computer architectures, and the onset of large-scale distributed computing.

Open transport switch: a software defined networking architecture for transport networks

There have been a lot of proposals to unify the control and management of packet and circuit networks but none have been deployed widely. In this paper, we propose a simple programmable architecture that abstracts a core transport node into a programmable virtual switch, that meshes well with the software-defined network paradigm while leveraging the OpenFlow protocol for control. A demonstration use-case of an OpenFlow-enabled optical virtual switch implementation managing a small optical transport network for big-data applications is described. With appropriate extensions to OpenFlow, we discuss how the programmability and flexibility SDN brings to packet-optical backbone networks will be substantial in solving some of the complex multi-vendor, multi-layer, multi-domain issues service providers face today.

Control Plane Architecture and Design Considerations for Multi-Service, Multi-Layer, Multi-Domain Hybrid Networks

In this paper we discuss key architecture and design considerations associated with the development of a control plane capable of dynamic provisioning in this heterogeneous multi-domain, multi-layer, multi-service hybrid network environment . We present a framework for addressing the heterogeneous nature of the hybrid networks via the development of a flexible set of mechanisms which address the key control plane functions of routing, path computation and signaling. An interoperable set of constructs are proposed based on GMPLS and Web service for seamless provisioning across heterogeneous data and control planes. This paper also includes a discussion of our recent design and implementation efforts to instantiate these concepts on ESnet, USN, and the Internet2 Networks.

Improving the bulk data transfer experience

Scientific computations and collaborations increasingly rely on the network to provide high-speed data transfer, dissemination of results, access to instruments, support for computational steering, etc. The Energy Sciences Network is establishing a science data network to provide user driven bandwidth allocation. In a shared network environment, some reservations may not be granted due to the lack of available bandwidth on any single path. In many cases, the available bandwidth across multiple paths would be sufficient to grant the reservation. In this paper we investigate how to utilise the available bandwidth across multiple paths in the case of bulk data transfer.

Bursting Data between Data Centers: Case for Transport SDN

Public and Private Enterprise clouds are changing the nature of WAN data center interconnects. Data center WAN interconnects today are pre-allocated, static optical trunks of high capacity. These optical pipes carry aggregated packet traffic originating from within the data centers while routing decisions are made by devices at the data center edges. In this paper, we propose a software-defined networking enabled optical transport architecture (Transport SDN) that meshes seamlessly with the deployment of SDN within the Data Centers. The proposed programmable architecture abstracts a core transport node into a programmable virtual switch that leverages the OpenFlow protocol for control. A demonstration use-case of an OpenFlow-enabled optical virtual switch managing a small optical transport network for a big-data application is described. With appropriate extensions to OpenFlow, we discuss how the programmability and flexibility SDN brings to packet-optical data center interconnect will be substantial in solving some of the complex multi-vendor, multi-layer, multi-domain issues that hybrid cloud providers face.

Traffic Optimization in Multi-layered WANs Using SDN

Wide area networks (WAN) forward traffic through a mix of packet and optical data planes, composed by a variety of devices from different vendors. Multiple forwarding technologies and encapsulation methods are used for each data plane (e.g. IP, MPLS, ATM, SONET, Wavelength Switching). Despite standards defined, the control planes of these devices are usually not interoperable, and different technologies are used to manage each forwarding segment independently (e.g. Open Flow, TL-1, GMPLS). The result is lack of coordination between layers and inefficient resource usage. In this paper we discuss the design and implementation of a system that uses unmodified Open Flow to optimize network utilization across layers, enabling practical bandwidth virtualization. We discuss strategies for scalable traffic monitoring and to minimize losses on route updates across layers. A prototype of the system was built using a traditional circuit reservation application and an unmodified SDN controller, and its evaluation was performed on a multi-vendor test bed.