Search for: All records

To interconnect research facilities across wide geographic areas, network operators deploy science networks, also referred to as Research and Education (R&E) networks. These networks allow experimenters to establish dedicated circuits between research facilities for transferring large amounts of data, by using advanced reservation systems. Intercontinental dedicated circuits typically require coordination between multiple administrative domains, which need to reach an agreement on a suitable advance reservation. The success rate of finding an advance reservation decreases as the number of participant domains increases for traditional systems because the circuit is composed over a single path. To improve provisioning of multi-domain advance reservations, we propose an architecture for end-to-end service orchestration in multi-domain science networks that leverages software-defined exchanges (SDX) for providing multi-path, multi-domain advance reservations. We have implemented an orchestrator for multi-path, multi-domain advance reservations and an SDX to support these services. Our orchestration architecture enables multi-path, multi-domain advance reservations and improves the reservation success rate from 50% in single path systems to 99% when four path are available. 

To interconnect research facilities across wide geographic areas, network operators deploy science networks, also referred to as Research and Education (R&E) networks. These networks allow experimenters to establish dedicated network connections between research facilities for transferring large amounts of data. Recently, R&E networks have started using Software-Defined Networking (SDN) and Software Defined Exchanges (SDX) for deploying these connections. AtlanticWave/SDX is a response to the growing demand to support end-to-end network services spanning multiple SDN domains. However, requesting these services is a challenging task for domain-expert scientists, because the interfaces of the R&E networks have been developed by network operators for network operators. In this paper, we propose interfaces that allow domain expert scientists to reserve resources of the scientific network using abstractions that focus on their data transfer needs for scientific workflow management. Recent trends in the networking field pursue better interfaces for requesting network services (e.g., intent-based networking). Although intents are sufficient for the needs of network operations, they are not abstract enough in most cases to be used by domain-expert scientists. This is an issue we are addressing in the AtlanticWave/SDX design: network operators and domain-expert scientists will have their own interfaces focusing on their specific needs. 

Poster Abstract: To interconnect research facilities across wide geographic areas, network operators deploy science networks, also referred to as Research and Education (R&E) networks. These networks allow experimenters to establish dedicated network connections between research facilities for transferring large amounts of data. Recently, R&E networks have started using Software-Defined Networking (SDN) and Software Defined Exchanges (SDX) for deploying these connections. AtlanticWave/SDX is a response to the growing demand to support end-to-end network services spanning multiple SDN domains. However, requesting these services is a challenging task for domain-expert scientists, because the interfaces of the R&E networks have been developed by network operators for network operators. In this paper, we propose interfaces that allow domain expert scientists to reserve resources of the scientific network using abstractions that focus on their data transfer needs for scientific workflow management. Recent trends in the networking field pursue better interfaces for requesting network services (e.g., intent-based networking). Although intents are sufficient for the needs of network operations, they are not abstract enough in most cases to be used by domain-expert scientists. This is an issue we are addressing in the AtlanticWave/SDX design: network operators and domain expert scientists will have their own interfaces focusing on their specific needs. 

New international academic collaborations are being created at a fast pace, generating data sets each day, in the order of terabytes in size. Often these data sets need to be moved in real-time to a central location to be processed and then shared. In the field of astronomy, building data processing facilities in remote locations is not always feasible, creating the need for a high bandwidth network infrastructure to transport these data sets very long distances. This network infrastructure normally relies on multiple networks operated by multiple organizations or projects. Creating an end-to-end path involving multiple network operators, technologies and interconnections often adds conditions that make the real-time movement of big data sets challenging. The Large Synoptic Survey Telescope (LSST) is an example of astronomical applications imposing new challenges on multi-domain network provisioning activities. The network for LSST is challenging for a number of reasons: (1) with the telescope in Chile and the archiving facility in the USA, the network has a high propagation delay, which affects traditional transport protocols performance; (2) the path is composed of multiple network operators, which means that the different network operating teams involved must coordinate technologies and protocols to support all parallel data transfers in an efficient way; (3) the large amount of data produced (12.7GB/image) and the small interval available to transfer this data (5 seconds) to the archiving facility requires special Quality of Service (QoS) policies; (4) because network events happen, the network needs to be prepared to be adjusted for rainy days, where some data types will be prioritized over others. To guarantee data transfers will happen within the required interval, each network operator in the path needs to apply QoS policies to each of its network links. These policies need to be coordinated end-to-end and, in the case where the network is affected by parallel events, all policies might need to be dynamically reconfigured in real-time to accommodate specific QoS policies for rainy days. Reconfiguring QoS policies is a very complex activity to current network protocols and technologies, sometimes requiring human intervention. This presentation aims to share the efforts to guarantee an efficient network configuration capable of handling LSST data transfers in sunny and rainy days across multiple network operators from South to North America. 

It can take a domain scientist weeks to set up a circuit, meeting for hours with IT administrators to figure out exactly what is needed, approvals from their own campus along with the remote campuses to set up a simple circuit to transfer data between campuses on an ongoing basis. Talking about networks may as well be a foreign language to many domain scientists. As such, we need to make it easier for domain scientists to allocate and configure resources for scientific applications without needing to understand the details of bandwidth, circuits, and port numbers. This session will discuss the challenges in supporting domain science applications across long distances and multiple management domains. We will discuss the AtlanticWave/SDX project and how it approaches this problem, making it possible for a domain scientist with little networking know-how to create paths across an intercontinental network while making network administrators' lives easier in the process. We will focus on the tools being developed to manage the network, along with a practical demonstration spanning multiple SDN switches. 

The SDN paradigm enables network operators to host multiple control planes in parallel, being an approach to support multiple network services. Supporting multiple control planes over production networks exposes the production environment to potential risks and increases operational complexity. To understand and mitigate these risks, we implemented procedures and tools that resulted in a more reliable network. This paper describes our experience and findings with the support of multiple control planes in a wide-area production network.