Data Transfer Nodes Research Articles

Numerical analysis of data transfer quality in the 3D multi-scale uncoupled concurrent model connected with DMR

Analysis of the quality of obtained results from a 3D multiscale concurrent finite element numerical model based on Digital Material Representation (DMR) concept is the main aim of the research. Particular attention is put on an influence of different number of data transfer nodes between subsequent scales on material behavior predicted by a micro DMR model. Conclusions are drawn based on results in the form of equivalent strain distribution, homogenized stress-strain curves and samples shape changes.

Long-haul secure data transfer using hardware-assisted GridFTP

Extreme-scale scientific collaborations require high-performance wide-area end-to-end data transports to enable fast and secure transfer of high data volumes among collaborating institutions. GridFTP is the de facto protocol for large-scale data transfer in science environments. Existing predominant network transport protocols such as TCP have serious limitations that consume significant CPU power and prevent GridFTP from achieving high throughput on long-haul networks with high latency and potential packet loss, reordering and jitter. On the other hand, protocols such as UDT that address some of the TCP shortcomings demand high computing resources on data transfer nodes. These limitations have caused underutilization of existing high-bandwidth links in scientific and collaborative grids. To address this situation, we have enhanced Globus GridFTP, the most widely used GridFTP implementation, by developing transport offload engines such as UDT and iWARP on SmartNIC, a programmable 10GbE network interface card (NIC). Our results show significant reduction in server utilization and full line-rate sustained bandwidth in high-latency networks, as measured for up to 100 ms of network latency. In our work, we also offload OpenSSL on SmartNIC to reduce host utilization for secure file transfers. The offload engine can provide line-rate data channel encryption/decryption on top of UDT offload without consuming additional host CPU resources. Lower CPU utilization leads to increased server capacity, which allows data transfer nodes to support higher network and data-processing rates. Alternatively, smaller or fewer DTNs can be used for a particular data rate requirement.

A Wireless Sensor Network for Hospital Security: From User Requirements to Pilot Deployment

Increasing amount of Wireless Sensor Network (WSN) applications require low network delays. However, current research on WSNs has mainly concentrated on optimizing energy-efficiency omitting low network delays. This paper presents a novel WSN design targeted at applications requiring low data transfer delays and high reliability. We present the whole design flow from user requirements to an actual pilot deployment in a real hospital unit. The WSN includes multihop low-delay data transfer and energy-efficient mobile nodes reaching lifetime of years with small batteries. The nodes communicate using a low-cost low-power 2.4GHz radio. The network is used in a security application with which personnel can send alarms in threatening situations. Also, a multitude of sensor measurements and actuator control is possible with the WSN. A full-scale pilot deployment is extensively experimented for performance results. Currently, the pilot network is in use at the hospital.

Dominating Set Theory based Semantic Overlay Networks for Efficient and Resilient Content Distribution

Recently overlay networks have emerged as an efficient and flexible method for content distribution. An overlay network is a network running on top of another network, usually the Internet. This network is formed by a subset of underlying physical nodes. The connections between the overlay nodes are provided by overlay links, each of which is usually composed of one or more physical link. These networks are employed in many settings to provide logical communication infrastructure over an existing communication networks. The main objective of the overlay network is to reduce routing path lengths, stretched by the overlay routing process. In the existing solutions developed, a kind of fixed infrastructure in the form of excessive message exchange is necessary to guarantee good overlay properties. The scope of our effort is to construct an overlay network based on Dominating Set Theory to optimize the number of nodes for large data transfer. Fast Replica algorithm is applied to reduce the content transfer time for replicating the content within the semantic network. A dynamic parallel access scheme is introduced to download a file from different peers in parallel from the Semantic Overlay Network (SON), where the end users can access the members of the SON at the same time, fetching different portions of that file from different peers and reassembling them locally. That is, the load is dynamically shared among all the peers. To eliminate the need for retransmission requests from the end users, an enhanced digital fountain with Tornado codes is applied. Decoding algorithm at the receiver will reconstruct the original content. In this no feedback mechanisms are needed to ensure reliable delivery. This paper analyzes the performance of sequential unicast, multiple unicast and fast replica with tornado content distribution strategies in terms of content replication time and delivery ratio. This paper also analyzes the impact of dominating set theory for the construction of semantic overlay networks.