Search

Large parallel file systems (PFSs) must service multiple independent workloads concurrently. Workloads often require different quality of service (QoS) of metadata I/O. However, existing PFSs are unable to differentiate metadata I/O from different workloads. There are many researches on QoS support of storage system, they do not focus on metadata I/O of PFSs as one metadata I/O can have several sub-operations and different sub-operations have different access size. In addition, users cannot specify metadata QoS requirements without a suitable QoS specification. This paper presents a novel scheme QoSFM (QoS Support for Metadata) to support metadata I/O QoS in PFSs. QoSFM allows users to specify workloads' metadata QoS requirements through our two-layered QoS specification mechanism. QoSFM can dynamically translate the workload-level goals into system resource needs such as service time per time window and sub-operation latency. Client-side controllers throttle the metadata I/O based on service time allocation to meet the throughput goal while metadata servers (MDSs) reorder the sub-operations based on deadlines to meet the latency goal. Extensive simulations show that QoSFM can effectively satisfy workload-level metadata I/O QoS requirements. Meanwhile, results also show that QoSFM has low overhead.

Metadata-intensive applications make metadata access the bottleneck of system. A metadata-intensive I/O performance optimization method by merging read/write requests is proposed. Extensive simulations show that aggregate throughput of intensive metadata access can be increased by up to 8.45 times and average response time can be decreased by up to 99.02 percent when the merging period and request interval is configured as 0.4ms and 0.025ms respectively.

The physical characteristics of current mobile devices impose significant constraints on the processing of 3D graphics. The remote rendering framework is considered a better choice in this regard. However, limited battery life is a critical constraint when using this approach. Earlier methods based on this framework suffered from high transmission frequency. We present a software solution to this problem with a key element, an Adaptive Splitting and Error Handling Mechanism that indirectly reserves the electricity in mobile devices by reducing the transmission frequency. To achieve this goal, a geometric relation is maintained that tightly couples several consecutive Levels of Detail (LOD). Adaptive Splitting can then approximate the LOD from a much coarser split base under the guidance of the relation. Data transmission between the server and mobile device occurs only when out-ranged LOD is about to be displayed. Our remote rendering architecture, based on the above approach, trades splitting process for transmission, thereby alleviating the problem of frequent data transmission.

Air traffic management information network carries many important data which have relations to flight safety of civil aviation. Aiming to fight against the risks of certification system in ATM information network, Diameter security authentication model is proposed and achieved. The system improves the security of identification, supports for mutual authentication, finally achieves security unity certification in ATM information network.

Handoff scheme provides continuous and seamless transfer when the mobile node moves around the overlay area of wireless networks. It is not practical to use the traditional Received Signal Strength Indication (RSSI) based methods to decide when to process handoff in Collaborative Virtual Environment (CVE), because the probe delay occupies most of the handoff delay time. In this paper, we propose a handoff scheme based on time interval indication (TII) in CVE over WLAN. Using the time interval between state protocol data units (PDUs) as a metric, can reduce the number of probes to decrease the probe delay in the processing of CVE. We take advantage of state forecast arithmetic to simulate object movement during processing of handoff. Numerical results demonstrate that the proposed approach reduces probe delay time in the entire operation of CVE.

Cementing quality directly influences exploitation of oil or gas layers in oil or water boreholes. Characteristic wave (or pseudo-Rayleigh wave) in acoustic full waveform can evaluate the cementing quality of the secondary interface in cased boreholes. Before evaluating the cementing quality of secondary interface, we must analyze bath propagation velocity and amplitude of the characteristic wave. To achieve this, we simulate the acoustic full waveform and 2-D spectrum (frequency-wavenumber spectrum) for various cementing quality of the secondary interface based on the real axis integration method, and analyze relation between bath velocity and amplitude of the characteristic wave and thickness of the water layer on the secondary interface. The simulating results for the waveform and 2-D spectrum of characteristic wave indicate that the velocity of characteristic wave is influenced by the velocity of bath the casing and cement annulus together and is small than that of cased wave. The velocity is increasing and up to the velocity of cased wave with the thickness of water layer. In efficient range of the thickness of water layer, the amplitude of characteristic wave is increasing with the thickness of water layer. Moreover, we validate the simulation results by comparing with experiment data in calibration pits. It is shown that the difference of amplitude or velocity for the characteristic wave can indicate the cementing quality of the secondary interface. These results are of great importance to build the standards of extracting the characteristic wave and criteria of evaluating the cementing quality of the secondary interface.

For supporting real-time interaction in distributed virtual reality application (DVR), it is common to distribute all geometry data to clients from server before the start of the application. For efficient replication, two schemes are used in general - prioritized transfer of objects and caching and prefetching technique. The existing caching and prefetching approaches are mostly based on the information of avatarpsilas geometric attributes. However, these information failed to determine which types of objects are more important to an individual avatar. To solve this problem, we propose a data management scheme using integrated information, including geometric attributes and non-geometric attributes. Then we give a prefetching and caching scheme according to the integrated information. In the end, the simulation experiment shows that our method can reflect the avatar's degree of attention to those objects in DVR better than other previous methods.