Networks of tree-ring data are commonly applied in statistical reconstruction of spatial fields of climate variables. The importance of elevation to the climatic interpretation of tree-ring networks is addressed using 281 station precipitation records, and a network of 79 tree-ring chronologies from different species and a range of elevations in the eastern Mediterranean. Cluster analysis of chronologies identifies 6 tree-ring groups, delineated principally by site elevation. Correlation analysis suggests several of the clusters are linked to homogenous elevational moisture regimes. Results imply that climate stations close to the elevations of the tree-ring sites are essential for assessing the seasonal climatic signal in tree-ring chronologies from this region. A broader implication is that the elevations of stations contributing to gridded climate networks should be considered in the design and interpretation of field reconstructions of climate from tree rings. Finally, results suggest elevation-stratified tree-ring networks as a strategy for seasonal climate reconstruction.

No description available.

Dendroclimatological research along a geographical gradient is important to understanding both spatial and temporal characteristics of climate influences on tree growth. In this study, three tree-ring width chronologies, obtained from field collection and previous research, were used to represent tree growth along a longitudinal transect from coast to inland in the southeastern U.S.: Hope Mills, located at the Atlantic Coastal Plain; Linville Mountain, located on the eastern side of the Appalachian Mountains; and Gold Mine Trail, located on the western side of the Appalachians. The variations of ring width indices in chronologies reflect extreme climatic events such as severe droughts or cold periods. Correlation and response function analyses were used to examine the climate-tree growth relationship at three sites. The temporal stationarity of climate signals was tested using moving interval analysis in DENDROCLIM2002. Winter temperature was the limiting climate factor for the western mountain site, while moisture was more important for tree growth in the eastern mountain and coastal area sites. However, all significant climate signals found in the trees were not stable over time. The tendency of a shift from precipitation signal to temperature signal is notable around the mid-20th century. Winter North Atlantic Oscillation (NAO) had positive correlations with radial growth at the two mountain sites, which might explain the winter temperature response by trees. The Atlantic Multidecadal Oscillation (AMO) showed an annual feature of associations with growth, and the multidecadal duration of significant correlations was also apparent. The Pacific-related Pacific Decadal Oscillation (PDO) and El Niño-Southern Oscillation (ENSO) also tended to influence tree growth. Along the coastal-inland transect, gradient features of climate oscillation signals did exist. Relationships changed with phase changes of the oscillations. Land-sea boundaries and high mountains may determine the climate response patterns in the Southeast. Other factors such as microenvironment, human disturbance, and biological reaction of trees to climate change also have influence. It is not reliable to use the composite chronology to study the effect of climate oscillations for the Southeast region. In the future, a large number of sample sites will be necessary to more extensively study the regional climate-tree growth relationship.

Thesis (Ph.D.) - Carleton University, 2007. / Includes bibliographical references (p. 210-215). Also available in electronic format on the Internet.

Diploma thesis deals with monitoring and diagnostics of transport networks. It focuses on basic diagnostic and surveillance tools, and tools which are developed in the project of Internet2. It is focused on the evaluation of the measurements performed by these tools with a focus on external factors and substandard conditions and their impact on the measurement results.

Ring networks are appropriate for the full range of network levels, including multiprocessor systems, local area computer networks and high speed backbones. The most well known and widely implemented examples are the IBM token ring and FDDI networks. Ring networks have the advantages of high channel utilisation and bounded delay if an n-limited service policy is used. The packet transfer delay, defined as the average time a packet spends in the network from the time it is generated until the time it is received at its destination node, improves with the number of rings on which a node is connected. However, many ring connections are not economically feasible since the cost of the ring interface increases with the number of rings. There has been an abundance of previous work on single token ring networks. A number of papers on slotted rings, register insertion rings and more complex ring architectures have also been published. However, there is very little existing literature on multiple ring networks as well as ring networks in clustered traffic environments, i.e. where nodes from the same cluster tend to communicate more with each other than with other nodes in the network. This thesis focuses on two network topologies that make use of multiple rings and are well suited to clustered traffic environments: the two-connected multiple ring (2-MR) and the destination removal double ring (DRDR). For the 2-MR network, three different practical token-based protocols are investigated in an attempt to optimise performance. It is further shown that significant performance improvements can be achieved by employing a slotted ring protocol rather than the token ring protocol. The DRDR network is also examined and its performance compared to the aforementioned architectures. For each of the six cases, both random and clustered traffic patterns are considered and compared. Analytical results are derived which are verified by results obtained from computer simulations. Furthermore, we look at exact methods of analysing ring networks. A mean value analysis of a single token ring network with a I-limited service discipline is performed, which clearly shows the complexity exact methods introduce. Finally, although it has been stated in the literature that an exact analysis of a multiple symmetrical token ring network is intractable, we present a novel Markov chain approach that gives exact results for near zero loads. / Thesis (M.Sc.-Electronic Engineering)-University of Natal, 1998.

Thesis (Ph.D.) - Carleton University, 2006. / Includes bibliographical references (p. 139-144). Also available in electronic format on the Internet.

Målet med detta arbete har varit att undersöka möjligheterna för en internetleverantör att dra nytta av VXLAN i sitt nätverk där ringformade topologier används. Ambitionen har varit att samtliga switchar i access- och distributionslagret kan konfigureras med samma VLAN. Dessutom behöver slutkunderna separe- ras helt och hållet i nätverket, trots att de är anslutna till samma VLAN. Ett av företagets krav är att trafiken i nätverket transporteras till ett specifikt interface i core-lagret. Genom att använda VXLAN skapas två klara fördelar. Dels att konfigurationen och administrationen av VLAN på switcharna förenklas och effektiviseras avsevärt, dels att trafiken i ringen transporteras över lager 3 till core-lagret och därmed kan skickas till ett logiskt interface. Detta medför att trafiken kan nå core-lagret på vilket fysiskt interface som helst, oavsett i vilken riktning trafiken transporteras i ringen. Genom att upprätta en testmiljö med ett antal switchar för att simulera nätverket i mindre skala, testas och verifieras att VXLAN fungerar och kan användas till det tänkta ändamålet. Resultatet visar att den nya konfigurationen med VXLAN kan påvisa en stabil och kostnadseffektiv plattform. Förtjänsterna innebär skalbarhet, enklare administration samt möjligheter att exempelvis utveckla nya system och tjänster, utbilda personalen, samt att förbättra underhållet av nätverket. / The aim of this work has been to investigate the possibilities for an Internet service provider to benefit from VXLAN in its network where ring topologies are used. The ambition has been that all switches in the access and distribution layer can be configured with the same VLANs. In addition, end customers need to be completely separated in the network, even though they are connected to the same VLAN. One of the company’s requirements is that the traffic in the network is transported to a specific interface in the core layer. Using VXLAN creates two clear benefits. On the one hand, the configuration and administration of the VLANs on the switches is greatly simplified and streamlined, and on the other hand, the traffic in the ring is transported over layer 3 to the core layer and thus can be sent to a logical interface. This means that traffic can reach the core layer on any physical interface, no matter in which direction the traffic is transported in the ring. By establishing a test environment with a number of switches to simulate the network on a smaller scale, it is tested and verified that VXLAN works and can be used for the intended purpose. The result shows that the new configuration with VXLAN can create a stable and cost-effective platform. The benefits include scalability, simpler administration and opportunities to for example develop new systems and services, train the staff, and improve the maintenance of the network

Pervasive use of GPUs across multiple disciplines is a result of continuous adaptation of the GPU architectures to address the needs of upcoming application domains. One such vital improvement is the introduction of the on-chip cache hierarchy, used primarily to filter the high bandwidth demand to the off-chip memory. However, in contrast to traditional CPUs, the cache hierarchy in GPUs is presented with significantly different challenges such as cache thrashing and bandwidth bottlenecks, arising due to small caches and high levels of memory traffic. These challenges lead to severe congestion across the memory hierarchy, resulting in high memory access latencies. In memory-intensive applications, such high memory access latencies often get exposed and can no longer be hidden through multithreading, and therefore adversely impact system performance. In this thesis, we address the inefficiencies across the memory hierarchy in GPUs that lead to such high levels of congestion. We identify three major factors contributing to poor memory system performance: first, disproportionate and insufficient bandwidth resources in the cache hierarchy; second, poor cache management policies; and third, high levels of multithreading. In order to revitalize the memory hierarchy by addressing the above limitations, we propose a three-pronged approach. First, we characterize the bandwidth bottlenecks present across the memory hierarchy in GPUs and identify the architectural parameters that are most critical in alleviating congestion. Subsequently, we explore the architectural design space to mitigate the bandwidth bottlenecks in a cost-effective manner. Second, we identify significant inter-core reuse in GPUs, presenting an opportunity to reuse data among the L1s. We exploit this reuse by connecting the L1 caches with a lightweight ring network to facilitate inter-core communication of shared data. We show that this technique reduces traffic to the L2 cache, freeing up the bandwidth for other accesses. Third, we present Poise, a machine learning approach to mitigate cache thrashing and bandwidth bottlenecks by altering the levels of multi-threading. Poise comprises a supervised learning model that is trained offline on a set of profiled kernels to make good warp scheduling decisions. Subsequently, a hardware inference engine is used to predict good warp scheduling decisions at runtime using the model learned during training. In summary, we address the problem of bandwidth bottlenecks across the memory hierarchy in GPUs by exploring how to best scale, supplement and utilize the existing bandwidth resources. These techniques provide an effective and comprehensive methodology to mitigate the bandwidth bottlenecks in the GPU memory hierarchy.