It has obvious advantage to adopt 802.1x authentication for network access *** paper analyzes802.1x protocol,EAP protocol and RADIUS protocol,and constructs AAA which is based on 802.1x authentication at the *** softw...
详细信息
ISBN:
(纸本)9781479919819
It has obvious advantage to adopt 802.1x authentication for network access *** paper analyzes802.1x protocol,EAP protocol and RADIUS protocol,and constructs AAA which is based on 802.1x authentication at the *** software the messages of the whole authentication process have been *** to AAA mechanism it analyzes EAP messages and RADIUS messages *** analysis of these messages provides technology strongly for particular research and further improvement,and it has important value for research and application.
In this paper, we propose a brand-new cache power reduction mechanism, Thread ID Cache (TID Cache), aiming to reduce dynamic power consumption of set-associative caches shared by multiple threads. TID tables are imple...
详细信息
In this paper, we propose a brand-new cache power reduction mechanism, Thread ID Cache (TID Cache), aiming to reduce dynamic power consumption of set-associative caches shared by multiple threads. TID tables are implemented at different levels of cache hierarchy to record thread IDs. When a thread accesses a cache way, the thread ID is updated to the corresponding entry in TID table. When the same thread accesses cache again, irrelevant ways owned by other threads can be avoided being accessed through comparing thread IDs in TID table. Experiments are conducted on TID Cache with the configuration of a 32KB 8-way L1 cache and a 128KB 8-way L2 cache. The results reveal that, when caches are shared by 16 threads, the proposed TID cache achieves 34% dynamic power reduction of L1 and 36% of L2 under the CloudSuite benchmarks, and the corresponding results are 42% of L1 and 48% of L2 under the SPLASH-2 benchmarks. Accordingly, the same proportion of dynamic power can be saved at the price of less than 0.8% area overhead of cache data arrays.
In this paper, we present a study on how to achieve Byzantine fault tolerance for collaborative editing systems with commutative operations. Recent research suggests that Conflict-free Replicated Data Types (CRDTs) ca...
详细信息
ISBN:
(纸本)9781467399869
In this paper, we present a study on how to achieve Byzantine fault tolerance for collaborative editing systems with commutative operations. Recent research suggests that Conflict-free Replicated Data Types (CRDTs) can be used to construct collaborative editing systems where concurrent update operations are commutative. This new approach is shown to avoid the complex issue of conflict resolution for concurrent updates to a shared document. The shared document is often modeled as a linear text buffer where each basic element is assigned a globally unique and totally ordered identifier. The linear text buffer constructed this way would constitute as a CRDT, which would make concurrent update operations issued by different users commutative. state convergence at all users can be achieved automatically as long as the users could receive the same set of operations irrespective of their relative ordering. However, it is not straightforward to guarantee state convergence in the presence of malicious users and external adversaries. In this paper, we carefully analyze the threats towards this type of systems, and propose a lightweight solution to achieve Byzantine fault tolerance with low runtime overhead. We define a set of correctness properties for such systems and prove that the proposed Byzantine fault tolerance mechanisms guarantee these properties.
The modular assembly of plasmonic nanoparticles enables the creation of a new class of two-dimensional nanoparticle sheets with novel optical *** exciting advances,undeveloped issues still remain in terms of nanoparti...
详细信息
The modular assembly of plasmonic nanoparticles enables the creation of a new class of two-dimensional nanoparticle sheets with novel optical *** exciting advances,undeveloped issues still remain in terms of nanoparticle sheet design and potential ***,we present our research work which describes a simplistic and general polymer-based approach to assemble core-shell Au@Ag nanocubes into periodic nanoparticle arrays.
The modular assembly of plasmonic nanoparticles enables the creation of a new class of two-dimensional nanoparticle sheets with novel optical signatures. Despite exciting advances, undeveloped issues still remain in t...
详细信息
The modular assembly of plasmonic nanoparticles enables the creation of a new class of two-dimensional nanoparticle sheets with novel optical signatures. Despite exciting advances, undeveloped issues still remain in terms of nanoparticle sheet design and potential applications. Here, we present our research work which describes a simplistic and general polymer-based approach to assemble core-shell Au@Ag nanocubes into periodic nanoparticle arrays.[1] These nanocubes are functionalized with polystyrene and spread on the surface of water droplet on a holey copper mesh. Upon water evaporation, the balance between the nanoparticle core-to-core attraction and polystyrene ligand-to-ligand repulsion forces led to the formation of a free-standing lattice-like nanosheet with ultimate thickness limit but macroscopic lateral dimensions. These sheets, which we termed as plasmene, exhibited unique novel properties such as localized gap and propagating plasmonics. Using focused ion beam(FIB) lithography, the superior robustness of these nanosheets allows it to be milled into one-dimensional nanoribbons and folded into geometrically well-defined three-dimensional origami structures such as cubes, hearts and even a ‘flying bird‘. Due to the structural homogeneity, mechanical flexibility and optical transparency of these sheets, we also demonstrated the application of these nanosheets as a surface-enhanced Raman scattering(SERS) substrate for chemical detection on topologically complex surfaces such as banknotes and coins.[2,3]
Object tracking is one of the most important components in numerous applications of computer vision. Much progress has been made in recent years. The Tracking-Learning-Detection (TLD) algorithm achieves excellent perf...
详细信息
In order to alleviate the problem raised in Gomez's method[1],where messages with different source-destination pairs have great probability choosing the overlapped paths therefore,new transmission bottleneck might...
详细信息
ISBN:
(纸本)9781510805750
In order to alleviate the problem raised in Gomez's method[1],where messages with different source-destination pairs have great probability choosing the overlapped paths therefore,new transmission bottleneck might be generated nearby,we present an intermediate node based deterministic fault-tolerant routing,named *** method selects the qualified intermediate node among all the candidates by its residual bandwidth without over-stressing any particular *** addition,deterministic routing is applied along both subpaths to guarantee deadlock-free and in-order message *** thereby achieves traffic workload balance across the network and graceful performance degradation even in the presence of faults.
In order to get better performance, lots of optimization methods are used in code transformation. When migrating SPMD to multi-core platform, vectorization is one key optimization to improve performance. Control flow ...
详细信息
Image steganography has been widely used in the domain of privacy protection, such as the storage and transmission of the secret images. This paper presents a novel image-hiding scheme to embed the secret image into a...
详细信息
A novel memory consistency model for thousand-core processors is presented. The model simplifies cache coherence for the full chip, and reduces cache design complexity. In addition, the model has the ability to descri...
详细信息
暂无评论