Table Lookup Time Research Articles

Deep-Learning-Based SDN Model for Internet of Things: An Incremental Tensor Train Approach

The Internet of Things (IoT) has emerged as a revolution for the design of smart applications like intelligent transportation systems, smart grid, healthcare 4.0, Industry 4.0, and many more. These smart applications are dependent on the faster delivery of data which can be used to extract their inherent patterns for further decision making. However, the enormous data generated by IoT devices are sufficient to choke the entire underlying network infrastructure. Most of the data attributes present little or no relevance to the prospective relationships and associations with the projected benefits foreseen. Therefore, order-based generalization mechanisms, known as tensors, can be used to represent these multidimensional data, thereby minimizing the flow table (FT) lookup time and reducing the storage occupancy. So, a novel IoT-train-deep approach for intelligent software-defined networking is designed in this article. The proposed approach works in four phases: 1) tensor representation; 2) deep Boltzmann machine-based classification; 3) subtensor-based flow matching process; and 4) incremental tensor train network for FT synchronization. The proposed model has been extensively tested, and it illustrates significant improvements with respect to delay, throughput, storage space, and accuracy.

A flamelet model for supersonic non-premixed combustion with pressure variation

A modified flamelet model is proposed for studying supersonic combustion with pressure variation considering that pressure is far from homogenous in a supersonic combustor. In this model, the flamelet database are tabulated at a reference pressure, while quantities at other pressure are obtained using a sixth-order polynomial in pressure. Attributed to merit of the modified model which compute coefficients for the expansion only. And they brought less requirements for memory and table lookup time, expensive cost is avoided. The performance of modified model is much better than the approach of using a flamelet model-based method with tabulation at different pressure values. Two types of hydrogen fueled scramjet combustors were introduced to validate the modified flamelet model. It was observed that the temperature is sensitive to the choice of model in combustion area, which in return will significantly affect the pressure. It was found that the results of modified model were in good agreement with the experimental data compared with the isobaric flamelet model, especially for temperature, whose value is more accurately predicted. It is concluded that the modified flamelet model was more effective for cases with a wide range of pressure variation.

Improving Routing Table Lookup in Software Routers

For the speed-up in the routing table lookup in software routers, we introduce a new data structure called, sTable that achieves space efficiency while causing low processing overhead without hardware parallelism. Using a hash table-like approach, sTable stores only next hop information to handle most packets within the on-chip fast memory while causing some accesses to the off-chip slow memory to handle lookup errors. Routing table lookup experiments using Click show that, causing less processing overhead and a smaller number of lookup errors than Bloom filter, sTable reduces the average table lookup time by 44% (i.e., from 119 ns to 67 ns) compared to Bloom filter.

Improving Bloom Filter Forwarding Architectures

To improve Bloom filter-based packet forwarding architectures, we introduce one additional Bloom filter that stores the correct next hop information for recent false positives so as to handle subsequent false positives. Through Click-based experiments, we show that the additional Bloom filter reduces the average table lookup time (i.e., from 120 ns to 93 ns) by handling most false positives within the fast memory.

A Quick Tables Look-up Optimization Algorithm for CAVLC Decoding

In this paper, a quick table look-up optimization algorithm is presented to solve the problems of long table look-up time for CAVLC decoding in H.264/AVC. The achievement of the new algorithm rests that we make full use of the hash table query and index technology to improve the table look-up speed for CAVLC decoding. The basic idea of the new algorithm is that we take the number of zero in code prefix calculated from input bit-stream as index key of the first level, the value of codeword suffix as index key of the second level, then through index key of the one and second index key above, we can quickly get the decoded codeword located in the third level in a hash table built, which can reduce a lot of table look-up time for CAVLC decoding in H.264/AVC. The simulation results show that our proposed schemes based on hash-index method can reduce about 40% table look up time for CAVLC decoding compared with TLSS method, without degrading video quality.

A High Efficient Tables Look-up Algorithm for CAVLC Decoding

In order to solve the problems of high table memory access and long table look-up time and big table storage space in the process of CAVLC decoding for H.264/AVC, a high efficient table look-up algorithm is presented in this paper. The contribution of this paper lies that we uses a program method to realize fully the no-table looking-up of codeword. Specifically, after finding the relationships existed in code length and numbers of 0 in code prefix and code suffix and code, we uses a program code method to realize entirely UVLCTs(Unstructured Variable Length Coding Tables) in CAVLC decoding, As a result, all decoded codewords in UVLCTS can be decoded and obtained easily through a program execution way instead of TLSS(table look-up by sequential search), which could save a lot of table memory access and reduce amount of table look-up time and save a large number of table storage spaces for CAVLC decoding. The simulation results show that our proposed scheme can save 100% table memory access, reduce about 45% table look-up time and save 2320 byte table storage space of table look-up in CAVLC decoding compared with TLSS method, without degrading video quality. Index Terms—Memory access; Table look-up; Storage space; CAVLC decoding; Program method

A Quick Tables Look-up Algorithm based on Hash Query for CAVLC Decoding

Aiming to solve the problem of long table look-up time in table look-up of CAVLC (Context-based Adaptive Variable Length Coding) for H.264/ AVC, a quick table look-up algorithm based on hash query (TLHQ) is proposed to reduce table look-up time for CAVLC decoding in this paper. The basic idea of the new algorithm is that it uses query technology of hash table to rapidly determine code length based on the number of zero in code prefix, As a result, it can save a lot of time of table looking-up and code judging. Experimental results show that our supposed new algorithm could reduce about 20% table look-up time and save 1056 byte storage space for CAVLC decoding in H.264/AVC.

A Tables Look-up Algorithm based on Program Code for CAVLC Decoding

Aiming to solve the problem of high memory access and long table look-up time in table look-up of CAVLC (Context-based Adaptive Variable Length Coding ) for H.264/ AVC , a efficient look-up algorithm based on program code is presented in table look-up for CAVLC decoding in this paper, based on the analysis of the structure of CAVLC code table .The basic idea of this algorithm is that a method based on program code is designed to express fully all the VLCTs (unstructured Variable Length Coding Tables ) in CAVLC decoding. As a result, the decoded codeword in VLCTs can be gotten easily in a program execution way which can save a lot of memory access and table look-up time spent in CAVLC decoding. The simulation results show that the proposed scheme can not only save 100% memory access of table look-up in CAVLC decoding, but also reduce 45% table look-up time in CAVLC decoding without degrading video quality, only adding 2.7kb storage space.

EXPERIMENTAL EVALUATIONS OF ALGORITHMS FOR IP TABLE MINIMIZATION

Reducing the size of IP routing tables is one of the most compelling scaling problems affecting the Internet because of massive growth of routing table entries, increased traffic, and the migration to 128 bit IPv6 addresses. Various algorithms for IP table minimization have been proposed in the literature both for a single and for multiple tables, also with the possibility of performing address reassignments. In this paper we first introduce two new compression heuristics, the BFM and its evolution called BFM-Cluster, that exploit address reassignments for the minimization of multiple routing tables, and then we experimentally evaluate their performances together with the already existing techniques. Since a main problem posed by the growth of the routing tables sizes is the consequent general increase of the table lookup time during the routing of the IP packets, the aim is twofold: to measure and compare the compression ratios of the different algorithms, and to estimate the effects of the compression on the lookup times by measuring the induced improvement on the time of the main algorithms and data structures for the fast IP address lookup from the original tables to the compressed ones. Our point is that the existing techniques are efficient in different situations, with BFM-Cluster heuristic outperforming all other ones.

Application and Research of Rotate-Lookup-Summation in Robot Motion

How to exercise reasonable motion control on robot becomes a new research focus in robot technology research. This paper proposes the Rotate-Lookup-Summation according to the deficiency of traditional look-up table. This method first confirm the controlling factors P1，P2….Pn that would affect the control volume Z, and then the multi-dimensional control volumes and multi-dimensional control factors would be projected to sub-dimensional space. Finally, the same controlling factors of different sub-dimensional space would be rotated to a single sub-dimensional space and establish a corresponding table. According to the single sub-dimensional space value of the controlling factors, the corresponding control variable could be found in table and eventually complete the control process. Experiments show that the method could ensure the integrity and accuracy of table, reduce the table memory capacitance and lookup time, so as to realize the control of look-up table in micro-device.

Performance comparison of bridge algorithms in interconnected local area networks

Abstract Problems of network planning and bridge design in interconnected Local Area Networks are studied by comparing and analyzing the performance of two bridge algorithms: Spanning Tree (ST) and Source Routing (SR), which are currently being specified by the IEEE 802 standards committee. The study supported by the extended LAN model simulation shows that the SR algorithm performs better under both steady-state compared to the ST algorithm. Also, in a real environment, it is pointed out that the table look-up time in ST algorithm and the route locating procedure in SR algorithm are not such critical factors as generally conceived by most people in the performance evaluation of bridging algorithms. Besides, network planning rules related to the topology of the extended network and bridging algorithms are given.

A novel approach for including band-structure effects in a Monte Carlo simulation of electron transport in silicon

We present a novel approach for including the effects of realistic silicon band structure in the simulation of electron transport with a Monte Carlo method. This will be achieved by an electron effective-mass and energy–over–wave-vector relation, which are derived directly from the density of states. Consistent with this model, the scattering rates as well as the equations of motion are determined by the density of states. With our approach the computation tables in three-dimensional momentum space can be replaced by one-dimensional tables in energy space. The necessary amount of memory size and table lookup time is therefore significantly reduced. The number of free parameters in our model is not higher than in the full band-structure model. We show by comparison with full band-structure Monte Carlo and experimental results that there is no loss in physical meaning by the use of the new method.

A pipelined recursive residue number system digital filter

The well-known advantages of pipelining as applied to Finite Impulse Response (FIR) Residue Number System (RNS) arithmetic digital filters is extended to the important area of Infinite Impulse Response (IIR) digital filters through a new technique based on augmentation of the IIR transfer function. Through this technique, pipelined IIR filters based on RNS Read-Only-Memory (ROM) table look-up techniques can be designed which offer throughput rates equal to the table look-up time of the ROM's. This high-speed realization can be achieved even though the recursive filter algorithm requires multiple delays in realizing the output of the filter. For the example of a typical second-order IIR filter, the pipelined structure represents a five-fold increase in speed over standard techniques. Higher order realizations will yield proportionately higher speed improvements. Although the new technique does increase somewhat the hardware complexity of the filter, the increase in speed will often justify the additional hardware. The paper discusses the basic technique, stability considerations, and hardware realizations.