No abstract

No abstract

No abstract

Data centers are increasingly relying on software defined networking (SDN) to orchestrate data transmission. To maximize network utilization, the SDN controller needs to frequently update the data plane as the network conditions change. Due to its asynchronous nature, data plane update may result in serious transient congestion and packet loss. Prior work strives to find a congestion-free update plan with multiple stages, each of which guarantees that there will be no congestion independent of the update order. This approach prevents the network from being fully utilized and requires solving a series of LP with scalability challenges.In this paper, we study the general problem of minimizing transient congestion during network update, given the number of intermediate stages. This exposes the tradeoff between update speed and transient congestion, which may be absorbed by switch buffers, and allows the operator to navigate a broader design space. We formulate the minimum congestion update problem (MCUP) as an optimization program, and propose heuristics to find the update sequence efficiently. Preliminary results show that our approach increases link utilization by 20% and reduces update time by 50% compared to prior work.

To implement the quality prediction scheme for batch processes, long short-term memory (LSTM) neural network is a feasible tool to handle with the process dynamics and nonlinearity. However, a global LSTM soft sensor suffers a decline in performance facing batch-to-batch variations. To overcome the batch diversity problem and take advantage of LSTM model, a multivariate trajectory based ensemble just-in-time learning strategy is proposed in this paper. Different trajectory based similarity measurements are designed to extract historical batch trajectories which share similar spatial positions and trends. For each selected trajectory, an online local LSTM soft sensing model is constructed and the real-time quality prediction result for each local model can be obtained. Then, a weighting parameter is determined for each model by cross validation. Bringing together quality prediction results from different local models, the ensemble prediction result can be finally figured out. Two case studies are carried out to prove the effectiveness of the proposed methodology including a fed-batch reactor and the fed-batch penicillin fermentation process.INDEX TERMS Batch production systems, Ensemble just-in-time learning, long short-term memory, multivariate trajectory analysis, soft sensor, quality prediction.