No abstract

Stream processing systems need to be elastically scalable to process and respond the unpredictable massive load spike in real-time with high throughput and low latency. Though the modern cloud technologies can help in elastically provisioning the required computing resources on-the-fly, finding out the right point-in-time varies among systems based on their expected QoS characteristics. The latency sensitivity of the stream processing applications varies based on their nature and pre-set requirements. For few applications, even a little latency in the response will have huge impact, whereas for others the little latency will not have that much impact. For the former ones, the processing systems are expected to be highly available, elastically scalable, and fast enough to perform, whenever there is a spike. The time required to elasticity provision the systems under FaaS is very high, comparing to provisioning the Virtual Machines and Containers. However, the current FaaS systems have some limitations that need to be overcome to handle the unexpected spike in real-time. This paper proposes a new algorithm called Elastic-FaaS on top of the existing FaaS to overcome this QoS latency issue. Our proposed algorithm will provision required number of FaaS container instances than any typical FaaS can provision normally, whenever there is a demand to avoid the latency issue. We have experimented our algorithm with an event stream processing system and the result shows that our proposed Elastic-FaaS algorithm performs better than typical FaaS by improving the throughput that meets the high accuracy and low latency requirements.

Quality of service (QoS) is widely adopted to characterize the performance of services invoked by users. For this purpose, the QoS prediction of services constitutes a decisive tool to allow end-users to optimally choose high-quality cloud services aligned with their needs. The fact is that users only consume a few of the broad range of existing services. Thereby, performing a highly accurate service recommendation becomes a challenging task. To tackle the aforementioned challenges, a hybrid deep learning technique is proposed to predict future QoS parameters. Initially, the data is collected from the WSDream data repository, however, the collaborating filtering (CF) technique is used to pre-process the collected data. Accordingly, the filtering technique resolves missing or undermined values and cleans the node. Additionally, this pre-processing technique reduces the error and promotes prediction efficiency. Subsequently, data clustering is performed by introducing a modified k-medoids algorithm, which groups the user services into different clusters. Accordingly, a future QoS is predicted based on the proposed hybrid deep learning method like convolutional neural network-long short-term memory. The proposed method is implemented using Python software and it evaluates response time, throughput, root mean square error, and mean absolute error (MAE). The proposed method is compared with the existing collaborative filtering-k medoids, support vector regression, self-exciting threshold auto-regressive, genetic guided clustering algorithm algorithms, and novel deep learning-based hybrid approach. The result of the proposed method presents higher accuracy and low consumption time. Therefore, the performance of the proposed method is 6% higher than the other existing methods. Accordingly, the MAE, mean absolute percentage error, mean absolute scaled error, and root mean squared error of the proposed method produces 7%, 7.5%, 2%, and 1.5% higher performance than the other existing methods.Subsequently, the precision and recall of the proposed method are approximately 3.5% and 3% high than the other state of art methods. The experimental results highlight that the viability of the proposed approach contrasted with the regular methodologies in real-time. Consequently, by introducing this proposed method the QoS prediction performance is improved and it provides higher-quality solutions.