I am pleased to announce two new papers at the 4th International Workshop on Cloud, IoT and Fog Systems (and Security) CIFS 2022 link, and the TASIR Workshop: Testbeds for Advanced Systems Implementation and Research (co-located with the 15th International Conference on COMmunication Systems & NETworkS (COMSNETS 2023)) link.
Title: Efficiently Improving the Performance of Serverless Applications with Microtask-based Scheduling
Authors: Sacheendra Talluri, Sven Lankester, Bogdan Ene, Jesse Donkervliet, Animesh Trivedi, Alexandru Iosup (VU Amsterdam)
Abstract: Serverless computing promises to make cloud computing cheaper and easier to use. However, serverless platforms use coarse-grained scheduling which decreases efficiency and application performance. We propose a fine-grained application model for serverless applications, and use it to design a scheduler to improve application performance and efficiency. We model serverless applications as being composed of microtasks, each with its own unique resource requirements. Microtasks are easily identified via distinct application phases like initialize, read, and process. We provide evidence for the existence of microtasks by experimentally evaluating a serverless online game. We design a scheduler that separates microtasks with different CPU requirements into different queues so that the appropriate amount of CPU cores could be allocated to each queue based on the CPU requirements of the microtasks in that queue. We implement and evaluate the design in an application-level proof-of-concept microtask-based scheduler and compare it to task-based scheduling commonly used by serverless platforms. For a distributed sort application, the microtask-based scheduler decreases application makespan by 37% and the duration of I/O based application stages by 81%, compared to task-based scheduling. Our work suggests that there is potential in extracting and using microtask information from serverless applications.
Title: Performance Measurement of Distributed Storage on Edge Devices
Authors: Priyam Loganathan (BITS Pilani, Goa, India), Dhruv Rauthan (BITS Pilani, Goa, India), Animesh Trivedi (VU Amsterdam, The Netherlands), Vinayak Naik (BITS Pilani, Goa, India)
Abstract: Edge computing and storage have been proposed as an alternative to current cloud computing infrastructures to address the latter’s limitations. However, edge computing services and support for applications based on edge infrastructures are still in the incipient stages. There are numerous storage services for efficient data storage in the cloud. In contrast, few such services exist at the edge. We look at the performance of a cloud storage service Cassandra when deployed on an edge network. In this paper, we analyze Cassandra’s performance in three different networks. We vary Cassandra’s attributes and the overall conditions in each network to simulate a real-world scenario and study the effect on the total time taken for a particular request. We (a) establish trends for different configurations, (b) discuss the reasoning behind them, and (c) suggest optimal configurations for the networks used, which are important for deploying Cassandra in an edge setting.