{"title":"Towards Efficient Scheduling of Concurrent DNN Training and Inferencing on Accelerated Edges","authors":"Prashanthi S K, Vinayaka Hegde, Yogesh L. Simmhan","doi":"10.1109/CCGridW59191.2023.00084","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00084","url":null,"abstract":"Edge devices are typically used to perform low-latency DNN inferencing close to the data source. However, with accelerated edge devices and privacy-oriented paradigms like Federated Learning, we can increasingly use them for DNN training too. This can require both training and inference workloads to be run concurrently on an edge device, without compromising on the inference latency. Here, we explore such concurrent scheduling on edge devices, and provide initial results demonstrating the interaction of training and inferencing on latency and throughput.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124565207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lasani Hussain, Mayank Rawat, Neeraj Kumar Yadav, S. Darak, Praveen Tammana, Rinku Shah
{"title":"Microservice-based in-network security framework for FPGA NICs","authors":"Lasani Hussain, Mayank Rawat, Neeraj Kumar Yadav, S. Darak, Praveen Tammana, Rinku Shah","doi":"10.1109/CCGridW59191.2023.00074","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00074","url":null,"abstract":"Recent advancements in programmable network hardware such as smartNICs (e.g., NPU, DPU, FPGA)1 and programmable switches (e.g., Intel Tofino) have created opportunities to save precious CPU cycles and achieve low latencies by offloading applications to these devices [1] –[3].","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"128 1-2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120928588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
P. V. Caderno, Feras M. Awaysheh, Yolanda Colino-Sanguino, Laura Rodriguez de la Fuente, F. Valdés-Mora, J. C. Cabaleiro, T. F. Pena, D. Gallego-Ortega
{"title":"OPERA-gSAM: Big Data Processing Framework for UMI Sequencing at High Scalability and Efficiency","authors":"P. V. Caderno, Feras M. Awaysheh, Yolanda Colino-Sanguino, Laura Rodriguez de la Fuente, F. Valdés-Mora, J. C. Cabaleiro, T. F. Pena, D. Gallego-Ortega","doi":"10.1109/CCGridW59191.2023.00038","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00038","url":null,"abstract":"genome Sequence Alignment Map The rapidly increasing demand for next-generation biotechnologies has enabled the development of DNA and RNA big data-oriented (BD) pipelines. The preprocessing stage requires sequencing and alignment tools that provide barcoding for error correction and increase accuracy during sequencing. Unique Molecular Identifiers (UMIs) promise a highly accurate bioinformatic identification of PCR duplication before the amplification stage. However, using alignment coordinates alone is Data-intensiveand challenging due to the increased demand for computational throughput, affecting the performance of the underlying resources. This paper proposes a highly scalable data scheduling and resource allocation framework called OPERA-gSAM for the genome Sequence Alignment Map (SAM). OPERA-gSAM, an OPportunistic and Elastic Resource Allocation, is an enabling big data platform (i.e., Apache Spark) for the next-generation massively parallel sequencing applications. We validate OPERA-gSAM scalability and efficiency using Genomics single-cell RNA sequencing. Our experiments demonstrate the usability and high efficiency of the proposed framework. Results show that OPERA-gSAM is up to 2.4× faster while consuming 50% fewer resources than the conventional pipeline using SAM and UMI tools.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116686162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shashwat Jaiswal, Suman Raj, Subhajit Sidhanta, Yogesh L. Simmhan
{"title":"A Lightweight, Mobility-Aware, Geospatial & Temporal Data Store for Multi-UAV Systems","authors":"Shashwat Jaiswal, Suman Raj, Subhajit Sidhanta, Yogesh L. Simmhan","doi":"10.1109/CCGridW59191.2023.00066","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00066","url":null,"abstract":"The meteoric rise in cutting-edge research in the area of cyber-physical systems has made it possible for UAVs to serve as computing infrastructure for processing the data collected from on-board sensors as well as ground-based IoT (Internet of Things) devices. To enable UAVs to perform edge computing tasks on the data received from cameras and other sensors in a near real-time manner, the input data or its associated metadata needs to be stored and processed locally, i.e., on the UAVs themselves. To that end, we present the design of DroneDB - a lightweight data storage engine that can process geospatial and time series data on a multi-UAV system comprising: A) a swarm of UAVs fitted with onboard computers, and B) ground-based fog servers. DroneDB is mobility-aware and has been optimized to perform efficient processing of different possible combinations of spatial and temporal queries, which are typically performed on time series data by real-world applications such as pollution, weather, and traffic monitoring.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127174293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research Posters Committee Members","authors":"","doi":"10.1109/ccgridw59191.2023.00014","DOIUrl":"https://doi.org/10.1109/ccgridw59191.2023.00014","url":null,"abstract":"","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130591541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Somin Song, Sahil Suneja, Michael V. Le, Byungchul Tak
{"title":"Sequence-based System Call Filtering for Enhanced Container Security, is it beneficial?","authors":"Somin Song, Sahil Suneja, Michael V. Le, Byungchul Tak","doi":"10.1109/CCGridW59191.2023.00057","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00057","url":null,"abstract":"One critical attack that exploits kernel vulnerabilities through system call invocations is the privilege escalation followed by the infamous container escape. The seccomp provides the first line of defense against it. However, it is known to be brittle since it operates at the granularity of the individual system call. Inadvertent filtering of necessary system calls may inhibit the correct execution while overly generous rules allow the attacks.We believe that, by looking at the sequence of system calls, we can achieve more accurate and effective blocking of attacks in containers. To this end, we analyzed the expected defensive power from applying the sequence-based filtering mechanisms by thoroughly analyzing a large set of collected kernel vulnerabilities to assess the feasibility.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134223239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Knowledge Enhanced Digital Objects: a Data Lake Approach","authors":"Yu Luo, Beth Plale","doi":"10.1109/CCGridW59191.2023.00064","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00064","url":null,"abstract":"Open science is gaining traction as a means to bring greater value to the products of research beyond the published manuscript - that is, to make data and software more accessible and usable. The FAIR principles [1] are a significant step forward in defining the characteristics needed of data for it to be shared, used, and reused.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133196032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Blockchain-based Security Management Framework for Cyber-Physical Systems","authors":"Debashis Das, Sourav Banerjee, Rakhi Chakraborty, K. Dasgupta, Pushpita Chatterjee, Uttam Ghosh","doi":"10.1109/CCGridW59191.2023.00021","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00021","url":null,"abstract":"Cyber-Physical Systems (CPS) are becoming increasingly common in various domains, including industrial control systems, smart cities, and healthcare. However, the security of CPS is a major concern due to its interconnected and heterogeneous nature. In this paper, we propose a Blockchain-based security management framework for CPS to address the security challenges of CPS. The framework leverages the unique features of Blockchain, such as decentralization, immutability, and transparency, to ensure the integrity and confidentiality of data exchanged among devices. We describe the architecture of the proposed framework and explain its key components, including Blockchain nodes, smart contracts, and security policies. Our work contributes to the field of CPS security by providing a novel approach to secure CPS using Blockchain technology. The proposed framework can be applied to various domains, including smart cities, healthcare, and industrial control systems. The framework can also be extended to handle more advanced security scenarios, such as dynamic security policies and real-time threat detection. In this framework, we build a decentralized controller-to-controller (C2C) communication system to make CPS more scalable, secure, private, and reliable. We make a miner-as-a-controller selection algorithm to choose a miner controller for a given session.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116013434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Martin Straesser, Stefan Geissler, T. Hossfeld, Samuel Kounev
{"title":"Power to the Applications: The Vision of Continuous Decentralized Autoscaling","authors":"Martin Straesser, Stefan Geissler, T. Hossfeld, Samuel Kounev","doi":"10.1109/CCGridW59191.2023.00058","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00058","url":null,"abstract":"Autoscaling has been one of the most active research areas since the beginning of the cloud computing era. Nearly all previously proposed approaches focus on decision-making based on averaged monitoring values of many service instances at fixed points in time. This limits responsiveness and can lead to service level objective (SLO) violations when the load suddenly increases. Our vision of continuous decentralized autoscaling avoids these issues by giving individual service instances the power to make scaling decisions in a distributed fashion. Each instance performs self-monitoring and evaluates its state. The service instance initiates upscaling if it detects an overload or downscaling if its load is below a specified threshold. By randomly determining scaling timing, we achieve quasi-continuous scaling behavior when multiple service instances are deployed. We discuss challenges regarding analytical modeling, simulation, and real-world evaluation of this approach.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116477683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On the timing of applying resource separation during DDoS attacks","authors":"Anmol Kumar, G. Somani","doi":"10.1109/CCGridW59191.2023.00036","DOIUrl":"https://doi.org/10.1109/CCGridW59191.2023.00036","url":null,"abstract":"From the past few years, DDoS attack incidents are continuously rising across the world. Most of these DDoS attacks lasted lesser than 10 minutes time. In the state-of-the-art solutions, we see a class of methods which advocate for the separation of resources to minimize the attack driven resource contention. Resource separation requires additional resources to implement the DDoS defense. In this work, we address an important practical aspect related to the timing of separation. For this, we experimentally compare five scenarios in which different types of separation methods are applied at different timings. We observe that the “always ON” strategy is best among all giving minimum response time and no request failures.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125049048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}