Distributed and Parallel Systems

,

Paper accepted: SimLess: Simulate Serverless Workflows and Their Twins and Siblings in Federated FaaS @2022 ACM Symposium on Cloud Computing

Titel: SimLess: Simulate Serverless Workflows and Their Twins and Siblings in Federated FaaS

Authors: Sashko Ristov, Mika Hautz, Christian Hollaus, Radu Prodan

2022 ACM Symposium on Cloud Computing

Abstract: Many researchers migrate scientific serverless workflows or function choreographies (FC) on Function-as-a-Service (FaaS) to benefit from its high scalability and elasticity. Unfortunately, the heterogeneous nature of federated FaaS hampers decisions on the most appropriate configuration setup to run FCs. Consequently, scientists must choose between accurate but tedious and expensive experiments or simple but cheap but less accurate simulations. Unfortunately, related work mainly supports either simulation models for serverfull workflow applications that run on virtual machines and containers or partial FaaS models for individual serverless functions that focus on execution time and neglect various kinds of federated FaaS overheads. Therefore, this paper introduces SimLess, an FC simulation framework across multiple FaaS providers to achieve accurate FC simulations with a simple and cheap parameter setup. Unlike the costly approaches that use machine learning over time series to predict the behavior of FCs, SimLess introduces two light concepts: (1) twins, representing the same code deployed with the same computing, communication, and storage resources, but in other cloud regions of the same FaaS provider, and (2) siblings, representing the same code deployed in the same region with different computing resources. The novel SimLess simulation model splits the round trip time of a function into several parameters reused among twins and siblings without running them. We evaluated SimLess with two scientific FCs deployed across 18 AWS, Google, and IBM regions. SimLess simulates the cumulative overhead with an average inaccuracy of 8.9% without significant differences between regions for learning and validation. Moreover, SimLess generates an inaccuracy of up to 9.75% for a low concurrency FC executed on a single region, with high concurrency of 2500 functions executed in other regions. Finally, SimLess reduces the parameter setup cost by 77.23% compared to the existing simulation approaches.

, ,

Narges Mehran got the student award to present the paper with the title of “Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum” at IEEE Cluster 2022.



Student travel award at IEEE Cluster 2022

Narges Mehran got the student award for presenting the paper titled “Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum” at IEEE Cluster 2022.

The presentation was on the 7th of September: https://clustercomp.org/2022/program/

,

Paper accepted: Towards Extreme and Sustainable Graph Processing for Urgent Societal Challenges in Europe @ IEEE Cloud Summit 2022

IEEE Cloud Summit 2022, https://www.ieeecloudsummit.org/

Authors: Radu Prodan, Dragi Kimovski, Andrea Bartolini, Michael Cochez,
Alexandru Iosup, Evgeny Kharlamov, Joze Rozanec, Laurentiu Vasiliu, Ana
Lucia Varbanescu

Abstract: The Graph-Massivizer project, funded by the Horizon Europe research and innovation program, researches and develops a high-performance, scalable, and sustainable platform for information processing and reasoning based on the massive graph (MG) representation of extreme data. It delivers a toolkit of five open-source software tools and FAIR graph datasets covering the sustainable lifecycle of processing extreme data as MGs. The tools focus on holistic usability (from extreme data ingestion and MG creation), automated intelligence (through analytics and reasoning), performance modelling, and environmental sustainability tradeoffs, supported by credible data-driven evidence across the computing continuum. The automated operation based on the emerging serverless computing paradigm supports experienced and novice stakeholders from a broad group of large and small organisations to capitalise on extreme data through MG programming and processing.

Graph-Massivizer validates its innovation on four complementary use cases considering their extreme data properties and coverage of the three sustainability pillars (economy, society, and environment): sustainable green finance, global environment protection foresight, green AI for the sustainable automotive industry, and data centre digital twin for exascale computing. Graph-Massivizer promises 70% more efficient analytics than AliGraph, and 30% improved energy awareness for ETL storage operations than Amazon Redshift. Furthermore, it aims to demonstrate a possible two-fold improvement in data centre energy efficiency and over 25% lower greenhouse gas emissions for basic graph operations.

,

Radu Prodan, Sashko Ristov and Andrei Amza participated in the CardioHPC project meeting

From August 16.-19.2022, a CardioHPC project meeting took place in Skopje, Macedonia. Radu Prodan, Andrei Amza and Sahsko Ristvo participated for AAU.

,

STSM-visit of Hamza Baniata @ ITEC

During the period Aug 1st –26th, 2022, Hamza Baniata, a PhD Candidate at the Department of Computer Science, University of Szeged, Hungary, has visited the institute of Information Technology of the University of Klagenfurt, Austria. Under the collaborative supervision by Prof.
Attila Kertesz (SZTE) and Prof. Radu Prodan (ITEC), Hamza has performed several research activities related to the simulation of Blockchain and Fog Computing applications, the enhancement of the FoBSim simulation tool, and the integration of Machine Learning with Blockchain technology. The visit was encouraged and funded by the European COST program under action identifier CA19135 (CERCIRAS), in which Attila, Radu and Hamza are active members. The scientific results of this research visit are currently being edited and finalized in order to be disseminated in an international scientific conference.

,

Coordination @ Horizon Europe “Massive Graph Processing of Extreme Data for a Sustainable Economy, Society, and Environment” (Graph-Massivizer) project accepted

Project lead/coordination: Radu Prodan
Project partners: IDC Italia, Peracton Limited, Institut Jozef Stefan, Sintef, Universiteit Twente, metaphacts GmbH, Vrije Universiteit Amsterdam, Cineca, Event Registry, Università di Bologna, Robert Bosch GmbH

Abstract: Graph-Massivizer researches and develops a high-performance, scalable, and sustainable platform for information processing and reasoning based on the massive graph representation of extreme data. It delivers a toolkit of five open-source software tools and FAIR graph datasets covering the sustainable lifecycle of processing extreme data as massive graphs. The tools focus on holistic usability (from extreme data ingestion and massive graph creation), automated intelligence (through analytics and reasoning), performance modelling, and environmental sustainability tradeoffs, supported by credible data-driven evidence across the computing continuum. The automated operation based on the emerging serverless computing paradigm supports experienced and novice stakeholders from a broad group of large and small organisations to capitalise on extreme data through massive graph programming and processing. Graph Massivizer validates its innovation on four complementary use cases considering their extreme data properties and coverage of the three sustainability pillars (economy, society, and environment): sustainable green finance, global environment protection foresight, green AI for the sustainable automotive industry, and data centre digital twin for exascale computing. Graph Massivizer promises 70% more efficient analytics than AliGraph, and 30% improved energy awareness for ETL storage operations than Amazon Redshift. Furthermore, it aims to demonstrate a possible two-fold improvement in data centre energy efficiency and over 25% lower GHG emissions for basic graph operations. Graph-Massivizer gathers an interdisciplinary group of twelve partners from eight countries, covering four academic universities, two applied research centres, one HPC centre, two SMEs and two large enterprises. It leverages world-leading roles of European researchers in graph processing and serverless computing and uses leadership-class European infrastructure in the computing continuum.

PHD candidate Kevin Theuermann defended his theses on 12th of July 2022: Trustworthy Service Composition Systems for E-Government

Kevin Theuermann (TU Graz) defended his theses on July 12, 2022.
Title: Trustworthy Service Composition Systems for E-Government
Prof. Radu Prodan was integrated as an external expert to survey the thesis and discuss it with the candidate as part of the defense.

,

Paper Accepted @ Journal of Ambient Intelligence and Humanized Computing

Title: MCred: Multi-Modal Message Credibility for Fake News Detection using BERT and CNN

Journal: Journal of Ambient Intelligence and Humanized Computing

Authors: Pawan Kumar Verma, Prateek Agrawal, Vishu Madaan, Radu Prodan

Abstract:

Online social media enables low cost, easy access, rapid propagation, and easy communication of information, including spreading low-quality fake news. Fake news has become a huge threat to every sector in society, and resulting in decrements in the trust quotient for media and leading the audience into bewilderment. In this paper, we proposed a new framework called Message Credibility (MCred) for fake news detection that utilizes the benefits of local and global text semantics. This framework is the fusion of Bidirectional Encoder Representations from Transformers (BERT) using the relationship between words in sentences for global text semantics and Convolutional Neural Networks (CNN) using N-gram features for local text semantics. We demonstrate through experimental results a popular Kaggle dataset that MCred improves the accuracy over a state-of-the-art model by 1.10%, thanks to its combination of local and global text semantics.

,

Paper accepted at IEEE Cluster 2022 Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum

Title: Matching-based Scheduling of Asynchronous Data Processing Workflows on the Computing Continuum

Heidelberg, Germany | September 6-9, 2022

https://clustercomp.org/2022/

Authors: Narges Mehran, Zahra Najafabadi Samani, Dragi Kimovski, Radu Prodan

Abstract: Today’s distributed computing infrastructures encompass complex workflows for real-time data gathering, transferring, storage, and processing, quickly overwhelming centralized cloud centers. Recently, the computing continuum that federates the Cloud services with emerging Fog and Edge devices represents a relevant alternative for supporting the next-generation data processing workflows. However, eminent challenges in automating data processing across the computing continuum still exist, such as scheduling heterogeneous devices across the Cloud, Fog, and Edge layers. We propose a new scheduling algorithm called C3-MATCH, based on matching theory principles, involving two sets of players negotiating different utility functions: 1) workflow microservices that prefer computing devices with lower data processing and queuing times; 2) computing continuum devices that prefer microservices with corresponding resource requirements and less data transmission time. We evaluate C3-MATCH using real-world road sign inspection and sentiment analysis workflows on a federated computing continuum across four Cloud, Fog, and Edge providers. Our combined simulation and real execution results reveal that C3-MATCH achieves up to 67% lower completion time compared to three state-of-the-art methods.

Paper Accepted @ The Journal of Supercomputing

Title: An Elastic and Traffic-Aware Scheduler for Distributed Data Stream Processing in Heterogeneous Clusters

Authors: Hamid Hadian, Mohammadreza Farrokh, Mohsen Sharifi, and Ali Jafari

Abstract:

Existing Data Stream Processing (DSP) systems perform poorly while encountering heavy workloads, particularly on clustered set of (heterogeneous) computers. Elasticity and changing application parallelism degree can limit the performance degradation in the face of varying workloads that negatively impact the overall application response time. Elasticity can be achieved by operator scaling, i.e., by replication and relocation of operators at runtime. However, scaling decisions at runtime is challenging, since it first increases the overall communication overhead between operators and secondly changes any initial scheduling that could lead to a non-optimal scheduling plan. In this paper, we investigate the problem of elasticity and scaling decisions and propose a DSP system called ER-Storm. To curb communication overhead, we propose a new 3-step mechanism for replication and relocation of operators upon detecting a bottleneck operator that overutilizes a worker node. The other challenge is to select the proper worker nodes to host relocated operators. By discretizing the input workload, we model the relocation of operators between worker nodes at runtime through a scalable Markov Decision Process (MDP) and use a model-free notion of reinforcement learning (Q-Learning) to find optimal solutions. We have implemented our propositions on the Apache Storm version 2.1.0. Our experimental results show that ER-Storm reduces the average topology response time by 20 to 60 percent based on the rate of input workload (low or high) compared to the R-Storm scheduler and the Online-Scheduler of Storm.