How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster
The 6th Workshop on Hot Topics in Cloud Computing Performance (HotCloudPerf 2023)
Laurens Versluis
Ph.D. student, Vrije Universiteit Amsterdam
Tech Lead MagnaData project
Research Focus
Scheduling complex data-driven workflows in datacenters
Contact
Links
Publications
Supervised
Talks
Teaching
Publications
PDF
Cite
Less is not more: We need rich datasets to explore
Future Generation Computer Systems
PDF
Cite
TaskFlow: An Energy- and Makespan-Aware Task Placement Policy for Workflow Scheduling through Delay Management
ICPE HotCloudPerf 2022
PDF
Cite
Empirical Characterization of User Reports about Cloud Failures
ACSOS 2021
PDF
Cite
A survey of domains in workflow scheduling in computing infrastructures: Community and keyword analysis, emerging trends, and taxonomies
Future Generation Computer Systems (FGCS), Volume 123, October 2021
OpenDC 2.0: Convenient Modeling and Simulation of Emerging Technologies in Cloud Datacenters
CCGrid 2021
Methodological Principles for Reproducible Performance Evaluation in Cloud Computing
IEEE Transactions on Software Engineering (TSE)
PDF
Cite
The Workflow Trace Archive: Open-Access Data from Public and Private Computing Infrastructures
IEEE Transactions on Parallel and Distributed Systems (TPDS) 2020
PDF
Cite
The SPEC-RG Reference Architecture for FaaS: From Microservices and Containers to Serverless Platforms
IEEE Internet Computing
PDF
Cite
Chamulteon: Coordinated Auto-Scaling of Micro-Services
IEEE ICDCS 2019 workshop program
PDF
Cite
The AtLarge Vision on the Design of Distributed Systems and Ecosystems.
IEEE International Conference on Distributed Computing Systems (ICDCS 2019)
PDF
Cite
A Reference Architecture for Datacenter Scheduling
International Conference for High Performance Computing, Networking, Storage and Analysis 2018 (SC18)
PDF
Cite
Serverless is More: From PaaS to Present Cloud Computing
IEEE Internet Computing, Sep/Oct edition, 2018
PDF
Cite
An Analysis of Workflow Formalisms for Workflows with Complex Non-Functional Requirements
HotCloudPerf 2018
PDF
Cite
A Trace-Based Performance Study of Autoscaling Workloads of Workflows in Datacenters
International Symposium on Cluster, Cloud and Grid Computing 2018
PDF
Cite
Massivizing Computer Systems: a Vision to Understand, Design, and Engineer Computer Ecosystems through and beyond Modern Distributed Systems
International Conference on Distributed Computing Systems 2018
PDF
Cite
The OpenDC Vision: Towards Collaborative Datacenter Simulation and Exploration for Everybody.
ISPDC 2017 (invited paper)
PDF
Cite
PDF
Cite
Less is not more: We need rich datasets to explore
Future Generation Computer Systems
PDF
Cite
TaskFlow: An Energy- and Makespan-Aware Task Placement Policy for Workflow Scheduling through Delay Management
ICPE HotCloudPerf 2022
PDF
Cite
Empirical Characterization of User Reports about Cloud Failures
ACSOS 2021
PDF
Cite
A survey of domains in workflow scheduling in computing infrastructures: Community and keyword analysis, emerging trends, and taxonomies
Future Generation Computer Systems (FGCS), Volume 123, October 2021
OpenDC 2.0: Convenient Modeling and Simulation of Emerging Technologies in Cloud Datacenters
CCGrid 2021
Methodological Principles for Reproducible Performance Evaluation in Cloud Computing
IEEE Transactions on Software Engineering (TSE)
PDF
Cite
The Workflow Trace Archive: Open-Access Data from Public and Private Computing Infrastructures
IEEE Transactions on Parallel and Distributed Systems (TPDS) 2020
PDF
Cite
The SPEC-RG Reference Architecture for FaaS: From Microservices and Containers to Serverless Platforms
IEEE Internet Computing
PDF
Cite
Chamulteon: Coordinated Auto-Scaling of Micro-Services
IEEE ICDCS 2019 workshop program
PDF
Cite
The AtLarge Vision on the Design of Distributed Systems and Ecosystems.
IEEE International Conference on Distributed Computing Systems (ICDCS 2019)
PDF
Cite
A Reference Architecture for Datacenter Scheduling
International Conference for High Performance Computing, Networking, Storage and Analysis 2018 (SC18)
PDF
Cite
Serverless is More: From PaaS to Present Cloud Computing
IEEE Internet Computing, Sep/Oct edition, 2018
PDF
Cite
An Analysis of Workflow Formalisms for Workflows with Complex Non-Functional Requirements
HotCloudPerf 2018
PDF
Cite
A Trace-Based Performance Study of Autoscaling Workloads of Workflows in Datacenters
International Symposium on Cluster, Cloud and Grid Computing 2018
PDF
Cite
Massivizing Computer Systems: a Vision to Understand, Design, and Engineer Computer Ecosystems through and beyond Modern Distributed Systems
International Conference on Distributed Computing Systems 2018
PDF
Cite
The OpenDC Vision: Towards Collaborative Datacenter Simulation and Exploration for Everybody.
ISPDC 2017 (invited paper)
PDF
Cite
Supervised Bachelor Theses
A Trace-Based Validation Study of OpenDC
A Trace-Based Validation Study of OpenDC
Talks
TaskFlow: An Energy- and Makespan-Aware Task Placement Policy for Workflow Scheduling through Delay Management
HotCloudPerf 2022
Methodological Principles for Reproducible Performance Evaluation in Cloud Computing
CompSys 2019
An Analysis of Workflow Formalisms for Workflows with Complex Non-Functional Requirements
HotCloudPerf 2018
A Short Introduction to Cloud Computing
VU HPC workshop
A Trace-Based Performance Study of Autoscaling Workloads of Workflows in Datacenters
CCGrid 2018, 1-4 May, Washington D.C.
ANANKE: a Q-Learning-Based Portfolio Scheduler for Complex Industrial Workflows
ICAC 2017
TaskFlow: An Energy- and Makespan-Aware Task Placement Policy for Workflow Scheduling through Delay Management
HotCloudPerf 2022
Methodological Principles for Reproducible Performance Evaluation in Cloud Computing
CompSys 2019
An Analysis of Workflow Formalisms for Workflows with Complex Non-Functional Requirements
HotCloudPerf 2018
A Short Introduction to Cloud Computing
VU HPC workshop
A Trace-Based Performance Study of Autoscaling Workloads of Workflows in Datacenters
CCGrid 2018, 1-4 May, Washington D.C.
ANANKE: a Q-Learning-Based Portfolio Scheduler for Complex Industrial Workflows
ICAC 2017
Teaching
Distributed Systems, MSc, VU (2017-ongoing)
Distributed Systems, MSc, VU (2017-ongoing)