Cargando…
Dynamic scheduling of virtual machines running hpc workloads in scientific grids
The primary motivation for uptake of virtualization has been resource isolation, capacity management and resource customization allowing resource providers to consolidate their resources in virtual machines. Various approaches have been taken to integrate virtualization in to scientific Grids especi...
Autores principales: | , , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2010
|
Materias: | |
Acceso en línea: | https://dx.doi.org/10.1145/1330555.1330556 http://cds.cern.ch/record/1294646 |
_version_ | 1780920861993205760 |
---|---|
author | Khalid, Omer Maljevic, Ivo Anthony, Richard Petridis, Miltos Parrot, Kevin Schulz, Markus |
author_facet | Khalid, Omer Maljevic, Ivo Anthony, Richard Petridis, Miltos Parrot, Kevin Schulz, Markus |
author_sort | Khalid, Omer |
collection | CERN |
description | The primary motivation for uptake of virtualization has been resource isolation, capacity management and resource customization allowing resource providers to consolidate their resources in virtual machines. Various approaches have been taken to integrate virtualization in to scientific Grids especially in the arena of High Performance Computing (HPC) to run grid jobs in virtual machines, thus enabling better provisioning of the underlying resources and customization of the execution environment on runtime. Despite the gains, virtualization layer also incur a performance penalty and its not very well understood that how such an overhead will impact the performance of systems where jobs are scheduled with tight deadlines. Since this overhead varies the types of workload whether they are memory intensive, CPU intensive or network I/O bound, and could lead to unpredictable deadline estimation for the running jobs in the system. In our study, we have attempted to tackle this problem by developing an intelligent scheduling technique for virtual machines which monitors the workload types and deadlines, and calculate the system over head in real time to maximize number of jobs finishing within their agreed deadlines. |
id | cern-1294646 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2010 |
record_format | invenio |
spelling | cern-12946462023-03-15T19:12:02Zdoi:10.1145/1330555.1330556http://cds.cern.ch/record/1294646engKhalid, OmerMaljevic, IvoAnthony, RichardPetridis, MiltosParrot, KevinSchulz, MarkusDynamic scheduling of virtual machines running hpc workloads in scientific gridsComputing and ComputersThe primary motivation for uptake of virtualization has been resource isolation, capacity management and resource customization allowing resource providers to consolidate their resources in virtual machines. Various approaches have been taken to integrate virtualization in to scientific Grids especially in the arena of High Performance Computing (HPC) to run grid jobs in virtual machines, thus enabling better provisioning of the underlying resources and customization of the execution environment on runtime. Despite the gains, virtualization layer also incur a performance penalty and its not very well understood that how such an overhead will impact the performance of systems where jobs are scheduled with tight deadlines. Since this overhead varies the types of workload whether they are memory intensive, CPU intensive or network I/O bound, and could lead to unpredictable deadline estimation for the running jobs in the system. In our study, we have attempted to tackle this problem by developing an intelligent scheduling technique for virtual machines which monitors the workload types and deadlines, and calculate the system over head in real time to maximize number of jobs finishing within their agreed deadlines.The primary motivation for uptake of virtualization has been resource isolation, capacity management and resource customization allowing resource providers to consolidate their resources in virtual machines. Various approaches have been taken to integrate virtualization in to scientific Grids especially in the arena of High Performance Computing (HPC) to run grid jobs in virtual machines, thus enabling better provisioning of the underlying resources and customization of the execution environment on runtime. Despite the gains, virtualization layer also incur a performance penalty and its not very well understood that how such an overhead will impact the performance of systems where jobs are scheduled with tight deadlines. Since this overhead varies the types of workload whether they are memory intensive, CPU intensive or network I/O bound, and could lead to unpredictable deadline estimation for the running jobs in the system. In our study, we have attempted to tackle this problem by developing an intelligent scheduling technique for virtual machines which monitors the workload types and deadlines, and calculate the system over head in real time to maximize number of jobs finishing within their agreed deadlines.arXiv:1009.4841oai:cds.cern.ch:12946462010-09-27 |
spellingShingle | Computing and Computers Khalid, Omer Maljevic, Ivo Anthony, Richard Petridis, Miltos Parrot, Kevin Schulz, Markus Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title | Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title_full | Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title_fullStr | Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title_full_unstemmed | Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title_short | Dynamic scheduling of virtual machines running hpc workloads in scientific grids |
title_sort | dynamic scheduling of virtual machines running hpc workloads in scientific grids |
topic | Computing and Computers |
url | https://dx.doi.org/10.1145/1330555.1330556 http://cds.cern.ch/record/1294646 |
work_keys_str_mv | AT khalidomer dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids AT maljevicivo dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids AT anthonyrichard dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids AT petridismiltos dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids AT parrotkevin dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids AT schulzmarkus dynamicschedulingofvirtualmachinesrunninghpcworkloadsinscientificgrids |