Download presentation
Presentation is loading. Please wait.
Published byPriscilla Joseph Modified over 9 years ago
1
IM&T Vacation Program Benjamin Meyer Virtualisation and Hyper-Threading in Scientific Computing
2
HPC Clusters A large set of connected computers Used for computation intensive workloads, rather than I/O orientated operations Each node runs own instance of OS http://www.redbooks.ibm.com/redbooks/pdfs/sg247287.pdf
3
CSIRO’s Bragg Cluster 128 compute nodes with 16 CPUs each 2048 cores in total 128GB of RAM per node 384 Fermi Tesla M2050 GPUs 172,032 streaming cores
4
What is Virtualisation?
5
Hypervisor Software which allows different and multiple operating systems to run on the underlying hardware Ensures all privileged operations are appropriately handled to maintain system integrity Invisible to operating system OS thinks it is running natively VMware ESXi Hypervisor used for this project
6
Benefits: Heterogeneous Clusters
7
Benefits: Live Migration Running jobs can be moved to other hardware Allows dynamic scheduling Preemptive failure/down time evasion
8
Checkpointing Status of OS, application and memory are saved at intervals Allows for easy failure recovery Software debugging Clean compute Security Run time/failure isolation Clean start Benefits
9
Performance Comparison Floating point operations per second
10
Performance Comparison Updates to random memory locations per second GUPs 100% 87.1% 54.8%
11
Performance Comparison MPI (message passing) latency
12
Hyper-Threading
13
Thread 1 Thread 2 Thread 1 Thread 2 Physical Cores Logical Cores (seen by OS) Thread 1 Thread 2 Physical Cores non Hyper-Threaded Hyper-Threaded Time Resource 1 Resource 2 Resource 3 Resource 4 Time Resource 1 Resource 2 Resource 3 Resource 4 Hyper-Threading Example
14
Performance Comparison Floating point operations per second
15
Performance Comparison Updates to random memory locations per second
17
References Tim Ho (2012, Nov.). CSIRO Advanced Scientific Computing User Manual [Online]. Available: https://wiki.csiro.au/display/ASC/ASC+Homepage https://wiki.csiro.au/display/ASC/ASC+Homepage (2013, Jan.). Top500 HPC Statistics [Online]. Available: http://www.top500.org/statistics/overtime/ http://www.top500.org/statistics/overtime/ (2012, Oct.). IBM Blue Gene #1 in Supercomputing [Online]. Available: http://www.03.ibm.com/systems/technicalcomputing/solutions/bluegene/index.html http://www.03.ibm.com/systems/technicalcomputing/solutions/bluegene/index.html (2013). Virtualize for Efficiency, Higher Availability and Lower Costs [Online]. Available: http://www.vmware.com/virtualization/virtualization-basics/virtualization-benefits.html http://www.vmware.com/virtualization/virtualization-basics/virtualization-benefits.html (2012). Tuning a Linux HPC Cluster: HPC Challenge [Online]. Available: http://www.ibm.com/support/publications/us/library/ http://www.ibm.com/support/publications/us/library/
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.