Download as PDF, PPTX
Uploaded byIan Lumb
High Performance Computing in the Cloud?
High performance computing (HPC) in the cloud is increasingly viable for various applications, particularly as the HPC-community develops strategies to mitigate latency issues. Notably, the integration of containerization and an emphasis on cloud-native architectures are emerging trends. However, challenges remain for traditional MPI applications in cloud environments, prompting exploration of alternative solutions.
More Related Content
Similar to High Performance Computing in the Cloud?
High Performance Computing in the Cloud?
- 1. Abstract High Performance Computingin the Cloud is viable in numerous use cases. Common to all successful use cases for cloud-based HPC is the ability embrace latency. Not surprisingly then, early successes were achieved with embarrassingly parallel HPC applications involving minimal amounts of data - in other words, there was little or no latency to be hidden. Over the fulness of time, however, the HPC-cloud community has become increasingly adept in its ability to ‘hide’ latency and, in the process, support increasingly more sophisticated HPC use cases in public and private clouds. Real-world use cases, deemed relevant to remote sensing, will illustrate aspects of these sophistications for hiding latency in accounting for large volumes of data, the need to pass messages between simultaneously executing components of distributed-memory parallel applications, as well as (processing) workflows/pipelines. Finally, the impact of containerizing HPC for the cloud will be considered through the relatively recent creation of the Cloud Native Computing Foundation.
- 2. High Performance Computing inthe Cloud? Ian Lumb ilumb@univa.com Ontario Association of Remote Sensing (O.A.R.S) Ryerson University, Toronto - November 10, 2015
- 4.
- 7.
- 10.
- 15. Latency is physicallya consequence of the limited velocity with which any physical interaction can propagate. https://en.wikipedia.org/wiki/Latency_(engineering)
- 16.
- 17. Current consumer deviceshave appallingly bad latency … It's the Latency, Stupid https://rescomp.stanford.edu/~cheshire/rants/Latency.html
- 19.
- 20.
- 22. … if youhave a network link with low bandwidth then it's an easy matter of putting several in parallel to make a combined link with higher bandwidth, but if you have a network link with bad latency then no amount of money can turn any number of them into a link with good latency. It's the Latency, Stupid https://rescomp.stanford.edu/~cheshire/rants/Latency.html
- 23. Latency is INEVITABLE!!! … andthe best we can do? Try to ‘HIDE’ it!
- 24. Latency is INEVITABLE!!! … andthe best we can do? Try to ‘HIDE’ it!
- 26. Definitions ● Public cloud ○Off-premise IT capabilities or applications, provided by others ● Private cloud ○ On-premise enablement of cloud capabilities with existing IT ● Hybrid cloud ○ Some combination of public and private clouds
- 28.
- 29. GPUs in theCloud? The Top Four Reasons 1. You can realize possibilities using the cloud a. You can scale up and scale out 2. You still realize the promise of GPU programmability a. … via HPC in the cloud 3. Your use of the cloud is transparent a. You’ve found ways to `hide’ latency i. Constraints apply for MPI apps 4. Your go-to apps still work in the cloud http://info.brightcomputing.com/Blog/bid/196290/The-Top-4-Reasons-You-Should-Try-Cloud-Based-GPUs-for-HPC
- 30.
- 31.
- 32.
- 34. HPC as aContainerized Cloud Based Service http://insidehpc.com/2015/11/ubercloud-delivers-cae-as-a-service-with-univa-grid-engine-container-edition/
- 36.
- 37.
- 41. Cloud Native ComputingFoundation (CNCF) ● For current applications and services ○ Uptake of cloud computing remains an afterthought from a systems-architecture perspective ● CNCF aims to introduce a cloud-native paradigm shift that emphasizes: ○ Containerization ○ Dynamic scheduling ○ Orientation around micro services ● Making use of Kubernetes as a ‘seed technology’ ○ #1 priority: Integrate the orchestration layer of the container ecosystem ● Univa is a Founding Member ○ Along with Google, IBM, Intel, Red Hat and numerous others ... https://cncf.io/
- 42.
- 43. MPI Apps Remaina Challenge ● … for ○ cloud use ○ containerization ● Constrain MPI apps to mitigate concerns with latency ○ Run HPC on-premise OR in a cloud, but not between ○ Containers? ■ Just say no??? ● Seek alternatives ○ Apache Spark ??? ○ Message busses ???
- 45. “The wonderful thing about standardsis that there are so many of them to choose from.” https://en.wikiquote.org/wiki/Grace_Hopper
- 46. Cloud Computing is bereftof standards!!!
- 47. Cloud Computing is bereftof standards!!! ...but, FLUSH with implementations!!!
- 48. Latency is INEVITABLE!!! … andthe best we can do? Try to ‘HIDE’ it!