Tag: hpc in the cloud
Many scientific applications can require significant coordination between large numbers of nodes. Massive effort is spent in the HPC arena to hide much of the latency from coordination with expensive low-latency networks and fine-tuned communication libraries. Such efforts have not yet been translated to the commercial cloud computing arena, which still typically provide systems with varying amounts of installed memory but rarely various quality interconnects.
Substantial benefits can be provided to the HPC cloud community if a high-performance network provisioning solution can be devised. In the light of his invited talk at ISC Cloud’10 in October this year in Frankfurt, Wolfgang Gentzsch asks Gilad Shainer from Mellanox for some insights about the cloud networking issue.
The announcement this morning that Amazon is offering Cluster Compute Instances for EC2 specifically for the needs of HPC users might just be that long-awaited game-changer when it comes to the viability of scientific computing in the public cloud. While it is fresh from a private beta and the results are promising, only time will tell to what degree users will snatch up this opportunity to have supercomputing power on demand.
Coming up with a list of the top ten threats and concerns in terms of security for HPC in the cloud is the easy part. However, putting that list in an organization-specific order is an exercise in prioritizing issues of not only security, but performance, applications in general, and overall goals.
When we start considering cloud, the primary driver seems to be economic (cost), thus we need to make sure we address any cost-related barriers associated with adoption of cloud as well as ensure that our expectations are honest and appropriate.
Today’s entry contains feedback on the present and future of HPC in the cloud from from Dr. Ignacio M. Llorente, Professor in Computer Architecture and Technology, and the Head of the Distributed Systems Architecture Research Group at Complutense University of Madrid.