Can HPC workloads be moved to the Cloud? Does it make sense to move them? There is not a single answer to this question.
Each workload presents different technical attributes (parallelism of application code, need for very fast access to large volumes of datas, capability to leverage accelerators and GPUs…), usage attributes (continuous base line, peak loads), user and resources geographical distribution, use of licensed software, internal HPC skills….
Our methodologies and tools integrates these factors to define workload “cloudability” and recommend the best strategic options through our “Workcloud Quadrant”.