Below is the architecture model of our computation and storage resources. Currently in the cluster room we have 3 HPC clusters (a.k.a. Supercomputers), called Hades, Lucifer and Baal, that are composed as following:
Because it will not do any good to compute without storage capacity, following storage volumes are also available:
Except for *ibpc_team and /scratch volumes, all above-mentioned volumes are replicated and/or distributed on a storage cluster currently composed by 11 servers.
/scratch volume on each node serves only to store temporary computing files.So, because it's cleared every night (deleting old job's directories), you should not try to use it for chaining jobs.
In order to get the best performance as possible, we choose a high throughput and low latency network technology: Infiniband QDR (40Gbs).
Below, the evolution year-per-year of the computing performance and storage: