top of page

What drives Disaggregated Computing?

Updated: Oct 1, 2022

The separation of the physical computer systems responsible for performing computational work (collectively known as “compute nodes”) and those responsible for storing digital data (“storage nodes”) is a common architecture for big data applications in large-scale deployments in enterprises and in public clouds. This deployment model enables the independent provisioning, scaling, and upgrading of compute clusters and storage clusters. Compute clusters may be created on-demand, as additions and changes may be made to the number of physical computer systems constituting nodes of the cluster (this flexibility is termed “elastic scaling”). In particular, nodes of a cluster may be transient in that they may be made available for inclusion in the cluster by a third-party only for a limited time, and only a short programmatic advance warning of their unavailability (for example, thirty seconds) may be given. An example of a transient node is Amazon’s EC2 Spot Instance.

Direct attached storage is not useful for any form of permanent or intermediate data due to the transient nature of deployed clusters and due to interruption of nodes, but can be useful for caching to exploit locality. Since clusters are transient permanent data would have to be stored outside the cluster in a shared repository, though can be cached in DRAM and local SSD storage. Intermediate data is also a problem, due to the failure of node or spot instance interruptions. One would have to replicate the data outside the cluster for recovery, though local caching in DRAM and SSD is useful.



bottom of page