Storage Clustering and Disaggregation

To increase cost-effectiveness and scale at a finer granularity, the hardware trend is heading toward disaggregated storage and computation architecture. Generally, this means that compute machines store and access temporary/permanent data on storage machines over the network. Disaggregated storage has a number of advantages over collocated storage, which limits each node to the space available on its local disks. It enables the separation of computation and storage resources within a computer server without affecting physical connectivity.


Disaggregated storage connects resources via a network fabric, giving you more flexibility when upgrading, replacing, or adding resources. It also enables servers to be built for future expansion, providing more storage efficiency, scale, and performance than traditional data storage without sacrificing throughput or latency. Disaggregated storage is a type of scale-out storage that consists of a group of storage devices that act as a logical pool of storage that can be allocated to any server on the network via a high-speed network fabric.


The performance of local storage is combined with the flexibility of storage area networks in disaggregated storage. Modern storage clusters are moving forward to meet 100Gbps network infrastructure as the demand for data access performance and latency grows. High-performance links between storage nodes are a key component for building a disaggregated storage cluster solution. With this saying, RoCE V2 is a technology that enables data movement between servers providing both flexibility and performance at scale.