Home > Storage > PowerScale (Isilon) > Product Documentation > Storage (general) > PowerScale OneFS Operating System > Performance
A large-scale storage system must provide the performance required for a variety of workflows, whether they be sequential, concurrent, or random. Different workflows will exist between applications and within individual applications. OneFS provides for all of these needs simultaneously with intelligent software. More importantly, with OneFS (see 0), throughput and IOPS scale linearly with the number of nodes present in a single system. Due to balanced data distribution, automatic rebalancing, and distributed processing, OneFS is able to leverage additional CPUs, network ports, and memory as the system scales.
To fully exploit locality and meet the needs of various workflows, OneFS provides a globally accessible and coherent cache across all nodes. Storage nodes can currently utilize up to 736 GB of RAM each, allowing a OneFS powered cluster to contain up to 181 TB of system memory (252 nodes). This memory is primarily used to cache data that has been placed on that particular storage node and is actively being accessed. This cache grows as more nodes are added to a cluster, allowing an increasing working set to continually remain in cache. In addition, various SSD configurations are also available for additional read caching. OneFS also allows the storage system administrator to specify the type of workload on a per-file or per-directory basis. It also indicates whether the access pattern to a particular file or directory is random, concurrent, or sequential. This unique capability allows OneFS to tailor on-disk layout decisions, cache-retention policies, and data prefetch policies to maximize performance of individual workflows.