Home > Storage > PowerScale (Isilon) > Product Documentation > Storage (general) > PowerScale OneFS Operating System > Scalability
Traditional storage systems that have a finite maximum size and must be replaced by a bigger storage array when the maximum performance or capacity is reached. In contrast, a OneFS powered cluster can linearly expand, or scale out, performance, capacity, or both, seamlessly increasing the existing file system or volume into petabytes of capacity. In addition, with the flexibility of OneFS, different node types can be mixed in a single cluster or pool by using OneFS SmartPools software. The automated tiered storage capability of SmartPools provides added flexibility and eliminates the need for “forklift” upgrades when different capacity or performance levels are needed. SmartPools (see 0) enables businesses and storage administrators to easily deploy a single file system to span multiple tiers of performance and capacity. This single file system automatically adapts to business data and application workflows over time.
As well as tiering data automatically across different nodes, SmartPools can also use solid state drives (SSDs) to accelerate metadata and file-based storage workflows. SSDs as a tier can be used within a pool to improve metadata or data access performance. Alternately, the SSDs in one tier are used to hold the metadata of files on other tiers—boosting the performance of the entire cluster, including nodes that have no SSDs.
OneFS also allows data to be moved to lower-cost cloud storage with the CloudPools functionality. CloudPools can seamlessly connect to Dell-based cloud storage and third-party providers, including Amazon S3, Alibaba, Google Cloud, and Microsoft Azure. CloudPools expands the SmartPools framework by treating a cloud repository as an additional tier. This enables older or data to be stored in a cold or frozen data tier or archive, thereby taking advantage of lower-cost, off-premises storage.
In addition to the F-series all-flash nodes and using SSDs with SmartPools as part of the file system, OneFS can also use SSDs as an integral part of its caching hierarchy. As such, an optional third tier of read cache, SmartFlash, is configurable on nodes that contain SSDs. SmartFlash is a persistent eviction cache that is populated from system memory (DRAM) as it ages out.
There are significant benefits to using SSDs for caching rather than as traditional file system storage devices. For example, when allocated for caching, the entire SSD will be used, and writes will occur in a very linear and predictable way. This provides far better utilization and also results in considerably reduced wear and increased durability over regular file system usage, particularly with random write workloads. The non-volatile nature of SSDs means that data cached by SmartFlash with persist even during node reboots. Using SSD for cache also makes sizing SSD capacity a much simpler prospect compared to using SSDs as a storage tier. SmartFlash is ideal for workloads such as rendering, HPC, CAD, and software design.
Adding capacity and performance capabilities to a cluster is significantly easier than with other storage systems—requiring only three simple steps for the storage administrator: adding another node into the rack, attaching the node to the InfiniBand network, and instructing the cluster to add the additional node. The new node provides additional capacity and performance since each node includes CPU, memory, and network. The Autobalance feature of OneFS will automatically move data across the InfiniBand network in an automatic, coherent manner so existing data that resides on the cluster moves onto this new storage node. This automatic rebalancing ensures the new node will not become a hot spot for new data and that existing data is able to gain the benefits of a more powerful storage system. The Autobalance feature of OneFS is also completely transparent to the end user and can be adjusted to minimize impact on high-performance workloads. This capability alone allows OneFS to scale transparently, on-the-fly, from 10s of terabytes to 10s of petabytes with no added management time for the administrator, or increased complexity within the storage system.
Allocating data with a single, scalable pool of storage is an often-understated benefit and added efficiency found in a single file system. Managing and selecting volumes that have the requisite amount of free space or manually moving data is time-consuming and inefficient. If chosen incorrectly, the performance demands of a particular workflow may not be satisfied by a particular volume. Also, if the organization cannot address a particular volume, or if the storage administrator cannot move data transparently and quickly, then storage efficiency will be sub-optimal. A OneFS powered cluster operates with storage utilization typically more than 80 percent and is, therefore, highly efficient.