Qumulo’s File Storage System
Your Choice of Operating Environments
Contents of this guide
- Overview of the Qumulo file system
- What’s included in Qumulo subscription?
- Your choice of environments
- Real-time quotas
- Continuous replication
- Scalable Block Store (SBS)
- Data protection with erasure coding
- Hot/cold tiering for read/write optimization
- Protocol support
- REST API
Your choice of operating environments
Qumulo software currently runs on Qumulo’s own QC-series of white box servers, as well as Hewlett-Packard Enterprise (HPE) Apollo servers, and Dell EMC PowerEdge, or in the cloud on Amazon Web Services (AWS). Because Qumulo’s file system is hardware-independent, it does not rely on any single hardware vendor; in fact, you can expect Qumulo to add support for servers from a number of additional manufacturers.
Qumulo’s software-only approach means that there are no dependencies on expensive, proprietary components such as NVRAM, InfiniBand switches and proprietary flash storage. Instead, Qumulo’s file system relies on hard disk drives (HDD) and solid state drives (SSD) with standard firmware that is available from the drive manufacturers.
You can also create Qumulo clusters in the public cloud. The nodes of cloud-based Qumulo clusters run the same Qumulo Core software as on-premise Qumulo clusters.
Unlike other cloud-based file storage systems, a Qumulo cluster running in the public cloud has no hard limits for performance or capacity. Both can be increased by adding nodes (compute instances and block storage). Qumulo is the only solution for cloud that lets you flexibly scale both capacity and performance.
When running in the cloud, Qumulo’s file system uses cloud block storage in a way that’s similar to the SSD/ HDD combination on premise. Qumulo uses low-latency block storage as a cache in front of economical higher-latency block storage.
On each node of a Qumulo cluster, Qumulo software runs in Linux user space rather than kernel space. Kernel mode is primarily for device drivers that work with specific hardware. By operating in user space, Qumulo can run in a wide variety of configurations and environments, and also deliver features at a much faster pace.
Running in user space also means Qumulo’s file system can have its own implementations of crucial protocols such as SMB, NFS, LDAP and Active Directory. For example, Qumulo’s implementation of NFS runs as a system service and has its own notions of users and groups separate from the underlying operating system on which it runs.
Running in user space also improves Qumulo’s reliability. As an independent user-space process, Qumulo is isolated from other system components that could introduce memory corruption, and the Qumulo development processes can make use of advanced memory verification tools that allow memory-related coding errors to be detected prior to software release.
By using a dual partition strategy for software upgrades, Qumulo can automatically update both the operating system and Qumulo Core software for fast and reliable upgrades. You can easily restart Qumulo’s file system without having to reboot the operating system, node or cluster.