More reliable than storage RAID

Redundancy is provided by multiple copies (replicas) of the data written synchronously across the cluster. Users set the number of replication copies. We recommend 3 copies as a standard and 2 copies for data that is less critical.

StorPool always distributes copies of a given data block to drives in different servers (or even in different racks). If a drive, server or even half of the cluster fails – the data is still available. It is much more reliable and efficient than storage RAID or traditional storage data protection scheme.

StorPool is a fully distributed system which protect against a variety of hardware and software-stack failures. End-to-end data integrity protects against phantom writes, misplaced writes, wrong data being returned by the underlying hardware and also protects against the majority of bugs. Parallelization give you quick recovery of data in the event of hardware failure.

End-to-end data integrity

The solution provides end-to-end data integrity. From the moment you give us data to store, we calculate a checksum and we store it with the data. This is better protection than systems which don’t have a data integrity mechanism (e.g. storage RAID), and also better than systems which only have a local data integrity mechanism (e.g. Ceph on btrfs).

Data protection

Data errors such as silent data corruption, phantom/partial/misplaced writes are all common for any current storage system, handling many terabytes of data. Our built in end-to-end data integrity ensures no such errors occur.

No single points of failure

Highly available due to its distributed architecture and no single points of failure. Instead of trying to make a single component like a hard disk or an SSD super-reliable (and thus super-expensive) StorPool uses cheap standard components and make them inter-exchangeable. This concept is used by leading hyperscale companies. Now even the most expensive storage systems usually have double redundancy – meaning two power supplies, two network interfaces, etc. per box. However some of the most critical components – like the motherboard or the backplane are a single component. StorPool takes this to the next level – we do not do redundancy on the component level, rather we can tolerate entire servers or racks to fail, without disruption of the storage service.

Simple hyper converged storage infrastructure

StorPool makes it fast and easy to set-up and manage data storage. If you deploy it in converged or hyper converged storage fashion you can also have a single building block for your data centers by combining compute and storage on the same node. This simplifies your entire IT infrastructure, while reducing its cost in times.

Designed with the customer in mind and a focus on simplicity:

• Simple, yet extremely robust architecture
• Usage of standard hardware: makes it simpler and cheaper to set-up and support
• Ease of use – StorPool is simple all the way – to implement, manage and grow
• No experts needed – a junior system administrator is usually able to operate StorPool with some basic training. No need for expensive specialist storage experts with 5 pages of certificates
• Hyperconverged on standard hardware – we know how important hyper-converged infrastructure has become, thus we developed a number of integrations, which makes building hyper-converged solutions fast and easy

When we designed StorPool’s distributed storage architecture we specifically though of simplicity. This is why we do not require custom kernels, we are compatible with the standard software stack, we use standard hardware. Although we are selective about some particular hardware components (mostly SSDs, NICs and HBAs), we still use standard mass-production components, familiar on the market. We always do our best to make everything as simple possible (but not simpler).

Scale-out storage

Seamless scale-out storage in capacity and performance. StorPool can grow without interruption and in incremental steps – one drive, one server and one network interface at a time.

StorPool allows you to scale-out storage – in other words to increase storage capacity gradually by adding just one drive or server at a time! Compare this with currently typical “fork-lift” upgrades (which are exactly what they sound – a fork lift gets in the datacenter to replace a big storage box with an even bigger one).

Scale-out

The system can be scaled by adding a single disk and/or a single server. And with any added disk/server the performance of the entire pool increases. This is done by spreading client’s data on all disks in the pool and parallel communication between the client and all servers. The performance is additionally increased by our tiering and load balancing mechanisms.

StorPool is a genuine scale out storage system that leverages the entire cluster of servers to give massively superior IOPS performance compared to other software storage solutions. StorPool is lightweight and works much closer to the hardware, thus it captures the raw IO speeds of the underlying hardware. StorPool also achieves unmatched performance through its superior network protocol and its copy-on-write on-disk format, while also accelerating random writes and mitigating the “I/O blender effect”.

Capacity and performance increase with any added drive / server

StorPool scales horizontally by adding low-cost servers and drives. Unlike most storage solutions, every time you add a server (CPU, RAM, drives /IOPS/ and network ports) you increase the performance of the entire StorPool system. StorPool takes care of spreading the data to all servers and keeping a prescribed number of copies. This has a significant impact on operating costs and complexity for our customers.

Individual drives in a StorPool cluster can be added and taken away with no impact on the service/data availability, provided sufficient copies have been made. The StorPool cluster is a flat hierarchy of drives and new drives can be added or removed at any time. When new drives are added to a StorPool cluster the capacity of these drives is simply added to the pool. StorPool automatically ensures that copies are placed in separate fault domains (typically separate servers), so there is no need to plan and configure RAIDs.

Add more drives to the storage cluster anytime you need!

StorPool is extremely flexible and offers a higher flexibility than other storage systems. It has sophisticated storage cluster and volume management, allowing the creation of multiple volumes each with a different placement policy (meaning copies) on any subset of drives from the pool. E.g. you can have volumes on SATA drives, other volumes on SSDs or volumes on mixture of HDD and SSD drives (hybrid). You can live-migrate (tiering) data from one set of drives to another set. Also each volume can also be configured for data locality – on the compute node that needs it, i.e. keeping data close to where it is used.

Why is StorPool more flexible than other solutions:

• Anything in StorPool can be changed online – online expansion/shrink, online changes, automatic self-healing, auto-load balancing, in-service upgrades
• No cluster downtime on hardware changes – you can add/remove hardware without this bringing the service down – forget about hot-swappable disks. You can hot-swap whole servers (or even racks)
• StorPool provides unmatched levels of operational flexibility. It also scales-out seamlessly in capacity and performance.
• Different deployment options – StorPool can be used as a stand-alone storage (on dedicated storage nodes. It can be deployed in converged (Compute + Storage on the same servers) or hyper-converged fashion. Each server in a StorPool storage cluster can be only client, only server or both.
• Fully distributed system – StorPool does not have dedicated metadata servers, master-slave or other lock-up mechanisms – this makes it extremely flexible and robust, but also very fast due to the massive parallelism in processing data which combines IOPS and throughput of all servers in the storage cluster

Individual drives in a StorPool cluster can be added and taken away with no impact on the service/data availability, provided sufficient copies have been made. The StorPool cluster is a flat hierarchy of drives and new drives can be added or removed at any time. When new drives are added to a StorPool storage cluster the capacity of these drives is simply added to the pool.

StorPool automatically ensures that copies are placed in separate fault domains (typically separate servers), so there is no need to plan and configure RAIDs. StorPool volume management also offers a higher flexibility than most storage systems. StorPool allows for the creation of multiple volumes each with a different placement policy out of any subset of drives from the pool. E.g. you might want some volumes on SATA drives, while other volumes on SSDs. Each volume can also be configured for data locality, keeping data close to where it is used.

SAN storage arrays and all-flash arrays are too expensive?

Data storage is still the single most expensive component in the datacenter. And companies now realize that with all these costs in their balance sheet they lose competitive edge to more innovative competitors. This is why we are not surprised by the recurring pattern we see – customers replace traditional SAN storage arrays, all-flash arrays and even other storage software with StorPool in order to bring down the cost of their storage system.

StorPool uses standard hardware which brings solid reduction in Total Cost of Ownership and increases Return on Investment. With StorPool 2x less hardware is needed: it makes Converged Storage+Compute possible (storage running on the same hypervisor/application servers). StorPool is superior to traditional SAN storage arrays or all-flash storage arrays in both technical and business terms.

Standard hardware

Unlike big SAN storage boxes from major storage vendors, StorPool uses commodity hardware. It is well selected, still standard / white label / OEM / white-box hardware. This is why a StorPool storage solution is simpler, easier to procure and cheaper to set-up, maintain and support. It does not require specialist workforce and specific certificates. The bottom line is it can be up to 10 times less expensive in the long run than a comparable dedicated storage hardware solution.

Increased utilization

StorPool can in practice run Compute and Storage on the same servers (converged/hyper-converged/Server SAN deployemnts). By doing so, companies can use one set of servers for both running applications and storing data. In this way, they save on CAPEX, space, utility costs and maintenance.

“Pay-as-you-go” pricing

StorPool has “pay-as-you-go” license model. It is one of the first storage vendors to align pricing of storage with the business demands for flexibility and operational expenses – we shift high CAPEX to affordable OPEX.

High-end SAN storage features

An often overlooked fact is that StorPool provides most of the features that you would expect from a mid- to high-end SAN storage array – thin provisioning, scale-out, data integrity, QoS (Quality of Service), snapshots and clones, multi-controller system, etc, etc. However, with most traditional SAN storage or all-flash/solid state storage vendors each of these features comes with a separate price tag. With StorPool they all come in the package.