Parallels Cloud Storage

Revolutionary Cloud Storage

Included in Parallels Cloud Server, Parallels Cloud Storage uses the existing hard drives in your server nodes, improving server availability and performance with elasticity and scalability at a fraction of the cost of competing solutions.

Learn more about why Parallels Cloud Storage is the ideal storage solution for hosters

Storage Optimized for the Cloud

  • Eliminate server downtime due to hardware failures
  • Deliver high availability servers
  • Scalable to petabytes

Learn more about how hosters can enhance their offerings with Parallels Cloud Storage 

High Performance

  • Distributes storage across all nodes, maximizing bandwidth
  • Enables zero downtime server migration and no-disruptive replication
  • Support for SSD caching to burst I/O performance and increase hardware utilization

Cost Effective

  • Reuses existing hardware and is as light as possible in terms of its resource footprint
  • Reduces maintenance and support costs

Key Features


High performance and scalable to petabytes


Expands easily when adding additional nodes and devices

Failure Tolerant and Redundant

Users can set the level of redundancy they require, and Parallels Cloud Storage will automatically detect failed nodes/drives and perform auto-recovery to maintain that specified level of redundancy. During failure recovery the storage cluster adjusts to provide continuous data access for the clients.

Auto Recovery

Automatically recovers damaged, lost, or inaccessible data replicas below the configured replication level if a drive/node fails.

High Availability

Ensures that the end user will not experience a loss of service if the server suffers a hardware failure.  In the event of a failure the system automatically switches to redundant data copies and can quickly restart the service on another physical machine.

Data Protection

Provides user data check-summing and verification. Also performs background data verification “scrubbing” to ensure that all data is still readable and correct.

SSD Caching

Bursts I/O performance by an order of magnitude, increasing hardware utilization and density.


When new storage servers are introduced into a cluster or come online after failure, the system will rebalance the storage across the total cluster.

Introductory Information

  1. 1.
    What is Parallels Cloud Storage?

    Parallels Cloud Storage is the storage technology delivered with Parallels Cloud Server. It allows service providers to build a reliable, and highly available, distributed shared storage cluster which performs similar to a SAN. However, Parallels Cloud Storage is much more cost effective since it uses existing hard drives in your current servers. This eliminates the need for the dedicated shared storage hardware normally required by SANs and provides the lowest cost cloud storage available. With Parallels Cloud Storage you can store and run Parallels virtual machines and containers, eliminate VPS downtime due to HDD and RAID failures, provide high availability servers, perform instant live migration of servers between physical hosts, and scale available storage beyond the locally available disk space.

  2. 2.
    Who is the ideal customer for Parallels Cloud Storage?

    Parallels Storage is designed for hosting providers who want to dramatically improve server availability and reliability while reducing costs.

  3. 3.
    Do I need to buy additional storage hardware with Parallels Cloud Storage?

    No, Parallels Cloud Storage eliminates the need for external storage devices typically used in SANs by converting locally attached storage from multiple nodes into shared storage.


  1. 4.
    How is the Parallels Cloud Storage licensed?

    Parallels Cloud Storage licensing is scalable to meet the needs of service providers. Licensing is based on the amount of enabled storage in 100 GB increments, to allow easy scalability as demand grows.

  2. 5.
    How can I buy Parallels Cloud Storage?

    Parallels Cloud Storage is delivered as a component of Parallels Cloud Server and is available for download from the Download Cloud Server page.

  3. 6.
    Is there an evaluation key for Parallels Cloud Storage?

    Yes, please contact our sales team for an evaluation key.

Scalability and Performance

  1. 7.
    Is there a limit on the number of storage nodes?

    There is no strict limit on number of storage nodes. However, we recommend limiting the servers in the cluster to a single rack to avoid any possible performance degradation that can occur with inter-rack communications.

  2. 8.
    What is the limit on the total disk space?

    Parallels Cloud Storage can scale to support up to 1 PB of effective available disk space, which means up to 3PB of physical disk space in the case of mirroring with 3 copies.

  3. 9.
    Can I add additional nodes or storage capacity to an existing Parallels Cloud Storage cluster?

    Yes, you can dynamically add and remove nodes from a Parallels Storage cluster to increase capacity or to remove node(s) for maintenance.

  4. 10.
    Can Parallels Cloud Storage be configured for remote replication between separate data centers?

    No, since we cannot guarantee link performance between data centers we do not support this at this time.

  5. 11.
    What is the expected performance of Parallels Cloud Storage?

    The performance will depend on the hard disks used and the network speed. However, the performance should comparable to locally attached storage and in some cases even better! Additionally Parallels Cloud Storage supports advanced SSD caching which increases the performance of commodity hardware by adding small, inexpensive SSD drives to the cluster for caching purposes.

  6. 12.
    What performance should I expect on 1Gbit ethernet?

    The 1Gbit ethernet network maximum speed is close to that of a single rotational drive, however, in most workloads random I/O access is prevalent and the network is not a bottleneck. Research with large service providers demonstrates that average I/O performance rarely goes above 20MB/sec due to randomization. Virtualization itself introduces additional randomization as multiple independent environments perform I/O access simultaneously. Regardless, 10Gbit ethernet will result in better performance and is recommended when possible.

  7. 13.
    Does the addition of new storage nodes or disks improve the overall performance for existing users?

    Yes, since data is distributed among all hard drives in the cluster, applications performing random I/O will experience an increase in IOPS when more drives are added to the cluster. Even a single client machine may get noticeable benefits by increasing the number of storage nodes and achieve performance far beyond traditional local attached storage.

  8. 14.
    Does performance depend on the number of replicas?

    Each additional data copy degrades write performance about 10%. However, it may improve read performance since Parallels Storage has more options to select a faster server.


  1. 15.
    How does Parallels Cloud Storage protect server data?

    Parallels Cloud Storage protects against data loss or temporary unavailability by mirroring data to a specified number of copies (replicas) placed on different machines to protect the data from single point hardware and control failures. To provide additional reliability Parallels Cloud Storage can be configured to maintain user data checksums and verify them on access and on an ongoing basis.

  2. 16.
    What happens when a disk is lost or a server node is unavailable for whatever reason?

    Parallels Cloud Storage automatically recovers from a degraded state to the specified redundancy level by replicating data on live nodes. Data access is still available during the recovery.

  3. 17.
    Will data survive intact in case of a power supply failure, where all nodes lose power at the same time?

    Yes, the data is secure and will be available when the power returns.

  4. 18.
    How fast does Parallels Cloud Storage recover from a degraded state?

    Since Parallels Cloud Storage recovers from a degraded state using all the available hard disks in the cluster this process is much faster than with traditional locally attached RAIDs. Overall this makes the reliability of the storage system significantly better as the probability of losing the only remaining copy of data during the recovery period is very small.

  5. 19.
    Can I change redundancy settings on the fly?

    Yes, you can change number of data copies and Parallels Cloud Storage will dynamically adjust by creating new copies or removing unneeded ones.

  6. 20.
    Do I still need to use local RAID?

    No, Parallels Cloud Storage provides the same built-in data redundancy as a mirror RAID1 array with multiple copies. However, for better sequential performance it is possible to use local stripping RAID0 exported to the Parallels Storage cluster.

  7. 21.
    Does Parallels Cloud Storage have redundancy levels similar to RAID5?

    No, it is not possible to build a reliable software-based RAID5 system without also including special hardware capabilities like backup power batteries. In the future Parallels Cloud Storage may provide RAID5 level redundancy for read-only data such as backups.

  8. 22.
    What is the recommended number of data copies?

    We recommend configuring Parallels Cloud Storage to maintain 2 or 3 copies, which allows the storage system to survive the simultaneous loss of 1 or 2 hard drives.

Compatibility and requirements

  1. 23.
    What are the hardware requirements for Parallels Cloud Storage?

    Parallels Cloud Storage does not require any special hardware and can run on commodity nodes with traditional SATA drives and 1Gbit ethernet networks. Some hard drives and RAID controllers ignore the FLUSH command to imitate better performance and must not be used as this may lead to file system or database corruptions. This is especially true for RAID controllers and SSD drives. Please consult with the drive manual to make sure you use reliable hardware.

  2. 24.
    Can I run Parallels Cloud Storage with one node?

    You can, but this does not provide any advantages over locally attached storage except for the ability to seamlessly grow to multi-node configurations and to take advantage of Parallels Cloud Storage caching features.