PeerFS

PeerFS is redefining the concept of peer-to-peer replication. It delivers cost-effective data protection for Linux-based enterprise applications, and unlike other replication solutions—that simply make copies of data elsewhere that you have to manually retrieve—PeerFS enables current, active and usable data everywhere it resides.

Linux applications, such as MySQL, PostgreSQL and Apache are playing an increasingly important role in today’s enterprise. As such, these applications increasingly require fault tolerant and other capabilities that compare with the traditional mission-critical applications they replace.

With its patented peer-to-peer replication technology, PeerFS delivers the highest standards of data protection and availability, load balance and safety. And, as a POSIX-compatible file system, PeerFS requires no changes in your application environment.

How does it work?

• PeerFS is the first true peer-to-peer file replication technology that allows multiple sources and multiple targets. This technology provides an unprecedented level of safety, with reliable file sharing on a global basis.
• READ operations are performed locally, without tying up expensive network bandwidth—only the changed bytes within a file are sent over your network.
• Full 256-bit AES encryption is configurable for each endpoint. PeerFS provides seamless failover among endpoints in the event of storage failures. Applications never stop running.
• PeerFS helps Linux applications meet the requirements for disaster recovery, data sharing and data movement that they currently lack. And it does so cost-effectively. PeerFS is implemented without changing existing applications, server or data storage hardware.

Local Access. Global Synchronization

PeerFS is a robust, laterally scalable infrastructure for moving, sharing and protecting valuable application data anywhere. You can access and modify databases or other files that are local to each of you, yet synchronized across 64 or more nodes.

File system technologies are receiving a lot of attention today because they can simplify managing data across long distances and across heterogeneous servers and data storage subsystems. Unlike global file system technologies that may require read and write activity to occur over long distance networks, PeerFS locates data close to users. Read activity occurs locally. Only write activity travels across the network.

Continuous Data Availability

Server clustering is a well-established solution to increase application availability. However, most high availability applications rely on a single RAID storage system to protect data. The RAID system protects the data, but does not assure access to that data in the event of disruptions to that single RAID system.

Assuring application availability requires a more robust approach. PeerFS provides continuous availability to data on disk by replicating it in multiple locations, and by providing multiple paths to reach it. PeerFS permits you to locate data close to any number of groups of users, reducing the possibility of data inaccessibility due to single points of failure. If you local copy of data becomes inaccessible for any reason, you will seamlessly and automatically be directed to an alternative source.

Application and Hardware Agnostic

Remotely Configurable and Manageable

PeerFS is designed to interface with any database or other application. It incorporates a patent-pending locking mechanism that allows multiple users to access the same data while preventing damage to that data. PeerFS presents itself as direct-attached storage or network-attached storage, depending on your needs. Unlike block-level replication products, PeerFS permits you to replicate data among any number of storage nodes that can include any type of storage hardware.

PeerFS can be installed and configured in as little as 15 minutes. Nodes can be added or removed from an existing consistency group on the fly. If nodes are included outside your firewall, communication can be encrypted.

Performance, Scalability and Flexibility

PeerFS operates as a native file system to Linux, and provides a means to synchronize data on as many as 64 storage nodes globally. It runs on single or multi-processor Intel-based Linux servers, and can be configured as your needs change to support new application or new locations.

OS Support

We are constantly adding to the family of supported operating systems.  For the most current list please visit our trial download page.  Download PeerFS.  Following is a sampling of environments supported in our most recent release:

• CentOS 4.4
• SuSE ES 10
• SuSE ES 9
• RedHat Fedora Core 5
• RedHat Fedora Core 4
• RedHat ES 4

Supported Applications

• MySQL
• PostgreSQL
• SUSE Openexchange Server
• Apache
• Others

Storage Interconnects

• Fibre Channel
  • 1 GB
  • 2GB
• Gigabit Ethernet
• Direct Attached
• iSCSI