Store and backup limitless data with a powerful management solution that offer
high availability, agility, and flexibility for your business storage needs.
StoneFusion®, the engine that drives StoneFly’s Intelligent iSCSI and Fibre Channel SAN
The award-winning StoneFusion® Network Storage Platform (NSP), StoneFly’s patented storage networking operating system architecture, brings in block level storage intelligence to the IP network core, StoneFly’s iSCSI & Fibre Channel SAN appliances deliver all the benefits of managed SANs: increased storage utilization through resource consolidation, storage provisioning, centralized access control, volume management, and advanced storage services including Active/Active clustering, mirroring, replication, encryption, snapshots, and Thin Provisioning. At its core, StoneFusion virtualizes physical storage resources into virtual volumes and enables advanced storage functions in a virtual network layer. StoneFusion is the foundation of all StoneFly SAN, Scale Out and Hyper-Converged product lines. With StoneFly the power of SAN is at your fingertips.
StoneFusion® Differentiators & Features:
- Fully Active/Active Clustered iSCSI SAN storage
- Fibre Channel SAN target
- Scale Out NAS
- Single Namespace
- No Metadata
- Add nodes without losing performance
- Scale Out (Horizontal) or Scale Up (Vertical)
- Support for iSCSI, NAS (CIFS/NFS), FC, SNMP Traps, UPS, Nagios, RAID Monitoring, Call Home
- Thin Provisioning
- StoneFusion® software deployed on Standard x86 / Intel Platforms
- Leverages off-the shelf, Best-in-Class Hardware for Adaptable Solutions
- Tightly coupled with Linux Internals including I/O
- Result: Broad Compatibility, High performance, Quick Product Development
- Advanced Copy-on-write snapshots
- Read/Write snapshots
- Advanced scheduling
- StoneFly VSS agent available
- Advanced mirroring
- Synchronous mirroring
- Campus mirroring (LAN/WAN) (iSCSI/FC mirroring)
- Multi-Site/Multi-Appliance Replication and Unified Central Management System
- Comprehensive enterprise wide performance reporting and statistics, call home, SNMP, etc.
- Over ½ million lines of code and growing
- Multi-Gigabit or 10Gb (CX4 or SFP+) support
- Tiered Storage (SATA, SAS, SSD) support
- Support for all Virtualization environments
- iSCSI port teaming, failover, and load balancing
- Multi-box management
- Horizontal or vertical scalability
- Wire speed Performance
- Volume based encryption (AES256)
- Data deduplication
- Dynamic / on-demand, no downtime volume or storage expansion
- Easy to use, Web-based User Interface
- Patented I/O Processing in iSCSI Target
- Conventional Storage Processing:
- Wait for complete I/Os before beginning processing
- Processes complete I/Os
- Transmits complete I/Os
- StoneFly Approach:
- Starts processing while I/O is in process
- Virtualization engine classifies communications to route them efficiently over the network
- Decreases latency
- Eliminates I/O bottlenecks
- Operates seamlessly behind the scenes
- Benefits: Efficient, concurrent processing, lower latencies
- Conventional Storage Processing:
Features & Benefits:
|IP Network-based SAN Intelligence||
Universal connectivity on IP networks
Supports iSCSI and optional FC SAN protocol(s)
Uses existing Ethernet network infrastructure
Lower total cost of ownership by enabling customers to cost-effectively build and manage a SAN that supports networking technologies they already have
|Disaster recovery and fault-tolerance||Supports corporate business continuity objectives with advanced features, such as Active-Active FailOver clustering, mirroring (local and remote data replication), data migration, and disk-to-disk backup & restore|
Users can access storage without needing to know where it is located or how it is configured
Maximizes utilization of storage resources and eliminates waste by freeing stranded storage, making it available on the IP network
Allows administrator to quickly respond to changes in storage requirements
Helps administrators consolidate backup processes for faster and more reliable backups
Supports disk-based backup for almost instantaneous restores
Integration with Microsoft’s Volume Shadow Copy Services for seamless snapshots
|Host Independent HTML-based Management||
Centralized storage management, control, and monitoring of logical storage volumes
Easy and intuitive to use
Reduction in storage administration overhead
|Storage Device Independence||
Works with SAS, iSCSI and Fibre Channel storage
Easily configured to work with standard off-the-shelf or legacy storage
|Access Control Lists||Allows administrator to designate which physical storage devices are accessible to individual hosts|
Learn more with the following White Paper
StoneFly’s active/active failover clustering technology enables Storage Concentrators to be deployed in mission critical environments. With seamless field-tested failover technology and load-balancing, Storage Concentrator Clusters increase availability and performance cost effectively, serving as the backbone of a SAN with no single point of failure.
Learn about Clustering technology with the following Datasheet
Mountable Read-Write Snapshot Volumes
Snapshots contain an image of the volume at the exact point-in-time that the snapshot was taken. The snapshot volume appears to any host machine as if it was a regular logical volume and can be mounted as read-only or read-write. Even after changes are made to the original volume, delta-based snapshots preserve a copy exactly as it existed when it was taken. Any read or write operations do not impact the primary volume. Snapshots persist across reboots and can be mounted and accessed just like any other volume.
The StoneFly appliance can automatically take snapshots at predetermined intervals using the Snapshot Scheduler. Administrators can configure the system to automatically take snapshots hourly, daily, weekly and monthly.
Snapshots are not replication, like StoneFly Reflection (Synchronous Mirroring). An exact copy would mean that five Snapshots require five times the storage space of the original. Instead, snapshots use Copy-On-First-Write (COFW) mechanisms that only copy data blocks if and when an application changes them. Often, this “copy” is of a pointer followed by an allocation of a data block.
Rollback is a disaster recovery feature that allows users to return to the last known “good” data point in the event of corruption, viruses, or lost files. Essentially, performing a rollback returns the Live Volume to the known state of one of the Snapshots. All writes that have been completed since the Snapshot was taken are overwritten so that the Live Volume appears just as it was when the Snapshot was taken. You can rollback multiple times to get the desired result. For example, if you rollback to find a lost file and the file isn’t there on the first snapshot used, you can rollback again using an older snapshot to locate the item required.
Learn more with the following White Paper:
Reflection Synchronous Mirroring
Reflection provides continuous access to a SAN volume without interruption to data availability, simplifying disaster recovery and reducing system downtime.
How it Works
StoneFly Reflection Synchronous Mirroring supplies host-independent, mirrored data storage that duplicates production data onto physically separate mirrored target images. This duplication occurs transparently to users, applications, databases, and host processors. The software duplicates block-level changes as they occur to one or more volumes: either to another local Storage Concentrator volume image or to another Storage Concentrator volume image at a campus location over standard IP connections.
High-Availability for Critical Applications
Reflection offers the benefit of protecting a critical volume from being a single point of failure, and providing continuous access to a volume without interruption to data availability when loss of access occurs to one of the images. The result is “Next Generation” performance, reliability and high data availability for critical applications.
The StoneFly Asynchronous Replication upgrade option allows data to be kept at two or more different locations. The local data volumes are synchronized with mirror volumes at the remote site at irregular intervals. The synchronization operations may be scheduled or allowed to happen at any time the local data volume falls quiet for a prescribed period of time.
Learn more with the following White Paper:
Asynchronous Replication can be used to replicate data between two or more geographically separate sites. The local data volumes are synchronized with mirror volumes at the remote site at irregular intervals. The synchronization operations may be scheduled or allowed to happen at any time the local data volume falls quiet for a prescribed period of time.
Who Needs Asynchronous Replication?
Asynchronous Replication is associated with a desire to have a level of data recovery in a remote location in the face of a disaster. Recently, more companies have been required to keep data at certain distances (in some cases hundreds of miles) away from the production site to insure protection of the data in the event a disaster at the local site.
How it Works
The StoneFly method for providing asynchronous mirrors includes the ability for the local physical or virtual StoneFly Storage Appliance to identify changes in a volume and periodically move those changes to a remote site. This process is done by coupling StoneFly Reflection Mirroring and Delta-Based Snapshot technology. The StoneFly Snapshot technology tracks the changes in the local (production) volume while the Reflection technology maintains the movement of the data to the mirror image at the remote site.
A simple view of the way StoneFly Asynchronous Replication works is to describe the two volumes that must be synchronized during the mirroring or replication process:
- The Remote (Replication) Volume
- The Production Volume
The Production Volume is the source of all changes in the replication process. Changes are captured through the use of snapshot technology that tracks all changed blocks in a volume. Once the time for a replication has been identified, a differencing process takes place and all changed blocks are transferred to the remote site. A special set of snapshots are established on the production volume to allow for the differencing process to take place. Once the snapshots have been established on the volume a second step is used to connect the volume to one or more of the available remote volumes.
StoneFly’s snapshot technology uses a separately allocated space in the Resource pool for tracking changes. The allocation of this space for the Production Volume is a buffer against problems with communications to the remote site. A larger snapshot allocation provides more space to hold changes while the communications link may be down.
The remote volume may also be enabled for snapshots. This is highly recommended to improve the accuracy and stability of the replicated data at the remote site. Each time a set of changes is sent to the remote volume, a snapshot is requested by the StoneFly Storage Appliance at the production site. These snapshots become recovery points at the remote site. The amount of Snapspace allocated at the remote site determines how many recovery points are possible. Allocating a larger Snapspace provides more recovery points up to limit defined by the software implementation.
StoneFly’s deduplication technology provides optimized data deduplication for increased storage efficiency. StoneFly’s data deduplication capability allows users to fit 5x to 137x more data (depending on the data mix) within the same storage footprint with minimal impacts to overall performance. Block segmentation optimizes efficiency based on data type.
Data Optimization & Efficiency
Of the available data optimization technologies, data deduplication offers the greatest potential to deliver substantial and recurring impact on the cost and manageability of data growth. By reducing storage capacity, organizations will enjoy related cost reductions in areas including storage licensing, storage management, power, cooling, and data center floor space. Data deduplication makes it much more affordable to store, analyze, manage and house your data.
Minimal Performance Impact
StoneFly’s deduplication takes place at the block level. It operates mostly outside the data path of the storage application software as a duplicate advisory service. StoneFly uses parallel deduplication, which greatly reduces the high-write latencies commonly seen with in-line deduplication systems. Featuring a high performance index engine, StoneFly’s deduplication handles heavy workloads and can rapidly identify duplicate data across TeraBytes of data.
How StoneFly’s Deduplication Works
StoneFly’s deduplication index utilizes advanced techniques to identify duplicates across massive storage pools using memory-resident information over 99% of the time, avoiding costly disk access and eliminating the largest single bottleneck in storage deduplication. StoneFly index operations are orders of magnitude faster than other deduplication solutions.
- StoneFusion operating system pushes new data and internal placement information to be deduplicated.
- Block-aware segmentation breaks larger objects into variable-sized chunks.
- Unique content fingerprints are computed.
- Patented indexing technologies determine if the chunk has been previously seen.
- Previous placement information is pushed asynchronously back to the StoneFusion operating system for unification.
Immediate Cost Savings
Independent validation has demonstrated that StoneFly’s deduplication software offers up to 7x data reduction for general purpose environments, and up to 137x data reduction for virtualization images such as those generated by VMware®. This allows users to greatly reduce their storage footprint along with all of the operating expenses involved.
Learn more with the following White Papers