Flexible, Scalable Data Storage
As businesses grow and storage requirements spiral upward, storage servers and arrays invariably proliferate throughout the organization. The presence of such disparate storage resources eventually becomes difficult to manage, and this often leads to under utilized (even lost) storage. In fact, it's common to see storage utilization at only about 50% resulting in wasted capital expenditures on new storage. Storage virtualization has emerged as an answer to this dilemma, allowing storage administrators to identify, provision and manage disparate storage area networks (SANs) as a single aggregated resource.
This is an important element of storage consolidation, easing management headaches and allowing higher levels of storage utilization, which in turn eliminates the expense of adding superfluous storage area networks.
What is Storage Virtualization?
Storage Virtualization is the pooling of physical storage from multiple storage area networks into what appears to be a single storage device that is managed from a central console. Storage virtualization is commonly used across multiple storage area network (SAN) providers.
The management of storage devices can be tedious and time-consuming. Storage virtualization helps the storage administrator perform the tasks of backup, archiving, and recovery more easily, and in less time, by disguising the actual complexity of the SAN.
SAN symphony-V software solves many of the difficult storage-related challenges raised by server and desktop virtualization in contemporary data centers and Cloud environments. The software forms an active, transparent virtualization layer across diverse storage devices to maximize the availability, performance and utilization of disk resources in IT organizations large and small.
The integrated set of centrally-managed data protection, provisioning, caching, replication and migration functions operates uniformly over different models and brands of storage equipment, assimilating current and future technologies non-disruptively. You’ll find that SAN symphony-V cost-effectively delivers uninterrupted data access, speeds up applications, and extends the life of your tiered storage investments, while giving you peace of mind.
SAN symphony-V forms a transparent, scalable virtualization layer across your storage infrastructure in order to enhance its capabilities and centralize its management. The many nuances and incompatibilities distinguishing different models and brands of storage no longer stand in the way of using them together.
You have complete freedom to mix and match storage devices, migrating between old and new in the background. SAN symphony-V works with all the popular models and brands of disks and disk systems supported on Windows Servers allowing you to balance between high-speed, premium-priced resources and lower cost, higher capacity gear.
The same flexibility applies to packaging. Larger environments generally use external disk arrays to house hundreds of drives in very high-density packages. On the other hand, compact applications in cramped rooms or hardened mobile enclosures can take advantage of drives housed inside the DataCore nodes. Still others start small with internal drives and expand externally. The software accepts direct-attached storage (DAS), SAN arrays and solid state disks / flash cards with equal ease, connecting to them via any of the standard disk interfaces.
Provisioning Disk Space
The SAN symphony-V administrator carves out virtual disks on demand from the physical disk pool according to the capacity, availability and performance needs of each workload.
For instance, some groups of virtual disks may be defined to be cached, synchronously mirrored, remotely replicated and thinly provisioned. In the background, the DataCore software draws on multiple devices and the necessary computing and network connections to meet those requirements. Auto-tiering functions dynamically determine which storage class will best meet the selected storage profile.
Hosts connect to the DataCore nodes over iSCSI and/or Fiber Channel just as they would connect to an external SAN disk array. Network ports and I/O channels may be configured from a wide range of host bus adapters (HBAs) and network interface cards (NICs) including iSCSI virtual SANs inside virtualized servers. The protocol between the hosts and the DataCore nodes may be different from what the nodes use to connect to the physical storage. For example, SAN symphony-V bridges iSCSI connections from hosts to trays of SATA and SAS drives.
For security purposes, hosts only see those virtual disks explicitly shaped and assigned to them over designated ports. In clusters, several hosts can share virtual disks even if the associated physical drives are not multi-ported (shareable).
To maximize disk utilization and eliminate wasted space, DataCore implements very granular, thin provisioning techniques and space reclamation features.
SAN symphony-V serves virtual disks to physical hosts running any of the popular open operating systems as well as virtual machines hosted by the mainstream server hypervisors.
The extensive PowerShell scripting library, SNMP traps, and task scheduling wizards simplify integration with other workflows and systems management tools. Real-time / historical charts, e-mail event notifications and reporting features provide additional insights into the health and performance of the virtualized storage infrastructure.
Uninterrupted Access Using Mirrored Nodes
SAN symphony-V shields applications from planned or unplanned outages in the underlying components by providing uninterrupted access to the virtual disks. It synchronously mirrors disk updates between nodes in separate rooms so they won’t be exposed to the same facility-related hazards.
Stretching the mirrored nodes up to 100 kilometers apart decreases the chances that ordinary misfortunes such as roof leaks, fires, air conditioning failures, construction crew mishaps or floods will affect both sites. Virtual disks in these high-availability (HA) configurations appears like a single, well-behaved, multi-ported shared drive, although it is really made up of two widely-separated mirror images updated in lock step.
Entire sites, nodes, disk subsystems, channels and other components of the environment may be taken out-of-service, upgraded, expanded and replaced without disturbing applications or scheduling downtime.
Replicate Offsite for Disaster Recovery
For additional offsite disaster recovery, or simply to migrate data between locations, SAN symphony-V replicates disks asynchronously over unlimited distances using conventional IP lines. You can also reverse the direction of replication to restore the original site after the danger has been averted. For additional confidence, remote restoration procedures may be tested regularly while normal production processing goes on at the primary data center.
Cache Overcomes I/O Bottlenecks
DataCore taps each node’s processing, memory and I/O resources to carry out advanced functions across all of the storage devices under its command. SAN symphony-V reserves up to 1 Terabyte (TB) of random access memory (RAM) per node for SAN-wide “mega caches”.
Thanks to the sophisticated multi-threaded caching algorithms, data written to or read from disks move swiftly into and out of the caches, harnessing the full potential of each node’s high-speed multi-processors. Caching makes application run faster than they would had they accessed the disks directly, whether connected to top-of-the-line storage systems or lower end gear.
Automatically Optimize Disk Access Across Tiers
Disks with different price/performance attributes can be organized into separate storage tiers. For example, the fastest tier may be composed of Solid State Disks (SSDs), with lower tiers made up of SAS and SATA drives, respectively. The same tier may include similar units from different manufacturers acquired over time.
SAN symphony-V automatically promotes frequently accessed disk blocks to faster tiers and demotes less active blocks to slower, more cost-effective drives based on the IT team’s criteria. You may override auto-tiering policies by explicitly defining which disks should be used for special workloads or confine the selection of tiers to a narrower set of resources.
Groups of disks may also be segregated for specific requirements, such as test / development or to isolate multiple subscribers (multi-tenancy) in public, hybrid and private clouds.
Advanced features such as caching, synchronous mirroring, asynchronous replication, snapshots, thin provisioning and CDP operate across mutually independent devices within the same tier or across tiers. For example, we find customers prefer to take online snapshots of virtual disks in tier 1 and place them in a tier 2 or tier 3 pool to avoid consuming premium resources for the backup copies. Similarly, tier 1 devices may be remotely replicated to a disaster recovery site that only has tier 2 devices, possibly from a different supplier.