Scale Computing for Virtualization

Scale Computing builds storage to meet the demands of virtualization using a scale-out architecture that ensures virtualization is not stalled by underperforming storage. Most importantly, this storage technology is available at prices that enable small to medium-sized enterprises to fully embrace virtualization, not just for consolidation cost savings, but also for improved availability and more dynamic IT service delivery.

About Virtualization

Virtualization is transforming the IT landscape, and with good reason. With virtualization, traditionally static resources such as networking, desktops, and servers are dynamically allocated and centrally managed. This flexibility has far-reaching business benefits:

• Get more out of existing resources. Non-virtualized servers rarely exceed 30% resource utilization. Consolidation into a virtualized environment can exceed 80% utilization of the hardware.
• Increase the availability of workloads. Virtualization provides methods to migrate workloads to different hardware in the event of planned maintenance or unplanned downtime.
• Improve IT service levels. Provision and deploy servers and desktops faster with reduced management overhead.
• Shrink operational costs. By reducing the physical footprint of the datacenter, costly data center build out are avoided and power and cooling costs shrink.
• Leverage legacy applications. Lengthen the life of legacy applications by operating them on virtual machines.

Scale Computing Tames the Virtualized Environment

But virtualization comes with its own set of challenges, especially for the storage underneath. Once virtualization is in place, it becomes very easy to provision new servers or desktops as needed. Resources previously underutilized are quickly applied to additional workloads, each with their own capacity and performance demands on the storage. Where the business may have required substantial time to acquire and deploy a new server in the past, it is deployed from thin air in minutes now.

Virtualization has shifted the limiting factor of the environment’s responsiveness and availability onto the storage containing the VMs.

Scale Computing offers scale-out storage capable of delivering the reliability, capacity and performance required of demanding virtualized environments. These advanced but highly cost-effective storage systems are extremely scalable, growing node by node to meet increased demands, with no disruption to the virtualized environment.

The Advantage of Scale-out

Legacy controller-based storage systems are designed with the ability to scale up capacity, but what about extra performance? Regardless of how much extra capacity is added, the performance is static, and the usability of VMs will only decrease as the controller is stressed, like adding more traffic to an overloaded freeway.

Scale Computing solves this problem by removing the controller bottleneck. A Scale storage cluster uses independent storage nodes, each contributing additional capacity to a single, unified storage pool, and each adding to the I/O bandwidth available. Collectively, the cluster delivers high-performance, redundant storage tailored for virtualization.

Enterprise Features for the SME

This type of architecture is possible using Scale’s innovative ICOS™ Technology included on every node. In addition to a true scale-out solution, the ICOS software provides an unprecedented level of storage features that are normally found only on expensive, enterprise-priced systems.

Intelligent Rebalance

An ICOS powered cluster ensures that the data is well protected. Data is striped and mirrored across multiple nodes in the cluster. This allows the cluster to stay fully available in the event of a drive or node failure. In the rare event a drive does fail, rather than requiring a hot spare and lengthy rebuild time from RAID parity volumes, ICOS automatically and quickly rebalances the lost data from the mirrored copy to remaining available space on the cluster. Whenever a replacement drive is added, the capacity is added to the cluster for new data.

Intelligent Pre-fetch

Rather than waiting for you to request data, the cluster anticipates what you may need, based on other data recently accessed, and then pre-fetches data into the cache. The result: significant performance gains in real-world usage.

Unified Storage

Many virtualization platforms support both shared block-level storage (SAN) and shared file-level storage (NAS). There are unique advantages to each, so don’t get trapped with only one option. Scale Computing storage clusters with ICOS use a Protocol Abstraction Layer to simultaneously offer iSCSI, CIFS, and NFS protocols with no extra licensing or gateway costs.