Clustered storage vendor Scale Computing today upgraded its scale-out storage cluster to include compute resources dedicated to applications and a hypervisor component that allows storage and memory to be easily managed and allocated.
Scale Computing calls its new HC3 appliance "infrastructure-in-a-box," saying it eliminates the need to buy virtualisation software, external servers and shared storage. That can mean significant reductions in both costs and complexity.
Scale Computing is part of an industry trend of all-in-one products built to reduce set-up time and ease management of virtual machine (VM) environments.
An HC3 four-node cluster
The company, which was launched in 2007, has so far shipped 1,000 clustered storage nodes to around 750 customers. But those appliances were only a precursor to the company's goal of creating an all-in-one appliance for small to mid-sized companies with upwards of 500 employees and up to to six IT generalists. Many such firms don't have virtualisation or storage specialists.
"We finally delivered on our vision for Scale Computing from the beginning, and why it's called Scale Computer and not Scale Storage," said Scale Computing CEO and co-founder Jeff Ready.
One of the new components added to create the HC3 is a virtualisation tab in the management console. It allows administrators to choose from a pool the compute, memory and storage capacity they want to assign to a given application. The console offers an administrator a single view of the application environment and the capacity assigned to it - whether those applications be an active directory, Microsoft Exchange or Web services.
"Now, when I want to deploy an application..., I just choose to create a VM, choose the CPU, memory and from a pool of resources, and then assign disk capacity," Ready said. "What we've done in bringing the concept of server and virtualization together with a shared storage system is we've eliminated a tremendous amount of complexity. Now, the paradigm is that applications will be communicating with storage that's direct attached."
The HC3 uses RedHat's KVM (kernel-based virtual machine) as its hypervisor on the backend; any platform that can communicate with KVM, such as Linux and Windows, can also run on the HC3, Ready said.
The HC3 has a starting price of $25,500, which includes three nodes with 96GB of RAM, three quad-core Intel CPUs and 24TB of storage. That configuration can support up to 30 VM workloads.
A four-node system, with 128GB of RAM, 16 quad-core Intel processors and 32TB of raw storage, has a price of $34,000.
Because the nodes are clustered, storage, CPUs or memory can be added or removed non-disruptively. Ready contends that a VM can be created in as little as 15 seconds on the HC3.
Jefferson Davis, technology and information systems manager at the Standard School District in Bakersfield, Calif., is beta testing the HC3. Davis said set up and management of the appliance was much simpler than that for rival offerings, and at a "much lower price point."
"HC3 saved us about 75% of what the other solutions quoted us for storage, licenses and additional servers," Davis said. "And, we looked at all of the 'usual suspects' before finding Scale Computing's HC3."
Jeff Boles, an analyst with market research firm the Taneja Group, said the HC3's virtual server configuration time can be an eighth as much time as needed to configure traditional vSphere infrastructures, and it doesn't need specialized virtualisation administrators.
"In our assessment, HC3 may take as little as one-tenth the effort to set up and install compared to traditional infrastructure, one-fourth the effort to configure and deploy VMs versus traditional infrastructure, and can eliminate the planning, troubleshooting and reconfiguration exercises that can consume as much as 25% to 50% of IT administrator time in the midmarket," Boles said.
Lucas Mearian covers storage, disaster recovery and business continuity, financial services infrastructure and health care IT for Computerworld. Follow Lucas on Twitter at @lucasmearian or subscribe to Lucas's RSS feed. His e-mail address is email@example.com.
Read more about data storage in Computerworld's Data Storage Topic Centre.