In recent years, hyperconverged storage has become really hot. According to Gartner, the hyperconverged integrated systems market will be worth $371.5 million in 2014 and could grow to nearly $5 billion by 2019. This increase, the analyst firm said, is being driven by demand for a more cost effective and operationally streamlined approach to virtual servers, virtual desktops and hybrid cloud infrastructure.
As a result, numerous vendors are bringing products, many of them appliances, to market in this space. Here are some of the candidates:
The HPE Hyper Converged 380 is a compute, software-defined storage and virtualization appliance built on the HPE ProLiant DL380 server. It incorporates several types of HPE software. The user experience relies on HPE OneView and reportedly makes it possible to deploy VMs with five clicks. Its software-defined architecture leverages HPE StoreVirtual VSA, and its analytics utilize HPE Cloud Optimizer to forecast required capacity and allocation of resources, which HPE says can cut overprovisioning of VMs by 90 percent. It is aimed at mid-sized and the remote office/branch offices (ROBO) market.
“Hyperconvergence presents an opportunity for mid-sized and ROBO enterprises who face the challenge of having to stay competitive at enterprise-scale, but with less resources,” said Ric Lewis, senior vice president and general manager, Converged Datacenter Infrastructure, HPE.
Maxta Storage Platform is a hypervisor-agnostic storage platform that supports private, public and hybrid clouds. Users can deploy software-only MxSP on top of any x86 servers to convert them into a converged compute and storage platform. MaxDeploy appliances offer essential the same arrangement, but the appliances are pre-integrated and delivered by Maxta distributers and channel partners. A configuration tool helps users figure out the MaxDeploy appliance that is the most appropriate for their needs.
“Some system vendors have integrated and re-branded Maxta into their own line of hyperconverged appliances,” said Yoram Novick, founder and CEO of Maxta. “Our pricing model is based on the number of CPU cores and the amount of storage capacity managed.”
Scale Computing HC3 clusters consist of multiple nodes built from three configuration tiers: HC1000, HC2000 and HC4000. They all converge virtualization, compute, storage and availability into a unified solution, with each having a different specification in terms of CPU, RAM and disks. Pricing for a mid-level, 3-node HC2000 starts at $35,100, which includes 18 CPU cores, 192 GB RAM, 7.2 TB raw storage capacity and a year of phone support and next hardware replacement.
“We are focused on providing virtualization for small and midmarket customers who can take advantage of the simplicity, scalability, availability and affordability of hyperconverged infrastructure,” said Scale Computing co-founder and chief evangelist Jason Collier. “We can make infrastructure an out-of-the-box solution for IT managers that don’t have the time or budget for dealing with complex hardware issues.”
Atlantis delivers a software-defined storage (SDS) platform called Atlantis USX and a pre-integrated hyperconverged appliance called Atlantis HyperScale that is based on the same technology. Atlantis USX uses a combination of existing data center infrastructure and hyperconverged building blocks, and it comes with inline data reduction to eliminate storage traffic before data is written to storage. In addition, its Stretched Cluster feature is said to enable zero recovery point objective (RPO) and recovery time objective (RTO) between multiple data centers up to 100 miles apart.
“With Stretched Cluster, businesses have the ability to tolerate failure of a data center without experiencing any disruption in application availability or loss of data,” said Seth Knox, vice president of product marketing, Atlantis Computing.
The Atlantis HyperScale appliance offers a choice of server platform from Cisco, Dell, HP, Lenovo or Supermicro, as well as multiple hypervisor platforms, with support included. The HyperScale CX-4, for example, is a two-node hyperconverged integrated system designed for ROBO and very small data centers. It features 4TB of all-flash storage and can be upgraded in the field to a 4-node CX-12 or CX-24 with 12TB or 24TB all-flash capacity. Pricing starts at $43,000 for a two-node Atlantis HyperScale CX-4 configuration including three years of support.
The Nutanix enterprise cloud platform is comprised of Nutanix Acropolis and Nutanix Prism. Acropolis converges compute, storage and virtualization resources and also integrates three key Nutanix technologies: Distributed Storage Fabric (DSF), App Mobility Fabric (AMF) and Acropolis Hypervisor (AHV) for virtualization.
“AMF allows you to move applications across different runtime environments based on business needs,” said Prabu Rambadran, director of product marketing, Nutanix. “Prism is a management solution built to simplify data center storage and virtualization operations.”
Gridstore’s appliance has up to four compute/storage nodes and can be expanded with additional appliances and/or Gridstore Storage Nodes for up to 256 total nodes. The company says it is best suited to low-latency and high transactional applications.
The All-Flash HyperConverged Appliance contains four hot-swappable servers and 24 solid state drives (SSDs) for a combined total of between 48 and 96 Intel V3 cores and between 23TB and 92TB of flash capacity in a 2U form factor.
“Compared to traditional infrastructure, the Gridstore All-Flash Hyperconverged Appliance delivers VMs that run five times faster, cost 65 percent less and reduce the data center footprint by 75 percent,” said Kelly Murphy, founder and CTO, Gridstore.
VSkyCube is Promise Technology’s hyperconvergence platform software that runs on commodity hardware. Ken Dai, the company’s general manager, said it allows storage usage to be defined and allocated on demand at application deployment time. Specific storage attributes, such as performance levels, protection options and others, can also be specified by individual applications.
Dell XC Series appliances offer Nutanix-powered hyperconverged infrastructure sitting on top of Dell PowerEdge servers. They are also certified to integrate well with the SAP NetWeaver technology platform running on Linux.
In addition, Dell is partnering with EMC to resell the VCE VxRail family of appliances. VxRail is the only hyper-converged appliance integrated, preconfigured and pre-tested for VMware environments. It combines EMC data services and systems management capabilities with VMware software.
SimpliVity hyperconverged appliances take the company’s OmniStack 3.5 software and combine it with Cisco UCS or Lenovo System x hardware. OmniStack includes Intelligent Workload Optimizer to help predict application performance using vSphere Distributed Resource Scheduler. It also supports Hyper-V environments within Microsoft Windows Server 2016. A web-based tool, OmniView, provides predictive analytics and trend analysis via customized dashboards.