Huawei FusionCube Hyper-Converged Infrastructure

Huawei FusionCube is a pre-integrated solution designed to simplify deploying a powerful, virtualized IT infrastructure. It features fully converged computing, storage and networking, with a highly automated virtualization and management system. The resulting system improves application and business efficiency beyond what was previously possible.

FusionCube reduces capital expenses as a result of higher utilization, less cabling, and fewer network connections, while also reducing operating costs by reducing administration labor through automated data center management. By unifying Ethernet and Fiber Channel networking, all components in the data center can be managed from a single console. FusionCube can even allow the consolidation of storage and network management teams. In addition to providing linear increases of computing and storage resources, FusionCube also provides a linear increase in system caching, by pooling memory and SSD storage across all devices in the cube, increasing application performance.

FusionCube is ideal for server virtualization, desktop virtualization, private clouds and application hosting. Expanding existing cloud data centers will also be much easier when using FusionCube.

Huawei FusionCube Highlights

 

Highly Converged Platform

FusionCube uses an innovative hardware platform composed of flexible slots for compute/storage blades and GE/10GE/FC switching modules. The compute and storage resource can be configured to adapt to various workloads. A single 12U chassis supports up to 64 CPUs and 12.3 TB of memory, making it especially suitable for situations requiring high compute density and virtualization. Converged storage and SSD cache accelerate application and database performance.

Elastic cloud platform

The cloud OS software virtualizes and schedules compute, storage and network resource and provides IaaS services such as elastic computing, load balancing and virtual private cloud.

The unified management system allows access to everything in FusionCube, from switch to VM, from storage volumes to applications, from provisioning to automation and security.

The scale of FusionCube systems that can be managed by the cloud platform range from a single box up to a complete data center. The management system also supports auto-discovery, so expanding the cloud is as simple as plugging in a new cube. Simple, but powerful.

Accelerated storage engine

The cluster scale-out storage engine provides storage virtualization of three-tier (Memory-SSD-DAS) storage and eliminates the vulnerable RAID controller while providing high availability and scalability.

The data is striped and stored across all (hundreds or thousands of) disks in FusionCube, increasing both the utilization of the disks’ storage and I/O performance for different applications. The storage engine delivers 3X-5X higher peak time IOPS to applications with same number of disks, improving performance 100%- 300%. Utilizing advanced features such as thin provisioning do not reduce performance. On-demand scale-out cluster architecture enables you to handle new virtualization needs without re-planning or rearchitecting.

FusionCube 9000

FusionCube 6000

 

Specifications

 

FusionCube 9000

Application Scenarios

Virtualization, cloud, hybrid load, and VDI

Data warehouse, database integration, and midrange computer replacement

Node Specifications

2-socket compute node (2 x E5-2600 V3 CPUs)

4-socket compute node (4 x E7-4800 V2 or E7-4800/8800 V3 CPUs)

2-socket compute and storage converged node (2 x E5-2600 V3 CPUs, 2–6 x PCIe SSDs with 2.4 TB/3.2 TB per SSD)

2-socket compute and storage converged node (2 x E5-2600 V3 CPUs, 12 x SAS HDDs)

Database node: 2-socket database node (2 x E5-2600 V3 CPUs), or 4-socket database node (4 x E7-4800 V2 or E7-4800/8800 V3 CPUs)

Storage node: Large-capacity storage node (12 x 1.2 TB/1.8 TB SAS disks, 3.2 TB SSD cache, 2 x 10 GE ports, 2 x 56G IB ports), High-performance storage node (4/6 x 3.2 TB SSDs, 2 x 10 GE ports, 4 x 56G IB ports)

Management node: 1 or 2 management nodes (2 x E5-2620 V3 CPUs, 128 GB memory, 2 x 10 GE ports)

Network

10 GE, IB 56G FDR switching (storage network)

10 GE, IB 56G FDR switching (storage network)

Storage

FusionCube distributed block storage

FusionCube distributed block storage

Maximum Specifications

256 server nodes

10 chassis

Management Software

FusionCube Center

FusionCube Center

Software

FusionSphere 5.1/VMware vSphere 5.5 or later

Oracle DB, IBM DB2, SAP Sybase IQ, and Dameng database

Cabinet Height

42U (customizable)

42U (customizable)

Power Consumption

6,000W (maximum per chassis)

6,000W (maximum per chassis)


FusionCube 6000

Model

FusionCube 6000VH

FusionCube 6000VS

Application Scenario

Virtualization and cloud

Virtualization and cloud

Basic Package

72 cores, 768 GB memory, 60 TB storage, and 2,400 GB SSD cache

48 cores, 384 GB memory, 36 TB storage, and 1,800 GB SSD cache

Expansion Package

24 cores, 256 GB memory, 20 TB storage, and 800 GB SSD cache

16 cores, 128 GB memory, 12 TB storage, and 600 GB SSD cache

Basic Package Configuration

3 nodes (configuration of each node: 2 x E5-2658A V3 CPUs with 12 cores, 8 x 32 GB DIMMs, 10 x 2 TB SATA HDDs, 800 GB SSDs, 2 x 10 GE ports), and chassis components

3 nodes (configuration of each node: 2 x E5-2630 V3 CPUs with 8 cores, 8 x 16 GB DIMMs, 6 x 2 TB SATA HDDs, 600 GB SSDs, 2 x 10 GE ports), and chassis components

Expansion Package Configuration

1 node (2 x E5-2658A V3 CPUs, 8 x 32 GB DIMMs, 10 x 2 TB SATA HDDs, 800 GB SSDs, and 2 x 10 GE ports)

1 node (2 x E5-2630 V3 CPUs, 8 x 16 GB DIMMs, 6 x 2 TB SATA HDDs, 600 GB SSDs, and 2 x 10 GE ports)

Network

10 GE

10 GE

Maximum Number of Nodes

256 nodes/64 chassis (4 nodes per chassis)

256 nodes/64 chassis (4 nodes per chassis)

Storage

FusionCube distributed block storage

FusionCube distributed block storage

Virtualization Platform

FusionSphere 5.1 or later/VMware vSphere 5.5 or 6.0

FusionSphere 5.1 or later/VMware vSphere 5.5 or 6.0

Management Software

FusionCube Center 2.5

FusionCube Center 2.5

Cabinet Height

4U/19 inches (447 mm x 175 mm x 898 mm)

4U/19 inches (447 mm x 175 mm x 898 mm)

Power Consumption

60% load: 1,000W (3 nodes), 1,300W (4 nodes)

60% load: 740W (3 nodes), 960W (4 nodes)

Additional Information

Basic package (100 VMs), expansion package (40 VMs)

VM specifications: 1 x vCPU, 4 GB memory, 150 GB storage (1 core = 2 vCPUs)

Basic package (40 VMs), expansion package (20 VMs)

VM specifications: 1 x vCPU, 4 GB memory, 150 GB storage (1 core = 2 vCPUs)

 

Market Analysis

 

Market Treand

Gartner Says Hyperconverged Integrated Systems Will Be Mainstream in Five Years.

The market for hyperconverged integrated systems (HCIS) will grow 79 percent to reach almost $2 billion in 2016, propelling it toward mainstream use in the next five years, according to Gartner, Inc.

HCIS will be the fastest-growing segment of the overall market for integrated systems, reaching almost $5 billion, which is 24 percent of the market, by 2019. Although the overall integrated systems market is growing, other segments of the market will face cannibalization from hyperconverged systems, Gartner analysts said.

Gartner defines HCIS as a platform offering shared compute and storage resources, based on software-defined storage, software-defined compute, commodity hardware and a unified management interface. Hyperconverged systems deliver their main value through software tools, commoditizing the underlying hardware.

Phase 1 is the peak period of blade systems (2005 to 2015), Phase 2 marked the arrival of converged infrastructures and the advent of HCIS for specific use cases (2010 to 2020), and Phase 3 represents continuous application and microservices delivery on HCIS platforms (2016 to 2025).

The third phase of integrated systems will deliver dynamic, composable and fabric-based infrastructures by also offering modular and disaggregated hardware building blocks, driving continuous application delivery and continuous economic optimization.

Despite high market growth rates, HCIS use cases have so far been limited, causing silos with existing infrastructure, according to Gartner. Its progression will be dependent on multiple hardware and software advances, such as networking and software-defined enterprises.

Ultimately, the underlying infrastructure will disappear to become a malleable utility under the control of software intelligence and automated to enable IT as a service (ITaaS) to business, consumer, developer and enterprise operations.

Vendor List

CRN has complied a very comprehensive list of hyper-converged infrastructure products. Here are they for you information:

Cisco HyperFlex

Cisco HyperFlex price HyperConverged Infrastructure

Cisco HyperFlex systems, Cisco's first hyper-converged infrastructure solution, combines Cisco UCS servers, Cisco networking and software-defined storage technology from SpringPath in a pre-integrated cluster that scales resources independently. Cisco said it can be up and running, including the networking, in less than an hour.

The Cisco HX Data Platform Controller combines the entire cluster’s SSDs and spinning disks into a single, distributed, multitier, object-based data store featuring real-time, always-on deduplication and compression to help cut storage costs by up to 80 percent without affecting performance. It also includes enterprise-class data management features, such as snapshots, thin provisioning and smart cloning.

Dell Updated Nutanix-Based Solution

Dell Nutanix Solution price

Dell, Round Rock, Texas, in April updated its Dell XC hyper-converged infrastructure appliances with the addition of Intel's new Broadwell Xeon E5-2600 v4 processors, which are now also available with Dell PowerEdge servers to boost performance. The XC series, first introduced in late 2014, is also now certified to be used in SAP's NetWeaver solutions.

Dell's XC series is based on the Nutanix Acropolis software stack, and is available as 1U or 2U appliances with up to four dual-processor nodes, choice of Intel processors, and flexible memory and storage configuration including SSDs and hard drive. The XC Series also includes all-flash, GPU, storage-only, and small form factor nodes to support a wide range of use cases and workloads.

EMC VCE VxRail

EMC VCE VxRail price

The VCE VxRail appliance family from Hopkinton, Mass.-based EMC, introduced in February, is an integrated hyper-converged infrastructure solutions specifically for VMware environments. VxRail Appliances bring together EMC's data services and systems management capabilities with VMware’s hyper-converged software -- including VMware vSphere, vCenter Server and VMware Virtual SAN -- in a single product family with one point of support.

The VxRail appliances are available in hybrid storage or all-flash versions, and come loaded with integrated EMC mission-critical data services including replication, backup and cloud tiering at no additional charge. The also integrate well with other EMC solutions, including EMC RecoverPoint for Virtual Machines, Virtual SAN active-active stretch clusters and vSphere Data Protection. List prices start at $60,000.

HPE Hyper-Converged Infrastructure Based On ProLiant DL380 Servers

HP ProLiant DL380 Hyper-Converged Infrastructure price

Palo Alto, Calif.-based Hewlett Packard Enterprise in March launched its latest entry into the hyper-converged infrastructure market, one aimed at providing an enterprise-grade solution in a market where commodity hardware is the norm.

The new offering, the Hyper Converged 380, combines HPE's enterprise-class DL380 servers and a software stack that includes HPE's StoreVirtual VSA software-defined storage application and the HPE OneView converged management and automation application. HPE called it a solution combining enterprise-class features and scalability with an easy-to-use software stack, particularly the HPE OneView management software, which the company said can spin up virtual machines in five clicks.

Nutanix Acropolis 4.6

Nutanix Acropolis price

San Jose, Calif.-based Nutanix in February updated its Nutanix Acropolis scale-out data fabric for storage, compute and virtualization with more than 25 enhancements to increase system performance by up to four times compared with earlier versions. The software powers up to 30,000 Microsoft Exchange mailboxes in 8U of rack space.

When used with the NX-9060-G4 all-flash appliance, which can be configured with up to six 1.6-TB SSDs and up to 512 GBs of memory across four nodes, Nutanix Acropolis can provide over 1 million storage IOPS in just 4U of data center space, the company said.

Oracle SuperCluster M7

Oracle SuperCluster M7 price

The Oracle SuperCluster M7 engineered system, introduced in October by Redwood Shores, Calif.-based Oracle, combines compute, networking and storage hardware with virtualization, operating system and management software in a complete, high-performance cloud infrastructure. The solution incorporates advanced processor-level security features, and works in a range of high-performance workload environments, from traditional ERP, CRM and data warehousing to e-commerce, mobile applications and real-time analytics.

The SuperCluster M7, based on Oracle's Sparc M7 processor, includes the Solaris operating system, Oracle VM Server for Sparc and the SuperCluster Management Toolset.

 

Supermicro Hyper-Converged Infrastructure Solutions

SuperMicro Hyper-converged Infrastructure product  price

San Jose, Calif.-based Supermicro is one of the top suppliers of industry-standard x86-based servers used in hyper-converged infrastructure solutions. Compute, network and storage resources are integrated to provide a complete solution when used with hyper-converged infrastructure software stacks from nearly any vendor.

Supermicro solutions can be configured for all-flash-based NVMe or SSD storage for tier-1 mission critical applications, hybrid flash storage and hard drive storage for second-tier or third-tier applications, or for disaster recovery or virtual desktop infrastructure applications

VMware Hyper-Converged Software (HCS)

Vmware Hyper-converged software VCE price

Palo Alto, Calif.-based VMware's Hyper-Converged Software (HCS), introduced in February, adds virtualized compute, storage and networking resources to standard x86-based servers. VMware HCS includes the company's vSphere hypervisor, Virtual SAN (VSAN) storage technology, and the vCenter Server unified management solution.

VSAN 6.2 delivers deduplication, compression and erasure coding capabilities to enable the building of all-flash hyper-converged infrastructure appliances. Also new with VSAN 6.2 is stretched clusters, quality of service controls on a per-virtual machine basis, and vCenter management.

VMware HCS is at the heart of the new VCE VxRail appliances. The software is also being used on more than 100 certified Virtual SAN Ready Nodes from all major server vendors

Lenovo Converged HX Series

Lenovo Converged HX price

The Lenovo Converged HX Series uses the Nutanix software stack to bring compute, storage and virtualization resources into a single appliance. It was designed to rapidly scale to meet the needs of small and midsize businesses and enterprise IT departments.

Beijing-based Lenovo in January introduced three models in the Lenovo HX Series. The HX7500 targets high-performance workloads such as Microsoft Exchange, Microsoft SharePoint and databases such as Microsoft SQL Server with higher-end processors, four SSDs and 20 2.5-inch HDDs. The HX5500 targets storage-heavy workloads such as file servers, Splunk and data center backups with up to 36 TBs of capacity. The HX3500 targets compute-heavy environments with a mix of two SSDs and six 2.5-inch HDDs.

 

Atlantis HyperScale CX-4 Atlantis HyperScale CX-4

Atlantis HyperScale CX-4

The Atlantis HyperScale CX-4 appliance from Mountain View, Calif.-based Atlantis Computing is a two-node hyper-converged system integrated with compute, all-flash storage, networking and virtualization for remote offices and branch offices and small data centers. Introduced just this past January, the HyperScale CX-4 appliance features a starting price of $43,000 for a two-node configuration with 4 TBs of effective all-flash capacity and three years of premium support. Partners get a simplified deployment process which pre-configures networking for each remote appliance to connect to a central Atlantis Manager virtual appliance. The Atlantis CX-4 can replicate data to another Atlantis hardware appliance or to a software-defined storage (SDS) solution at a secondary site.

Breqwatr Cloud Appliance

Breqwatr Cloud Appliance

Toronto-based Breqwatr is targeting enterprise private cloud requirements with the Breqwatr Cloud Appliance, a subscription-based hyper-converged cloud appliance that provides Infrastructure-as-a-Service capabilities without the need for an on-premises solution. Under the hood, the company has packed a curated version of OpenStack delivered in a 2U, four-node, shared nothing hyper-converged cluster that scales linearly as nodes are added. It supports NetApp and SolidFire solutions for expanded capacity or performance requirement, and is slated to support other Cinder-compliant arrays from EMC, Hewlett Packard Enterprise and Hitachi Data Systems in the near future.

The Breqwatr Cloud Appliance offers a complete cloud solution in an on-premises appliance with automation, orchestration, cloud management and a consumerized self-service portal.

DataDirect Networks (DDN) SFA14KE

DataDirect Networks DDN SFA14KE price

The DDN SFA14KE, introduced in November by Chatsworth, Calif.-based DataDirect networks, uses a combination of high-performance Intel processors, embedded networking, and NVMe SSDs to deliver 60 GBs per second of throughput and 6 million IOPS. That translates to 600 GB/ss and 60 million IOPS in a single rack. The SFA platform can be configured for up to 7 petabytes of capacity per rack, and scales out to over 100 petabytes of capacity.

Fujitsu Primeflex For VMware VSAN

Fujitsu Primeflex price Vmware VSAN

Under the Primeflex for VMware VSAN brand, Fujitsu offers a variety of validated server-based Virtual SAN Ready Node configurations for deploying hyper-converged infrastructures. For all configurations, Fujitsu provides the hyper-converged software stack, including VMware vSphere and Virtual SAN, already pre-installed on a flash boot device.

The Primeflex for VMware VSAN CX configuration variant is based on Fujitsu Primergy CX400 M1 systems providing a modular, high-density design for scale-out hyper-converged infrastructure solutions. Each Primergy CX400 M1 system is configured with up to four half-wide dual-socket servers, delivering a total of 8 Intel Xeon processors, 64 DDR4 memory DIMMs, and 24 disks, all in a 2U enclosure.

Gridstore HyperConverged Appliance

Gridstore HyperConverge price

The HyperConverged Appliances from Mountain View, Calif.-based Gridstore include compute and storage in a single system, and allow nodes to be either compute and storage or storage-only. The solution scales from three nodes to 256 notes. The company's software architecture delivers native Windows integration, per-virtual machine I/O control, and elastic and independent scaling of resources.

Gridstore’s HyperConverged Appliances deploy quickly for customers looking at infrastructure refreshes or at specific workloads such as virtual desktops, SQL Server, test/dev, private cloud and ROBO (remote office and branch office).

Hitachi Data Systems Hyper Scale-Out Platform

Hitachi Data System Hyper Scale-out price

HDS' Hyper Scale-Out Platform (HSP), introduced by the Santa Clara, Calif.-based company about a year ago, condenses server computing, storage, networking and server virtualization into a hyper-converged scale-out appliance to help customers implement a data lake. It integrates high-performance NFS storage with Hadoop, Spark and Pentaho software to provide users insight from mixed data-in-place analytics workloads.

HSP brings together Pentaho data integration and business analytics server software to streamline data blending, Hortonworks Data Platform for certified Hadoop data services, and HDS' own distributed file system and management tools for high-speed data ingest and sharing of data between virtual containers and Hadoop applications.

Infrascale Cloud Failover Appliance

Infrascale Cloud Failover Appliance price

Los Angeles-based Infrascale last August introduced its Cloud Failover Appliance, a hyper-converged infrastructure solution providing organizations of all sizes with quick restoring of critical systems, applications and data.

Delivered as a physical or virtual appliance, the Cloud Failover Appliance includes disaster recovery software, virtualization services, backup software, WAN acceleration, data archiving with retention policies, and cloud failover services. Critical business applications can be virtualized and spun up from the appliance or within a public or private cloud in a guaranteed 15 minutes. Data can be intelligently spilled over to the cloud, and all data is replicated to the cloud and automatically removed locally based on policies.

Maxta Storage Platform

Maxta Storage Platform price

The Maxta Storage Platform from Sunnyvale, Calif.-based Maxta is a hypervisor-agnostic resilient storage platform for virtualized data centers that turns standard servers into a converged compute and storage solution by leveraging server-side flash and disk drives to optimize performance and capacity. Maxta's distributed architecture works in VMware vSphere or KVM environments to build hyper-converged infrastructure solutions with any industry-standard x86-based servers and any storage configuration. The solutions allow compute and storage resources to be independently scaled.

Promise VSkyCube

Promise VSkyCube price

Promise Technology, Santa Clara, Calif., in April introduced its new VSkyCube hyper-converged infrastructure solution family for streamlining data center and IT operations. The Promise VSkyCube family of five appliances scales from one to 32 nodes, and comes fully integrated with the KVM hypervisor.

On the storage side, VSkyCube manages capacity with SSD caching, thin-provisioning, tiering and dynamic volume expansion, and protects data with RAID5-fashion data striping across nodes (RAIN). It can automatically detect partial failure conditions at block, disk and node levels, and trigger a healing process without disrupting operations.

VSkyCube also includes a centralized single-pane-of-glass management system.

 

Riverbed SteelFusion

Riverbed SteelFusion price

San Francisco-based Riverbed's SteelFusion hyper-converged infrastructure converges server, storage and network infrastructure into a single appliance targeting branch office applications using centralized data. SteelFusion consolidates storage to the data center or an Amazon Web Services or Microsoft Azure cloud to reduce the storage footprint at remote locations.

SteelFusion allows the development of “stateless” branch services, letting users access applications running locally in the branch offices while primary data is stored in the data center to reduce the risk of data loss. This lets SteelFusion deliver local performance and capacity without the need for a storage-intensive branch office solution.

 

Scale Computing HC3

Scale Computing HC3 price

Indianapolis-based Scale Computing’s HC3 and its HyperCore architecture aim to provide highly available and scalable compute and storage services via intelligent software automation and architecture simplification. Rather than using a virtual storage appliance, Scale Computing’s HC3 integrates storage within the kernel to streamline the I/O path for virtual machines while decreasing the required hardware resources, the company said. HyperCore puts intelligence and automation in the software layer to take advantage of low-cost commodity hardware components. These components are clustered together into a single unified and redundant system to operate similar to a redundant and elastic private cloud, with additional nodes automatically incorporated into the cluster.

SimpliVity OmniStack

SimpliVity OmniStack Price

Westborough, Mass-based SimpliVity’s hyper-converged platform, OmniStack, delivers enterprise-class performance and availability. Storage, compute, network efficiency, WAN optimization, data protection and performance, and global unified management capabilities are packed into a 2U, x86-based server with sharable resources.

In addition to SimpliVity's own OmniCube hyper-converged appliance, the company provides its software stack to technology partners like Cisco and Lenovo.

 

Sphere 3D V3

Sphere 3D V3 price

The V3 hyper-converged appliances from San Jose, Calif.-based Sphere 3D feature a workload-optimized virtualization architecture to help ease application deployment and reduce the need for specialized IT staff. V3 Appliances can be dropped into an existing VMware environment and begin serving up workloads in under an hour.

The V3 line includes V3 Smart Nodes for data centers, corporate headquarters, or very large regional campuses and offices; V3 End User Computing (EUC) Nodes for larger remote and branch offices with up to 200 users; V3 Flex Nodes for smaller businesses and branch locations with as few as 10 users; and V3 Desktop Cloud Orchestrator for managing virtual desktop pools on V3 server appliances.

Stratoscale Symphony

Stratoscale Symphony price

Stratoscale Symphony from Hertzliya, Israel-based Stratoscale is a software-only hyper-converged infrastructure solution that runs on any commodity server hardware. Introduced in December, it integrates compute, storage and networking into a shared resource pool, with built-in administrator control and management and user self-service.

Stratoscale Symphony offers full support for OpenStack APIs. The software is based on a microservices architecture that builds up the control and data planes. Microservices are distributed across the nodes of the cluster for resiliency, high availability and continuous scalability. The complete solution can be non-disruptively scaled out by adding additional servers to the cluster.

ZeroStack Cloud Platform

ZeroStack Cloud price

The ZeroStack Cloud Platform, introduced in March by Mountain View, Calif.-based ZeroStack, targets midmarket and large enterprise partners with what the company calls the industry’s first zero-touch private cloud solution.

The solution includes two elements: The ZeroStack Cloud Building Block hyper-converged infrastructure system includes compute, clustered storage, software-defined networking and distributed management software in a 2U, 4-node enclosure. They are connected through OpenStack APIs, and include a symmetric self-healing, scale-out architecture. The ZeroStack Cloud Portal provides a user and administrator Web interface that provides real-time monitoring and predictive analytics, a self-service environment, and a built-in app store with application templates and pre-defined images to simplify deployment.

 

Huawei FusionCube 9000 and 6000

Huawei FusionCube is a hyper-converged infrastructure, which consists of computing, storage, network, virtualization, and management systems. 

High performance, low latency, fast deployment, and built-in distributed storage 

Database support: Oracle, DB2, Sybase IQ, SQL Server, and SAP HANA

Virtualization support: FusionSphere, VMware, and KVM

For more information on Huawei FusionCube, please check Product documentation at ActForNet knowledge base

Access online  lab/demo

Quote on FusionCube 9000

Quote on Huawei FusionCube 9000 for Database Infrastructure

Quote on Huawei FusionCube 6000

Quote on Huawei FusionCube 6000 Typical COnfiguration

Quote on FusionCube Storage Software

Quote on FusionServer LCS

Contact us:Tel: 1-866-318-6636 (US/CANADA)

     Email: [email protected]