Cobus Smit – Updated 10 Feb 2020
Huawei FusionCube Info
Huawei FusionCube -Huawei FusionCube solution claims to be “Any Workload, One System” Let’s have a look at their HCI offerings.
Huawei FusionCube is an enterprise-grade data center hyper-converged infrastructure. It applies to scenarios such as an enterprise private cloud, virtualization platform, desktop cloud, branch offices, and data warehouse. FusionCube hyper-convergence delivers industry-leading performance by leveraging distributed storage and powerful hardware acceleration technologies. FusionCube has the largest market share in China and enabled Huawei to enter the Challengers quadrant in the Gartner Magic Quadrant for Hyper-Converged Infrastructure.
Their different HCI hardware platforms below. Updated Feb 2020
To download their Specification – click here
The above specs are according to Huawei. The 9000 model allows you to get up to 32 processors in one chassis, 12 terabytes of RAM, and 172 terabytes for data storage. FusionCube 6000 is a smaller solution – up to 8 processors here, up to 2 terabytes of RAM per module and up to 280 terabytes of storage.
The cluster scale-out storage engine provides storage virtualization of a three-tier (Memory-SSD-DAS) Software defined storage.
The data is striped and stored across all disks in FusionCube, increasing both the utilization of the disks’ storage and I/O performance for different applications.
The on-board storage system software combines the local storage resources of general-purpose servers into fully distributed storage pools, providing block, object, or file storage services to the upper layers of the stack via industry-standard protocols—SCSI, iSCSI, NFS, SMB, FTP, HTTP, S3, Swift, Cinder, RESTful API, and Manila. FusionStorage provides IOPS, bandwidth, and expansion capabilities for unstructured, structured, and semi-structured data. Active-active clustering, redirect-on-write snapshots, thin provisioning, remote replication, erasure coding, and other functionality work together to meet the needs of enterprises. Huawei provides open APIs via standardized protocols, integrating into Open-stack cloud-based architectures and Hadoop systems.
Another feature around their storage is Infiniband (only on E9000) and caching. It uses SSD SSD PCI cards. All of their HCI solutions run FusionStorage which is their software-defined solution. They have their external block storage, as well called oceanstor. Let’s recap, For their Public cloud they use FusionSphere (Cinder-block and Swift-object and Manilla-file) from Openstack. For their external block they use the Huawei Oceanstor portfolio and for their SDS they use FusionStorage.
If we have a look at their integrated storage solution, then it is pretty much the same as a lot of other bigger HCI vendors out there like vSAN from VMware, NDFS from Nutanix, HDX Data Platform from Cisco and Direct spaces from Microsoft. This gives them an additional advantage over some smaller HCI players. The recent FusionStorage platform delivers high performance and availability at consistently low response times in a mixed workload and virtualized environments.
They support most major virtualization platforms out there and have their own Huawei FusionSphere based on KVM just like some other vendors out there like Nutanix as their virtualization is also based on KVM. This with their own FusionStorage software defined storage makes them stand out and enable them to directly compete against Nutanix and DELL EMC on a whole Huawei solution. And on top of that they can compete with AWS, Azure and GCP on a public cloud level.
You also have some choice and do not have to use everything they have and can customize your solutions where you cannot on some other mostly smaller HCI vendors.
Following the announcement of its AI strategy and full-stack, all-scenario AI solutions in 2018, Huawei launched the AI-Native database GaussDB and the highest-performance distributed storage FusionStorage 8.0 on May 2019 in Beijing.
GaussDB uses the integration of AI capabilities into distributed databases. This is making the self-O&M, self-tuning, self-diagnosis, and self-healing possible. In online analytical processing (OLAP), online transaction processing (OLTP), and hybrid transaction/analytical processing (HTAP) scenarios, GaussDB uses the optimality theory to create the industry’s first reinforcement learning self-tuning algorithm, improving tuning performance by over 60%.
GaussDB supports multiple deployment scenarios, including local deployment and deployment on private or public clouds. On HUAWEI CLOUD.
Some of their solutions hardware below:
- FusionCube 9000
- FusionCube 6000
- FusionCube 6000C
- FusionCube 2000
FusionServer RH Series Rack Servers
- FusionServer RH8100 V3 Rack Server
- FusionServer RH5885H V3 Rack Server
- FusionServer RH5885 V3 Rack Server
- FusionServer 5288 V3 Rack Server
- FusionServer RH2288H V3 Rack Server
- FusionServer RH2288 V3 Rack Server
- FusionServer RH1288 V3 Rack Server
- N2000 V3 Series NAS Storage System
FusionServer E Series Blade Servers
- E9000 Converged Architecture Blade Server
- FusionServer CH242 V3 Compute Node
- FusionServer CH226 V3 Storage Expansion Compute Node
- FusionServer CH225 V3 All-Flash Compute Node
- FusionServer CH222 V3 Storage Expansion Compute Node
- FusionServer CH220 V3 I/O Expansion Compute Node
- FusionServer CH140 V3 Compute Node
- FusionServer CH121 V3 Compute Node
- FusionServer CH140L V3 Compute Node
- FusionServer CH121L V3 Compute Node
FusionServer X Series High-Density Servers
- FusionServer X6800 Data Center Server
- FusionServer XH628 V3 Server Node
- FusionServer XH622 V3 Server Node
- FusionServer XH620 V3 Server Node
- FusionServer X6000 High-Density Server
- FusionServer XH321 V3 Server Node
Server Management Software
- iBMC, Server Intelligent Management System
- FusionServer Management & Integration Tools
This is a proper HCI solution and is growing and integrating rapidly. Just be sure you check support for your country and region.
They use FusionCube builder as their installation tool and you can say everything is in there and not dependent on any other vendor. Even If you want to go the the public cloud you can keep everything Huawei
My opinion on Huawei is ” watch this space” they have all the right solutions, hardware and software required to build a full scale cloud to compete with Dell Technologies and Nutanix on a private cloud model and with AWS, Azure and GCP on a public cloud level. The future for Huawei is looking good and on this specific Hyper-Converged topic as well. They are growing fast and have the potential to be the top player in the world.
“Some images are the Property of Huawei Technologies“