このサイトはCookieを使用しています。 サイトを閲覧し続けることで、Cookieの使用に同意したものとみなされます。プライバシーポリシーを読む>
このサイトはCookieを使用しています。 サイトを閲覧し続けることで、Cookieの使用に同意したものとみなされます。プライバシーポリシーを読む>
OceanStor 100D
Huawei OceanStor 100D (originally called FusionStorage) is an intelligent, in-house developed distributed storage product with scale-out to support the business needs of both today and tomorrow. One cluster provides industry standard interfaces, such as block, HDFS, object, and file storage, for upper-layer applications. This greatly simplifies operations complicated by siloed storage systems, offering you everlasting operations for services, efficient processing for diverse data, and economical storage for mass data.
Block storage provides standard Small Computer System Interfaces (SCSIs) and Internet SCSIs (iSCSIs) and storage resource pools that support on-demand resource provisioning and elastic capacity expansion, making the storage platform ideal for private clouds, containers, virtualization platforms, and databases.
HDFS storage provides a decoupled storage-compute big data solution using native HDFS for on-demand resource configuration, slashing Total Cost of Ownership (TCO) and delivering a favorable user experience. While support for coexisting coupled and decoupled storage-compute architectures ensures seamless evolution, HDFS storage has been widely adopted for big data applications in finance, carrier, and e-government projects.
Object storage, enabling a single bucket to house up to 100 billion objects without compromising performance, is extensively used in scenarios involving bank check images, audio and video recordings, medical images, B2B electronic documents, and Internet of Vehicles (IoV).
OceanStor 100D uses E2E reliability assurance approaches at I/O, device, system, and data center levels to offer highly available storage services for complex enterprise applications. Specifically, OceanStor 100D supports E2E data integrity check and self-healing. It provides comprehensive sub-health detection and self-healing for disks, nodes, and networks. System reliability remains uncompromised even if four nodes or four cabinets simultaneously fail. OceanStor 100D also offers active-active and replication solutions to ensure data center reliability.
Huawei OceanStor 100D offers best-in-class performance of 140,000 IOPS/node at a consistent latency of 1 ms*, thanks to key technologies such as FlashLink® performance acceleration and AI Fabric lossless network. Stable performance is possible even when 100 billion objects are hosted in a single bucket. It implements converged management of multiple storage services, such as block, HDFS, object, and file, as well as efficient data usage. What's more, it allows you to run both old and new generations of hardware, and multiple platform hardware products for intelligent system management throughout the lifecycle.
Each cluster can support between 3 and 4096 nodes, to allow for on-demand and elastic scalability. To preserve current investments, performance and capacity increase on a linear scale. Meanwhile, data redundancy protection of elastic Erasure Coding (EC) utilizes 91.6% of disk space* for a 2.75x improvement over the traditional three-copy mode, while dynamic deduplication and compression provide high availability and efficiency.
| Model |
OceanStor 100D |
||
|
Comprehensive Features |
|||
| System Architecture | Fully distributed architecture | ||
| Scalability | 3 to 4096 nodes | ||
| Multiple Storage Pools | 1 to 128 pools | ||
|
Detailed Features |
|||
| Storage Service Type | Block Storage | HDFS Storage | Object Storage |
| Supported Protocols | SCSI, iSCSI, and OpenStack Cinder | HDFS | Amazon S3 and NFS |
| Data Redundancy Protection | EC and multi-copy modes | EC | EC |
| Key Features | • Snapshot • Linked clone • Cross-cluster active-active protection • Asynchronous replication • DIF • Thin provisioning • QoS • Dynamic deduplication and compression • Decoupled storage-compute deployment • Coupled storage-compute deployment |
• Multiple namespaces, with Kerberos and LDAP authentication • Directory- and namespace-level quotas • Data tiering • DIF • Recycle bin • Compatibility with the local HDFS |
• Unified namespace, with multiple tenants and resource pools • Small-object aggregation • QoS • Data lifecycle management • Data consistency check • Asynchronous replication • Object-level deduplication • WORM |
| Network Types | 10 GE or 25 GE TCP/IP 25 GE RoCE 56 Gbit/s or 100 Gbit/s InfiniBand |
10 GE or 25 GE RoCE 10 GE or 25 GE TCP/IP |
10 GE or 25 GE TCP/IP |
| Compatible Platforms and Components | VMware vSphere, Microsoft Windows Hyper-V, OpenStack, and containers* | Mainstream big data components: MapReduce, Hive, HBase, and Spark | Multiple backup software products: Commvault, NBU, Atempo, Veeam, and Rubrik |
| Recommended Node Models | F100 all-flash, P100 performance, and C100 capacity models | P100 performance and C100 capacity models | P100 performance and C100 capacity models |
* For more information, please contact your local Huawei sales representatives.
China Merchants Bank: OceanStor Distributed Storage Hosts Critical and Emerging Workloads
China Mobile: OceanStor Distributed Storage Accelerates Core Workload Upgrades
