Issue / 1.1
Date / 2014-05-31
HUAWEI TECHNOLOGIES CO., LTD.
Copyright © Huawei Technologies Co., Ltd. 2014. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior written consent of Huawei Technologies Co., Ltd.
Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective holders.
Notice
The purchased products, services and features are stipulated by the contract made between Huawei and the customer. All or part of the products, services and features described in this document may not be within the purchase scope or the usage scope. Unless otherwise specified in the contract, all statements, information, and recommendations in this document are provided "AS IS" without warranties, guarantees or representations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the preparation of this document to ensure accuracy of the contents, but all statements, information, and recommendations in this document do not constitute a warranty of any kind, express or implied.
Huawei Technologies Co., Ltd.
Address: / Huawei Industrial Base
Bantian, Longgang
Shenzhen 518129
People's Republic of China
Website: /
Email: /
Issue 1.1 (2014-05-31) / Huawei Proprietary and Confidential Copyright © Huawei Technologies Co., Ltd. / 1
Contents
1 Overview
1.1 Project Background
1.2 Live Network Status
1.3 Customer Requirements
1.4 Construction Scale and Objectives
2 Design Basis and Principles
2.1 Solution Design Basis
2.2 Solution Design Principles
2.3 System Design Scheme
2.4 Topologies
2.5 Software and Hardware Configurations
3 Highlights and Advantages
3.1 Advantages of HUAWEI Enterprise Storage System Design
3.2 Innovative Product Design
3.3 High Performance
3.3.1 High-Speed Bus and I/O Channel
3.3.2 Multiple Controllers, Multi-Core CPUs, and Smart Matrix
3.4 Flexible Scalability
3.4.1 OceanStor 18500
3.4.2 OceanStor 18800
3.5 Robust Reliability
3.6 Storage Efficiency Technologies
3.6.1 SmartThin
3.6.2 SmartTier
3.6.3 SmartPartition
3.6.4 SmartQoS
3.6.5 SmartVirtualization
3.6.6 SmartMotion
3.6.7 VMware VAAI Integration
3.6.8 RAID 2.0+
3.7 Data Replication Technologies
3.7.1 HyperSnap
3.7.2 HyperClone
3.7.3 HyperCopy
3.7.4 HyperReplication/S
3.7.5 HyperReplication/A
3.8 Standard Compliance
3.9 Wide Compatibility
3.9.1 Compatibility with Operating Systems
3.9.2 Compatibility with Storage Products
Issue 1.1 (2014-05-31) / Huawei Proprietary and Confidential Copyright © Huawei Technologies Co., Ltd. / 1HUAWEI OceanStor 18500/18800 Enterprise Storage System V100R001 Technical Proposals Template / 3 Highlights and Advantages
1 Overview
This chapter consists of four sections. It analyzes and evaluates the customer's live network.You can merge or delete some of the sections based on actual requirements.
1.1 Project Background
Introduce the project background.
[Example]
As XX's (customer name) core services rapidly develop and data volumes soar, the existing storage system is unable to address increasing requirements on performance, reliability, scalability, efficiency, security, and manageability. Therefore, XX's current IT system requires restructuring and multi-controller high-end storage systems are needed.
1.2 Live Network Status
You are advised to add a network topology to the description of the live network structure.The topology indicates the quantity and types of existing devices.
[Example]
XX sets up more than 10 service systems such as Customer Relationship Management (CRM) and Enterprise Resource Planning (ERP), uses Windows, Linux, and UNIX, deploys Oracle and SQL databases, and employs VMware and Hyper-V virtual machines. Data of some service systems is stored on low-performance storage devices. The storage performance and efficiency of core service systems are far from satisfactory. This solution enables central data storage for core service systems. A HUAWEI OceanStor 18500/18800 enterprise storage system stores the data of XX's YYY service system and implements disaster recovery for core service systems.
1.3 Customer Requirements
Introduce the requirements on the solution design or product.
[Example]
XX has the following requirements on the new storage system:
1.Provides 100 TB of available capacity in the initial deployment and can be expanded to 500 TB.
2.Provides xxx IOPS.
3.Possesses the remote disaster recovery capability with arecovery point objective (RPO) of not longer than yy seconds, and recovers service data rapidly at disaster recovery sites.
1.4 Construction Scale and Objectives
This section describes requirements of the new network.
[Example]
According to XX's storage requirements, an analysis is made as follows:
1.An enterprise storage system is deployed in the data center to store the data of all core service systems.
2.The embedded remote replication capability of the enterprise storage system implements remote disaster recovery for the data of core service systems.
Issue 1.1 (2014-05-31) / Huawei Proprietary and Confidential Copyright © Huawei Technologies Co., Ltd. / 1HUAWEI OceanStor 18500/18800 Enterprise Storage System V100R001 Technical Proposals Template / 3 Highlights and Advantages
2 Design Basis and Principles
This chapter briefly explains the design basis and principles of the solution.You can merge some of the sections based on actual requirements.
2.1 Solution Design Basis
Describe the solution design basis.
[Example]
This document designs the system solution based on the XX Bidding Specifications.
2.2 Solution Design Principles
Describe the solution design principles.
[Example]
Based on XX's servicerequirements and application system characteristics, it is recommended that the storage system comply with the following rules:
Efficient
Supports thin provisioning that automatically allocates storage space for data writevolumes, preventing low space utilization that traditional disk allocation faces. Therefore, the storage purchase costs decrease.
Supports the intermixing of solid state drives (SSDs), SAS disks, and NLSAS disks. The high performance of SSDs and the large capacity of NLSAS disks are fully utilized. Hotspot data is automatically stored on SSDs and cold data is stored on NLSAS disks, striking a balance between high performance and cost conservation.
Controls the quality of service (QoS), enabling core services to obtain storage resources in time to meet their performance requirements.
Employs cache partitioning, providing sufficient cache resources to core services and preventing low-priority services from occupying excessive cache resources.
Supports heterogeneous storage virtualization that integrates existing storage systems. The function improves the data efficiency of existing systems, simplifies storage management, and maximizes the return on investment (ROI).
Secure and reliable
Supports multiple controllers to satisfy the requirements on excellent performance and robust reliability.
Adopts components that are capable of working continuously for 24/7 hours under heavy workloads.
Delivers sufficient redundancy and fault-tolerance capabilities.
Employs advanced technologies and mechanisms to ensure data reliability.
Has strong environments adaptability, undemanding on operating environments.
Provides effective measures to secure login and access, protecting the network against attacks.
Prevents data loss in the event of an unexpected power failure. After the power supply recovers, the storage system automatically starts and resumes connections.
Advanced
Complies with international standards.
Keeps in line with development trends of storage technologies and the IT industry. All used product models have a high-volume production.
Adopts cutting-edge technologies, making the storage system constantly up-to-date.
Presents an industry-leading processing capability and meets future upgrade requirements.
Open
Supports international standard network storage protocols and open application protocols.
Is interoperable with mainstream servers.
Supports mainstream operating systems, volume management software, and application programs.
Enables customized management and maintenance by being integrated with third-party management platforms.
Provides sufficient scalability that suits future expansion.
Supports disks made by mainstream vendors.
Maintainable
Provides a multilingual user interface that is easy to use.
Supports the management of permission, logs, and faults as well as automatic alarm reporting for faults.
Simplifies device installation and usage, requiring no maintenance expertise.
Supports on-demand and online system capacity expansion without interrupting services.
Supports hitless upgrade of system functions.
Implements web-based and centralized management.
Scalable
Supports online controller expansion.
Uses standardized components for flexible replacement and capacity expansion.
Complies with related rules and regulations.
Green
Meets the requirements on environmental protection and energy saving with its green characteristics such as low noise, low power consumption, and zero pollution.
Uses unleaded components.
Employs energy-saving technologies.
Obtains environmental management certification, uses recyclable packaging materials, and supports reutilization.
2.3 System Design Scheme
Explain the design solution in detail.You can merge or delete sections based on actual requirements.
[Example]
2.4 Topologies
This section introduces the network topology.
[Example omitted]
The following figure shows the storage network diagram.
The configuration characteristics are as follows:
Centralized storage of all critical service data
Support for Fibre Channel and IP networks
Intermixing of SSD, SAS, and SATA disks
Advanced disk spin-down technology
Snapshot and remote replication
Thin provisioning, dynamic storage tiering (DST), and cache partition
2.5 Software and Hardware Configurations
This section provides a list that describes the devices, interfaces, and software to be configured.
[Example omitted]
Location / Device Name / Specifications and Model / Quantity / RemarksXX central equipment room / 18500, 18800 / 18500 x-controller, xxx GB memory
18800 x-controller, xxx GB memory / x
18500/18800 disk bay / xxx
UltraPath
1 TB SATA disk
DST software, cache partition software, and QoS software
Issue 1.1 (2014-05-31) / Huawei Proprietary and Confidential Copyright © Huawei Technologies Co., Ltd. / 1
HUAWEI OceanStor 18500/18800 Enterprise Storage System V100R001 Technical Proposals Template / 3 Highlights and Advantages
3 Highlights and Advantages
This chapter describes features and advantages of the solution.
3.1 Advantages of HUAWEI Enterprise Storage System Design
Introduce the benefits the HUAWEI enterprise storage system design brings to the customer:
[Example]
1.Provides xx TB of available capacity in the initial deployment and can be expanded to xx TB.
2.Provides xx IOPS.
3.The disaster recovery is implementing using HUAWEI HyperReplication technology, which delivers an RPO of only three seconds.
3.2 Innovative Product Design
HUAWEI OceanStor 18500/18800enterprise storage system is an ideal storage platform for the new generation of data centers (requiring storage virtualization, private cloud, and hybrid cloud). It featuresrobust reliability, flexible scalability, and outstanding efficiency. It applies to storage of critical services in various sectors including finance, government, energy, manufacturing, transportation, education, and telecommunication.
3.3 High Performance
3.3.1 High-Speed Bus and I/O Channel
As the OceanStor 18500/18800adopts the new-generation PCIe 2.0 bus technology and the SAS 2.0 high-speed I/O channel technology, it provides up to 1024 GB/s of system matrix bandwidth, laying a solid foundation for the excellent performance of the storage system.
3.3.2 Multiple Controllers, Multi-Core CPUs, and Smart Matrix
The solution adopts multiple controllers with multi-core CPUs to deliver a powerful processing capability. At the same time, the Smart Matrix architecture enables each controller to access the caches of other controllers and select the best path to deliver I/O requests using the UltraPath software. In doing so, the system can reach its optimum performance.
The LUN balancing technology evenly divides LUN space to the controllers. Services on a LUN can utilize the cache and disk resources of multiple controllers, balancing service workload on each controller.
When multiple paths are available from a host to a LUN, UltraPath preferably selects the path between the LUN and its owning controller. If no optimum path is available, the system automatically finds the corresponding controller of the LUN service after I/O requests are delivered to the storage system. Then the Smart Matrix architecture transfers the I/O requests to the corresponding controller.
3.4 FlexibleScalability
3.4.1 OceanStor 18500
The OceanStor 18500 in standard configuration has one system bay and can be expanded to four system bays.
When the OceanStor 18500 is equipped with 3.5-inch disk enclosures, disk bays can be added. The OceanStor 18500can have two disk bays.
Disk enclosures can be expanded in a bay and disks can be expanded in a disk enclosure.
Storage pools, thick LUNs, and thin LUNs are all expandable online. Cache is embedded in each controller and can be replaced or expanded online.
3.4.2 OceanStor 18800
The OceanStor 18800 in standard configuration has one system bay and can be expanded to eight system bays.
When the OceanStor 18800 is equipped with 3.5-inch disk enclosures, disk bays can be added. The OceanStor 18800 can have two disk bays.
Disk enclosures can be expanded in a bay and disks can be expanded in a disk enclosure.
Storage pools, thick LUNs, and thin LUNs are all expandable online. Cache is embedded in each controller and can be replaced or expanded online.
3.5 Robust Reliability
This solution provides sufficient means to guarantee a solid system reliability.
Component redundancy / Hardware components have backup counterparts for redundancy, including controllers, power supplies, fans, batteries, switching matrixes, and links. Each engine adopts 1+1 power supply design to ensure reliability.Disk RAID protection / Multiple RAID levels are available for disks, including RAID 10, RAID 5, and RAID 6.
Disk reconstruction / If a disk becomes faulty, data on this disk is regenerated through reconstruction and then written onto other disks to recover the disk.
If a disk is replaced online, the storage pool balances service workload on all disks.
Prevention of mistaken disk removal / If a disk is reinserted within two minutes after being removed, the disk automatically restores services and continues providing services. Data written after the disk is removed is rewritten into the disk by the system according to system logs. If the disk is reinserted longer than two minutes after being removed, the system automatically rewrites data into the disk according to system logs, reducing reconstruction workload to speed up restoration.
Online FRU replacement / Front-end interface modules, back-end interface modules, management modules, controllers, fans, power modules, and batteries of engines are all hot-swappable.
Power failure protection / If power fails, batteries supply power and dirty data in cache is written onto coffer disks. Permanent power failure protection is implemented.
DIF verification / End-to-end data integrity field (DIF) verification is implemented throughout I/O paths from hosts, to disk arrays, and to disks.
Redundant component design minimizes the impact on services in the event of component faults. All redundant components can be replaced online, including controllers, power supplies, fans, batteries, interface modules, PCIe switches, and disks.
Location / FRU / Redundancy / Fault ImpactEngine / Controller: / 1+1 / Performance decrease by less than 30%.
Power module / 2+2 / No impact
Fan module / 5+1 / No impact
BBU module / 2+2 / No impact
Interface module / 1+1 / No impact
PCIe switch / 1+1 / No impact
2 U disk enclosure / Expansion board / 1+1 / No impact
Power module / 1+1 / No impact
Fan module / 1+1 / No impact
4 U disk enclosure / Expansion board / 1+1 / No impact
Power module / 2+2 / No impact
Fan module / 5+1 / No impact
If a disk becomes faulty, a hot spare disk can replace the disk and take over its services.
Disk reconstruction / Global hot spare space is available. If a disk fails, its data can be generated through reconstruction and then written onto other disks. After the faulty disk is replaced online, the system automatically adds the new disk to the storage pool and balances data on all disks.Prevention of mistaken disk removal / If a disk is reinserted within two minutes after being removed, the disk automatically restores services and continues providing services. Data written after the disk is removed is rewritten into the disk by the system according to system logs. If the disk is reinserted longer than two minutes after being removed, the system automatically rewrites data into the disk according to system logs, reducing reconstruction workload to speed up restoration.
The OceanStor 18500/18800 uses data coffer disks to protect data in the event of power failure.When AC power loses, the batteries supply the system with power and cache data is written onto coffer disks to prevent data loss.After the external power supply resumes, controllers write data from coffer disks back to the cache.
3.6 Storage Efficiency Technologies
3.6.1 SmartThin
The SmartThin (providing thin provisioning) integrated in the OceanStor 18500/18800 series uses the copy-on-write technology to allocate storage space at KB level (from 8 KB to larger). This technology promotes space utilization and makes storage space visible to service systems larger than the actual storage space. The SmartThin supports standard space reclamation ports for VxFS reclaim, VMware VAAI, and Windows 2012 TP platforms. Combined with host agent software, these ports implement space reclamation for these platforms, further boosting storage space utilization. Also, the SmartThin supports switchover of fat LUNs and thin LUNs.
3.6.2 SmartTier
With implementation of large-scale centralized data storage, customers are facing new challenges such as increased complexity in centralized data management, poor efficiency, and rising management costs. The SmartTier automatically calculates the I/O hotspot statistics of hotspot data at different periods of time to detect the values of data. Data of higher values is automatically stored on advanced storage media and data of low values are migrated to more economical storage media. The migration process is transparent to services.