NetApp与OpenStack部署及运维全面指南

需积分: 9 65 浏览量更新于2024-07-20 收藏 2.76MB PDF 举报

本指南深入探讨了OpenStack在与NetApp技术协同部署和运营中的关键要素，特别针对OpenStack Liberty（2015年冬季）版本。OpenStack是一个模块化的云计算服务集合，它允许用户构建和管理自己的云基础设施。NetApp作为主要贡献者，提供了针对OpenStack Cinder块存储服务的支持。第1章执行摘要概述了作者和贡献者，强调了这份文档的重要性，旨在确保OpenStack在结合NetApp存储解决方案时的顺利实施。NetApp的存储系统包括Data ONTAP、FASS（File Access Service）系统以及E-Series和EF-Series等高效存储解决方案，这些都在指南中得到了详细介绍。第2章对OpenStack进行了概述，介绍了OpenStack基金会的角色，以及其不同版本的迭代发展，如从Icehouse到Liberty。NetApp在OpenStack生态系统中的贡献主要体现在提供稳定且兼容的存储服务，特别是Cinder块存储服务。Cinder是OpenStack的核心组件，负责为虚拟机提供持久化块存储，对于任何云环境来说，存储性能和可靠性至关重要。第4章详述了OpenStack Block Storage Service（Cinder），着重于其功能和部署。Cinder支持多种后端存储选项，包括NetApp的存储解决方案，使得管理员可以根据需求选择最合适的存储配置。它提供了灵活的存储管理和扩展能力，有助于提升云平台的整体性能和可用性。在后续章节中，指南还将涉及OpenStack的网络部署、安全设置、性能调优以及故障排除等实践操作，帮助读者规划、部署和维护一个融合NetApp技术的OpenStack云环境。这是一份全面的指南，旨在确保OpenStack项目的成功实施，并最大限度地利用NetApp的存储技术优势。

computing. Clustered Data ONTAP leverages core NetApp software such as

WAFL® (Write Anywhere File Layout), RAID-DP®, and NetApp Snapshot. RAID-

DP, a high-performance implementation of RAID 6, protects against double-

disk failures, and transparent node failover automatically bypasses any failed

components with no interruption in data availability. In addition to having

no single point of failure, clustered Data ONTAP supports the expansion or

reconfiguration of the storage infrastructure while online, enabling applications to

run uninterrupted as more storage capacity, processing power, and/or throughput

is added.

Enable continuous operations. Clustered Data ONTAP is configured for

continuous operation with the use of high-performance and modular NetApp

storage components. Each system consists of one or more FAS building blocks

where each building block is a high-availability pair of controllers (storage nodes).

Multiple controller pairs form a single, integrated cluster. Clustered Data ONTAP

uses Ethernet technology — Gigabit(Gb) and 10Gb — for server connections

and for interconnecting FAS controllers. Servers can also be connected through

InfiniBand through the use of a gateway device. Each controller can support

any mix of high-performance SAS and cost-effective SATA disk drives. Data

can move nondisruptively between nodes or between different tiers of disk

as performance requirements change. This capability makes sure that data

center and IT administrators can maximize performance where needed while

simultaneously improving capacity utilization.

Data ONTAP operating in 7-Mode

While clustered Data ONTAP is the preferred operating mode for nearly all

new Data ONTAP installations, and is NetApp’s focal point for future delivery

of additional enhancement and innovation, there are a significant number of 7-

Mode based systems in existence and a variety of valid operational considerations

that require ongoing use. While NetApp has provided Cinder driver enablement

for 7-Mode systems, NetApp recommends that clustered Data ONTAP be used

whenever possible.

3.2. Description of FAS Systems

The FAS family is a unified storage systems family with integrated data protection,

robust data management, and built-in storage efficiency for virtualized, shared

infrastructure and business applications. FAS storage is excellent for data

management efficiency, virtualized infrastructure, shared workloads, and IT as a

service delivery. FAS systems are often referred to as “filers.” FAS systems can be

configured with a variety of media types including hybrid (Flash, SSD, SAS, and

SATA) as well as all flash configurations.

Detailed information on the individual options within the FAS family is available

at: http://www.netapp.com/us/products/storage-systems/.

Note

NetApp’s OpenStack Block Storage drivers interact directly with Data

ONTAP and are abstract to the specific FAS platform. Any given

hardware platform supported by a version of Data ONTAP that in turn

is supported by a particular OpenStack Block Storage driver. Refer to

Section 4.6.9, “Data ONTAP Configuration” for more information on

prerequisites for Data ONTAP.

3.3. Description of E-Series and EF-Series

Systems

The family of E-series and EF-series systems provides performance-efficient, high-

density block storage aimed primarily at application-driven workloads.

NetApp E-Series systems are designed to provide:

• Simple, low-touch administration

• Flexible, reliable SAN storage

• Extreme capacity and density

• High performance GB/s or IOPS

• Performance, power, and space efficiency

• Consistent, low latency

• Consistently high IOPS and throughput

• Enterprise-class capabilities (reliability, availability, manageability)

The NetApp EF-Series is an all-flash storage array that brings together extreme

performance and enterprise-grade reliability to create a system optimized for

latency-sensitive workloads. Building on a heritage of performance leadership, its

core architecture has been proven in the world’s most demanding and complex

computing environments. Its field-proven design is the culmination of 20-years

of industry knowledge focused on designing enterprise-class storage. Leveraging

experience from nearly 1,000,000 systems shipped, the fully-redundant EF-Series

all-flash array is architected to provide the highest levels of reliability, availability

and data protection.

The E and EF-Series run on the enterprise-proven SANtricity® software platform.

SANtricity is designed to combine sustainable, low-latency performance with

enterprise-class availability and protection. Its streamlined firmware is well-suited

for the extreme demands of latency-sensitive workloads. And SANtricity helps

keep I/O flowing with automated path failover, online administration, advanced

data protection, proactive monitoring and repair, non-disruptive upgrades, and

extensive diagnostic capabilities.

E-Series’ Dynamic Disk Pools (DDP) provide an alternative to standard RAID

groups. They simplify protection by removing the complexity of configuring RAID

groups and allocating hot spares. Utilization is improved by dynamically spreading

data, parity, and spare capacity across all drives in a pool, reducing performance

bottlenecks due to hot-spots. Additionally, should a drive failure occur, DDP

enables return to optimal state significantly faster than RAID6, while reducing the

performance impact during the reconstruction of a failed drive. DDP also offers

greater protection from multiple drive failures by prioritizing the reconstruction of

the most critical segments.

The actual blocks provisioned in support of a Cinder volume reside on a single

Cinder backend. Starting in the Havana release, a Cinder volume can be migrated

from one storage backend to another within a deployment of the Cinder service;

refer to Section 4.7.3, “Cinder Command Line Interface (CLI)” for an example of

volume migration.

The cinder manage command allows importing existing storage objects that are

not managed by Cinder into new Cinder volumes. The operation will attempt

to locate an object within a specified Cinder backend and create the necessary

metadata within the Cinder database to allow it to be managed like any other

Cinder volume. The operation will also rename the volume to a name appropriate

to the particular Cinder driver in use. The imported storage object could be a file,

LUN, or a volume depending on the protocol (iSCSI/FC/NFS) and driver (Data

ONTAP operating in 7-mode, clustered Data ONTAP, or E-Series) in use. This

feature is useful in migration scenarios where virtual machines or other data need

to be managed by Cinder; refer to the section called “Cinder manage usage” for

an example of the cinder manage command.

The cinder unmanage command allows Cinder to cease management of a

particular Cinder volume. All data stored in the Cinder database related to the

volume is removed, but the volume's backing file, LUN, or appropriate storage

object is not deleted. This allows the volume to be transferred to another

environment for other use cases; refer to the section called “Cinder unmanage

usage” for an example of the cinder unmanage command.

Snapshot

A Cinder snapshot is a point-in-time, read-only copy of a Cinder volume.

Snapshots can be created from an existing Cinder volume that is operational and

either attached to an instance or in a detached state. A Cinder snapshot can serve

as the content source for a new Cinder volume when the Cinder volume is created

with the create from snapshot option specified.

Backend

A Cinder backend is the configuration object that represents a single provider of

block storage upon which provisioning requests may be fulfilled. A Cinder backend

communicates with the storage system through a Cinder driver. Cinder supports

multiple backends to be simultaneously configured and managed (even with the

same Cinder driver) as of the Grizzly release.

Note

A single Cinder backend may be defined in the [DEFAULT]

stanza of cinder.conf; however, NetApp recommends that the

enabled_backends configuration option be set to a comma-

separated list of backend names, and each backend name have

its own configuration stanza with the same name as listed in the

enabled_backends option. Refer to Section 4.6.1, “Cinder” for an

example of the use of this option.

剩余150页未读，继续阅读

woricjiajia

粉丝: 0
资源: 11

NetApp与OpenStack部署及运维全面指南

openstack-utils工具包

centos-release-openstack-pike-1-0.el7.x86_64.rpm

jenkins-openstack-deployment-plugin:一个用于 openstack 部署的 jenkins 插件

ansible-deployment-openstack-demo

Docker-containers-deployment-with-OpenStack-Heat:如何通过简单的步骤使用 OpenStack Heat Docker 化您的应用程序

hadoop-openstack-swifta:hadoop-openstack-swifta

kitchen-openstack-testing:厨房-openstack-testing coobook

Openstack-实验室：Openstack-实验室

OpenStack-EC2-Driver:OpenStack-EC2-驱动程序

最新资源