No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

OceanStor 18500 V3 and 18800 V3 Mission Critical Storage System V300R003 Basic Storage Service Guide for File 16

"This document describes the basic storage services and explains how to configure and managebasic storage services."
Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Basic Storage Principles

Basic Storage Principles

Storage systems provide storage space for application servers. The 18500 V3/18800 V3 storage systems use the block virtualization technology to support dynamic allocation and expansion of storage resources in storage pools. This shortens the response time for data reads/writes in the storage pools and the reconstruction time after a disk fails.

Basic Concepts

Get yourself started with the following basic concepts.

  • Disk domain: consisting of the same type or different types of disks. Disk domains are isolated from each other. Therefore, services carried by different disk domains do not affect each other in terms of performance and faults.
  • Storage pool: container of storage resources, which is created under a disk domain. The storage resources used by application servers are all from storage pools. Based on the storage media, a storage pool can have three storage tiers, including the high performance tier, performance tier, and capacity tier.
  • Storage tier: a set of storage media providing the same performance in a storage pool. Storage tiers are used to manage storage media with different performance and provide appropriate storage space for applications having different performance requirements.
  • CHUNK: CK for short, consecutive physical spaces of a fixed size on a disk.
  • CHUNK Group: CKG for short, a logical set of CHUNKs on different disks. A CHUNK Group has the properties of a RAID group.
  • Block virtualization: a new type of redundant array of independent disks (RAID) technology. Block virtualization divides disks into multiple CHUNKs of a fixed size and organizes them into multiple CHUNK groups. When a disk fails, all the disks on which the other CHUNKs in the same CHUNK group as the CHUNKs on the failed disk are located participate in the reconstruction. This significantly increases the disks involved in the reconstruction, eliminating the performance bottleneck in the reconstruction of traditional RAID groups and improving the data reconstruction speed. In addition, block virtualization distributes data to all the disks in a storage system and leverage the I/O processing capability of the storage system.
  • Extent: An extent is a logical storage space with a fixed size divided from a CKG. The size ranges from 512 KB to 64 MB. The default size is 4 MB. Extent is the smallest unit (granularity) for data migration and hotspot data statistics collection. It is also the smallest unit for space application and release in a storage pool.
  • Grain: In file system mode, extents are further divided into grains. The size of grains ranges from 4 KB to 64 KB. The default size of a grain is 64 KB. Grains are basic units that constitute a file system.
  • Hot spare space: the space used for faulty block data reconstruction in block virtualization. When a CHUNK is faulty, the system lets a CHUNK of the hot spare space take over and instructs the other CHUNKs in the CHUNK group to perform data reconstruction using the hot spare space. This ensures data integrity and read/write performance.
  • Reconstruction: A process of restoring the data saved on a faulty disk to hot spare CKs and replacing the CKs on the faulty disk with the hot spare CKs. During data reconstruction, valid data and parity data must be read and processed to restore the data saved on a faulty disk to hot spare space, thereby ensuring data security and reliability. Traditional reconstruction technologies allow only all disks in the same RAID group as the faulty disk to participate in reconstruction. RAID 2.0+ technology enables all disks of the same type and in the same disk domain as the faulty disk to participate in reconstruction, boosting data reconstruction speed and shortening data recovery duration.

    Data on other disks is read for reconstruction. To prevent reconstruction failures, service interruption, and data loss, do not remove other disks of the same type as the faulty disk in the disk domain where the faulty disk resides.

  • Quota: An administrator can set the number of files and space size for different directories. The quota management feature developed by Huawei is called SmartQuota.
  • Quota tree: A quota tree is the root directory of a file system. File quantity and storage space under a quota tree can be managed.
  • Thin file system: A thin file system is logical space accessible to a host, which is configured with an initial capacity when being created and dynamically allocated required storage resources when its available capacity is insufficient.
  • Thick file system: A thick file system is logical space accessible to a host, which is allocated a fixed capacity of storage resources according to the capacity specified when being created using the thin provisioning technology.

Storage Pool Structure

Figure 1-1 shows the structure of a storage pool.

Figure 1-1  Structure of a storage pool

A storage pool consists of three storage tiers at most. Each storage tier is constructed by the same type of storage media.

  • Tier 0 is the high performance tier, which is composed of solid-state drives (SSDs). Tier 0 provides the highest performance at a high cost. It is used to store frequently accessed data.
  • Tier 1 is the performance tier, which is composed of serial attached SCSI (SAS) disks. Tier 1 provides high performance at a moderate cost. It is used to store less frequently accessed data.
  • Tier 2 is the capacity tier, which is composed of Near Line SAS (NL-SAS) disks. Tier 2 provides moderate performance and a large capacity per disk at a low cost. It is used to store a large amount of data and seldom-accessed data.

Block Virtualization Process

Figure 1-2 shows the block virtualization process.

Figure 1-2  Block virtualization process

  1. The storage system divides the storage media in a disk domain into fixed-size CKs.
  2. CKs in each storage tier are grouped into CKGs and hot spare space based on the RAID policy and hot spare policy specified on DeviceManager.
  3. The storage system divides CKGs into extents.
  4. The storage system divides extents into grains based on the file system block size configured on DeviceManager.

Quota Configuration

You can configure different quantities and space sizes for different directories for efficient storage resource utilization. Figure 1-3 shows how to configure a quota.

Figure 1-3  Configuring quota

Working Principle of SmartQuota: A storage system employs hard quotas (including hard quotas of capacity and files) to restrict the maximum number of resources available to each user. The process is as follows: In each write I/O operation, check whether the accumulated quota (Quotas of the used capacity and file quantity + Quotas of the increased capacity and file quantity in this operation) exceeds the preset hard quota. If the accumulated quota does not exceed the preset hard quota, the follow-up operations can be performed. Otherwise, the write I/O operation fails. After the write I/O operation is allowed, add the incremental capacity and file quantity to the previously used capacity and file quantity. Then, update the quota (Latest capacity + Latest file quantity) and enable the quota and I/O data to be written into the file system. The I/O operation and quota update succeed or fail at the same time, ensuring that the used capacity is correct in each I/O check.

ensuring that the used capacity is correct in each I/O check. If a directory quota, user quota, and group quota are concurrently configured in a shared directory in which you are performing operations, each write I/O operation will be restricted by the three quotas. You must check each type of quota. If the hard quota of one type of quota does not pass the check, the I/O will be rejected.

User Permission Control

You can assign different user permissions for the same directory, so that the users can only access the directory within their specified permissions. Figure 1-4 shows user permission control.

Figure 1-4  User permission control

Users with the full control permission can not only read and write directories but also have permissions to modify directories and obtain all permissions of directories. Users with the forbidden permission can view shared directories but cannot perform operations in any directory.

Space Occupation upon Data Writes

Users use the redirect-on-write (ROW) technology to write data into files in the following two scenarios:
  • If the size of the data to be written is an integral multiple of the file system block size, the data will be written to a new location and the space of old data will be released (if the old data contains the created snapshots, the space of old data will not be released).
  • If the size of the data to be written is not an integral multiple of the file system block size, the old data will be read and then written to a new location with the new data, and the space of old data will be released (if the old data contains the created snapshots, the space of old data will not be released).
The ROW technology is used to save data to a new location, enabling quick data writes. Figure 1-5 shows how to modify data.
Figure 1-5  Data writes process

Write mode shifts from write back to write through

Generally, the write mode of file system in a storage system is write back by default. However, the write mode will become write through in the event of a fault.
Table 1-1  Scenarios where the write mode of file system shifts from write back to write through and recommended actions

Symptom

Scenario

Impact and Recommended Action

The temperature of a controller exceeds the upper limit.

  • If an overtemperature alarm is generated, the write mode of file system shifts from write back to write through.
  • If the ambient temperature becomes unacceptable or a component in a storage system malfunctions, the write mode of file system shifts from write back to write through.

Impact

The write mode of service objects on the entire engine becomes write through.

Recommended action

Check the external refrigerating system, fan modules, and air channels to locate the overtemperature causes and rectify faults.

BBUs on an engine malfunction.

  • Dual-controller storage device: If two BBUs malfunction and an alarm is generated, the write mode of file system shifts from write back to write through.
  • Four-controller storage device: If two or more BBUs malfunction and an alarm is generated, the write mode of file system shifts from write back to write through.

Impact

The write mode of service objects on the entire engine becomes write through.

Recommended action

  • Check whether BBUs are properly inserted.
  • Check whether the BBUs break down. If the BBUs break down, replace them with the spare parts.
  • Check whether the power of the BBUs is insufficient. If the power of the BBUs is insufficient, wait until the BBUs are fully charged.

The coffer disks of an engine malfunction.

  • Dual-controller storage device: If both coffer disks break down, the write mode of file system shifts from write back to write through.
  • Four-controller storage device: If all coffer disks of controllers A and B or controllers C and D break down (the controllers in the first row are controllers A and B and the controllers in the second row are controllers C and D), the write mode of file system shifts from write back to write through.

Impact

The write mode of service objects on the entire engine becomes write through.

Recommended action

Check whether the coffer disks are faulty. If the coffer disks are faulty, replace them with spare parts.

A controller malfunctions.

By default, the write mode of file system remains write back within a certain period (192 hours) in which a single controller malfunctions. If the fault is not rectified within this period, the write mode of file system shifts from write back to write through.

Impact

The write mode of service objects on the entire engine becomes write through if the fault persists after 192 hours.

Recommended action

  • Replace the faulty controller at the off-peak point in time during the write back protection period.
  • If the spare part is unavailable during the delay protection period of write through, add a proper period to the delay after assessing risks to prevent write through from adversely affecting service performance.

The remaining capacity of a storage pool is smaller than the reserved capacity.

An alarm is generated indicating that the capacity usage of a storage pool exceeds the threshold and reminding you of expanding the capacity.

Impact

The write mode of thin file system and thick file system with value-added features shifts from write back to write through.

Recommended action

Expand the capacity of the storage pool.

The remaining capacity of a file system is smaller than the threshold.

If the remaining capacity of a file system is smaller than the threshold, an alarm will be generated. If the remaining space is about to be exhausted, the file system's write mode changes to write through by default.

NOTE:
When the following case occurs, the data write mode of file system shifts to write through due to insufficient file system capacity, however, it may not generate the alarm of the remaining capacity of file system is smaller than the threshold:

The used capacity of file system + the capacity that the grain occupies + the capacity of dirty data in cache > the total capacity of file system

the capacity that the grain occupies including space debris, space prefetch, delayed release space.

Impact

After the write mode of the file system changes to write through, the performance of the file system deteriorates.

Recommended action

Expand the capacity of the file system.

Translation
Download
Updated: 2019-08-14

Document ID: EDOC1000084081

Views: 35970

Downloads: 83

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next