当前位置:网站首页>CNOOC, desktop cloud & network disk storage system application case
CNOOC, desktop cloud & network disk storage system application case
2022-07-29 06:06:00 【Taocloud Avenue】

CNOOC is 1982 year 2 month 15 A super large state-owned enterprise established with the approval of the State Council in Japan , It is the largest offshore oil and gas production operator in China . The company's main business segments include oil and gas exploration and development 、 Professional technical services 、 Refining and sales 、 Natural gas and power generation 、 Financial services, etc .
With the expansion of the company , Conventional PC And the office mode of decentralized storage of data is difficult to adapt to high efficiency 、 Simple and convenient 、 Energy saving 、 Safe office needs , Therefore, the company has passed continuous verification , Decided to build a cloud desktop and network disk system , Centralize resources and data , Comprehensively improve office efficiency and reduce management complexity . How to store such huge data is to build a reliable 、 One of the key points of an efficient desktop cloud and network disk system .
“ Facing the soul of the customer's father, three questions , Look at the small TAO How to respond tactfully ”
CNOOC : In the future, with the expansion of business and personnel , The amount of data will continue to grow , How to meet the storage needs of existing massive data , It can also meet the growing capacity demand of data ?
The road is cloudy :XDFS Distributed unified storage system provides online capacity expansion , It can meet the growing capacity demand of desktop cloud and network disk .
Efficient and convenient expansion , It can simplify the system personnel's operation of expanding the storage resource pool , Improve work efficiency , So as to reduce operational risk .
CNOOC : Most of our cloud desktops are for office workers , The startup time is relatively fixed , Therefore, a large number of boot operations in a period of time will cause high performance requirements for storage in a certain period of time , There will also be some data that will be accessed frequently , And it is possible that in a certain period of time, most users access the storage at the same time and read and write concurrently . In this case , Will the …
The road is cloudy : You can rest assured that ,XDFS The storage resource pool can be deployed in a mixed manner SSD and HDD Hard disk , Also covering documents 、 High bandwidth transmission, etc , Fully ensure that the storage resource pool provides sufficient performance guarantee , To meet diversity , Different types of data access support .
CNOOC : Our disk is very important data , If suddenly damaged, the damage will have a great impact on the office , Do you have any countermeasures ?
The road is cloudy :XDFS Provide mature 、 Stable storage resource pool , In terms of software and hardware matching 、 Equipment operation reliability 、 The system fault recovery ability and other aspects will be strictly tested and verified , To ensure that the storage resource pool provides safe and reliable data storage support for cloud desktop and network disk systems , In case of hardware failure , Minimize the risk of business interruption and data loss .
“ Click on the tongue , The product is hard, there is no way , So small TAO Overnight solutions “
Network topology

Program Overview
According to users' needs and combined with users' actual application scenarios , In this project XDFS Distributed storage system , Build online expansion 、 Highly reliable 、 High performance storage resource pool , Provide massive services for cloud desktop and network disk systems 、 Efficient 、 Scalable storage resource pool service .

️SSD and HDD Hybrid configuration , Realize tiered storage , Desktop system disk 、 Data disk 、 The network disk data is stored in different specific spaces according to different data characteristics ; The system disk of cloud desktop adopts SSD, To improve desktop startup speed , Avoid boot storm , At the same time SATA Disk as the data disk of the desktop , thus Improve the overall cost performance ;
️ Data for network disk , adopt XDFS Automatic migration of cold data and temperature data , Ensure that high-frequency access data is always stored in high-performance SSD On disk , Data that has not been accessed for a long time is stored in SATA In hard disk , Meet the performance requirements of users' concurrent reading and writing ;
️ adopt The double copy party fully guarantees the security of the system , Even if a storage node fails, it will not cause data loss and system damage ; adopt The snapshot backs up the system disk and template , Fully ensure data security .
Program advantages
️ Provide a safe and reliable self-developed storage platform , It also provides excellent performance and simple configuration management , It greatly simplifies the operation and maintenance of users ;
️ Double copy fully guarantees data security , The distributed architecture can realize horizontal expansion and fully meet the growing storage capacity requirements of users ;
️ Data hierarchical storage , Effectively avoid the impact of the desktop system boot storm , And the performance bottleneck of frequent access to network disk data , Ensure that users have a good experience of cloud desktop and network disk system .
“ The customer's father's feedback is coming !”
Customer feedback
CNOOC :XDFS The distributed unified storage system meets the business needs of cloud desktop and network disk systems , And performance 、 Configuration management and other aspects have achieved better optimization effect , It simplifies the operation and maintenance of business personnel , It provides a safe and reliable data protection mechanism for CNOOC .
边栏推荐
- 【语义分割】Fully Attentional Network for Semantic Segmentation
- [semantic segmentation] full attention network for semantic segmentation
- 通过简单的脚本在Linux环境实现Mysql数据库的定时备份(Mysqldump命令备份)
- 四、One-hot和损失函数的应用
- Reporting Services- Web Service
- Nifi changed UTC time to CST time
- Technology that deeply understands the principle of MMAP and makes big manufacturers love it
- tensorflow中tf.get_variable()函数详解
- [target detection] KL loss: bounding box progression with uncertainty for accurate object detection
- Android studio login registration - source code (connect to MySQL database)
猜你喜欢
![[convolution kernel design] scaling up your kernels to 31x31: revising large kernel design in CNN](/img/71/f3fdf677cd5fddefffd4715e747297.png)
[convolution kernel design] scaling up your kernels to 31x31: revising large kernel design in CNN

【Transformer】ATS: Adaptive Token Sampling For Efficient Vision Transformers

迁移学习—Geodesic Flow Kernel for Unsupervised Domain Adaptation

GAN:生成对抗网络 Generative Adversarial Networks

【pycharm】pycharm远程连接服务器

fastText学习——文本分类

Yum local source production

【Transformer】AdaViT: Adaptive Vision Transformers for Efficient Image Recognition

研究生新生培训第一周:深度学习和pytorch基础

Exploration of flutter drawing skills: draw arrows together (skill development)
随机推荐
Use of xtrabackup
mysql 的show profiles 使用。
一、常见损失函数的用法
[semantic segmentation] full attention network for semantic segmentation
有价值的博客、面经收集(持续更新)
tensorflow中tf.get_variable()函数详解
Rsync+inotyfy realize real-time synchronization of single data monitoring
【Transformer】AdaViT: Adaptive Tokens for Efficient Vision Transformer
clion+opencv+aruco+cmake配置
四、One-hot和损失函数的应用
【Transformer】SegFormer:Simple and Efficient Design for Semantic Segmentation with Transformers
迁移学习—— Transfer Feature Learning with Joint Distribution Adaptation
迁移学习——Transitive Transfer Learning
IDEA中设置自动build-改动代码,不用重启工程,刷新页面即可
yum本地源制作
Lock lock of concurrent programming learning notes and its implementation basic usage of reentrantlock, reentrantreadwritelock and stampedlock
迁移学习笔记——Adapting Component Analysis
[DL] build convolutional neural network for regression prediction (detailed tutorial of data + code)
Spring, summer, autumn and winter with Miss Zhang (2)
二、OCR训练时,将txt文件和图片数据转为lmdb文件格式