solution > solution-detail

Inspur’s Big Data and High-Performance Computing Solution

[summary]

Abstract: The data platform developed by Inspur creates integration and a centralized management of data, which makes managing the system platform more centralized and simplified, thus reducing the burden on administrators. In addition, the automatic operation of servers, storage and networks also reduces a lot of potential human errors, which ensures the reliability of the system and improves efficiency.

Demands and Challenges

The Yellow Sea Ocean Observation Site and the East China Sea Oceanology Observation Buoy Site are the most important components of China’s offshore ocean observation and studies system. Both sites collect data every ten minutes through the ground-based data receiving station in Qingdao. The data includes weather data (temperature, humidity, barometric pressure, wind speed, etc.), hydrological data (waves, water temperature, salinity, ocean currents, etc.) and water quality data (temperature and salinity, turbidity, Chlorophyll, etc.), all of which are collected day after day and year after year. The ocean observation data, collected from a fixed point and in a long period, will be stored to support the development of marine science and regional economies. 

These extensive data feeds are stored in the data center of the Institute of Oceanology in the Chinese Academy of Sciences. The center, which is based on Inspur’s storage, provides a steady flow of data for oceanology.

The massive data reserves urgently await a high-performance storage system.

The Institute of Oceanology in the Chinese Academy of Sciences is an experimental unit for the national patent strategy of SIPO (State Intellectual Property Office of the P.R.C). Over more than 60 years, the institute has been focusing on the theories and key technologies related to high-quality, efficient, and sustainable development of “blue (marine) agriculture”. The institute also studies the dynamics of marine environments and ecosystems (as well as the dynamics of ocean circulation and shallow waters), geological evolution at continental margins, and the effects of resources and environments, etc. A lot of innovative and foundational work has been completed, and more than 900 results have been achieved. The institute has made great contributions to China’s economic development and national security, and to the development of oceanology.

Oceanography is a complicated field of science which involves fundamental research as well as applied research. It also requires efficient and reliable processing of massive data. Therefore, the establishment of a marine information system is of crucial importance. The Institute of Oceanology’s high-performance computing center has a high-performance computing cluster that can reach speeds of 10 trillion times per second, which facilitates the study of marine information.

As marine science goes deeper and the society’s demand for marine information increases, higher requirements are placed for the amount and effectiveness of marine information needed. To adapt to the new situation, the Institute of Oceanology decided to build a large marine information center. Based upon the center, a centralized system for information storage, processing and publishing will be established to provide marine information services for the institute, as well as for wider society. Under this circumstance, it is crucial that the process of establishing the system involves the creation of a highly reliable storage system that can not only store and protect massive data, but also ensures that concurrent access to data can be provided for multiple servers, multiple users and multiple application systems.

Inspur’s Big Data and High-Performance Computing Solution

5 data platforms for mass storage.

In accordance with the institute’s demands, Inspur developed a solution for establishing a data center that was accepted by the institute. The data center was called the management platform for the R&D of marine science data products, as well as data collection and distribution. It was based on Inspur’s storage AS8000, and it achieved the goals of centralized storage and the management of data blocks and file data. In the latter stage, data from 5 platforms (namely offshore observation networks, open sea voyages, subsurface buoy observation, and a large amount of historical research and analyses) will be gradually transmitted and stored in the data center so as to achieve centralized storage and management. Meanwhile, the data center will also manage and operate the ground-based stations used by the offshore observation networks, receive and process data from the offshore observation networks, and release data products.

The Inspur AS8000’s high performance enables it to reach a speed of 500,000 reads and writes per second, enough to handle even the institute’s large amount of concurrency. In terms of scalability, the AS8000’s storage resources can be accessed seamlessly by new storage devices. Moreover, as capacity increases, the performance of data storage will also increase in a linear way to meet the demand for mass storage.

Safe and reliable storage provides the data center with a dose of reassurance.

The data security of the institute’s data center is an important issue. Since quantitative data collection concerns the evaluation of the whole institute, the accuracy, comprehensiveness and security of the data are of crucial importance.

For instance, the comprehensive observation data collected by the research vessels of the Chinese Academy of Sciences (from the coastal waters of China to the Western Pacific Ocean) concerns the impact the anomalies associated with the Western Pacific Warm Pool have on China’s weather. The data will also show the influences that nearby oceans have on China’s offshore environment, and will also inform future research on marine biodiversity, the deep sea environment and life detection. Once the data is lost, the losses will be impossible to estimate.

As the pioneering Inspur storage product in the cloud era, the Inspur AS8000 adopts cloud storage technology, which provides the institute’s data center with a dose of reassurance. The AS8000 adopts a fully redundant architecture and is immune from single points of failure. In terms of security, it is impeccable.

The AS8000’s performance is also excellent in terms of data backup and protection. Local backup of the institute’s data can be completed whilst using the AS8000’s functions, such as snapshot and volume replication. However, when natural disasters such as fires and earthquakes occur, data stores that rely on local backup and the system will still be seriously damaged. As it is different from general storage, the Inspur AS8000 can provide advanced data protection and achieve data disaster tolerance on the basis of storage. Data copies will be stored in different places so that when unstoppable disasters happen, data can be recovered quickly. This will ensure that the institute’s precious data will not be lost and guarantee that the institute can continue to operate smoothly. 

In terms of business continuity, the institute’s large amount of original data and systems can be easily transferred to the new storage system without shutting down the applications, thus achieving truly seamless data migration.

Inspur’s Solution and its Value

Efficiency is improved and the project lifecycles are shortened.

 “In the previous information platform, the management and storage of data was loose, which made it difficult for research departments to access data. Now all data will be stored together in the data center. The data can be retrieved with one click of the mouse, which is very convenient for our researchers. Their efficiency has been improved a lot,” said the director of the institute’s data center.

Every day, the institute collects and stores a large amount of data, and the loose data storage methods of the past only hindered the efficiency of scientific research. The data platform developed for the institute by Inspur allows for the integration and centralized management of data, which makes managing the system platform more centralized and simplified, thus reducing the burden placed on administrators. In addition, the automatic operation of servers, storage and networks also reduces a lot of potential human errors, which ensures the reliability of the system and the improvement of efficiency. The life cycles of the institute’s scientific investigation projects have been reportedly shortened by nearly 30% as a result of the speed of storing and accessing data being significantly increased. As a result, the institute can devote more time and energy to marine science and will soon become an institution with international influence. 

中科汇联承办,easysite内容管理系统,portal门户,舆情监测,搜索引擎,政府门户,信息公开,电子政务