China has one of the world’s lowest crime rates. A study by ECNU (East China Normal University) reveals that: Crime rate in China, namely the number of criminal events happened in every 10,000 people in a year, accounts for only 360 per 10,000: only 3% of Switzerland, 3.25% of New Zealand, 3.5% of Finland, 3.8% of Denmark and 5% of America. For example, crime rate in Shanghai is lower than Munich, the alleged safest city in Germany. Other cities such as Beijing and Guangzhou also exhibit a higher safety index than other countries. For that, we owe it to our public security system that has done substantial work to maintain social security and stability, such as:

A dense web of monitors has been knitted to watch out for our citizens and prevent criminal activities in advance. In 2010 alone, 10 million new monitors have been put in use.

Quick police response. If someone dare molest another in a bus, simply dial 110, the police will show up at next bus stop to apprehend the suspect. In Beijing, it has specified on how fast local police should respond to an emergency report. Police should arrive in 5 minutes after they receive an emergency report in the eight urban areas and central areas in rural areas and towns, 10 minutes in urban and rural copula and outskirt areas. For remote rural areas and mountain areas, though time limit has not been given, the police should arrive as soon as possible.

What a robust big data solution can offer for our public security is unimaginable. With data growing rapidly to TB or even PB level, it has brought new challenges and higher requirements upon system application and management than any other traditional backup assignment. A public security system contains a wide array of data types: videos, criminal information and so on. Such data is of great value to maintaining our social security and must be protected 24*7.

The public security IT system has built a mature data center in 2006. Its internal structure is super complex and contains many vertical applications. As new technologies like mobile internet and IoT(Internet of Things) have diversified data source, data relations become even more complicated. Also, as a special proprietary network designed for safeguarding social security, it needs to frequently interact with other data systems to retrieve information and coordinate operations. For example, if a driver from Guangdong province broke traffic rules in Beijing, the authority has to go through a lot of procedures to extract data.

For another example, during a financial fraud investigation, if you input a name and search citywide, the list may be short; but if you widen you search to nationwide, the result may be overwhelming. More tech issues will surface with time going on.

New changes have brought along new challenges, which a traditional technological approach cannot tackle. Not long ago, a tech expert working for a public security IT system shared his work experience. The public security system can be divided into:

  Social security related business system,

  Criminal related business system,

  Economic investigation related business system,

  Transport related business system,

  Command related business system,

  Video business system.

In terms of data storage, data will be stored at where it is produced. To be specific, data produced from a municipal system will stay within this city, data produced from a provincial system will be stored by this province, data from a ministerial system will be stored by the nation.

Data reuse can be categorized into three levels:

Internal reuse: general queries, data replication, database account, and content search

Cross-region reuse: service request, whole-network search, check and verify

External reuse: cross-departmental information share, government website, people’s livelihood

“From my point of view, there is no critical difference between 1 million and 100 million data. Public security data is experiencing an exponential growth.”

Traditional technologies have hit its bottleneck as it can no longer realize computing models of current software. To bridge between public security network with government websites and other systems, it involves not only the pile-up of pics, videos and other forms of traditional business data, but also application of modern technologies such as distributed storage, graphical display, in-memory database and so on. “Upon such urgent demands, the public security system should join hands with more software developers and IT providers to resolve this issue together” as is addressed by the expert.

We had handled a project in public security system not long ago. The Bureau of Public Security of Province A was looking for a system featuring local backup with off-site synchronization to storage media.

The Bureau of Public Security in Province A has two major database systems: the basic platform database integrates information resources from all police divisions in the province, consisting of two Linux physical servers connected by Oracle RAC; the macro information database collects business information from all public security systems, also consisting of two Linux physical servers connected by Oracle RAC. Note that any malfunction to these two databases should be reported to Ministry of Public Security, so an all-round protection should be deployed to prevent any accidents.

However, there may be such problems as:

I) Possible damage to key business system server

Its key business systems (macro information database, basic platform database) all run on a X86 Linux server. Though Oracle RAC is deployed, considering the ratio of a X86-architecture server going south increases dramatically after a few years of use, there may be circumstances that all x86 servers malfunction at the same time and result in data loss.  

II) It used to adopt a single and old-fashioned backup approach that is neither efficient nor user-friendly

III) It had no off-site disaster recovery. Once an emergency or disaster happens to local data center, partial or all business data may be lost.

IV) It had no automatic disaster recovery drills. No disaster recovery drills means there’s no way to verify the availability of backup data.

The data center has purchased high-end domestic backup appliances, storage media and DBackup. With independently developed hardware-software, both key systems and Oracle RAC databases can be protected real-time.

Automated full/incremental backup with compression and deduplication features to improve efficiency

Deploy appliances in the disaster recovery center to realize asynchronous synchronization with the other data center, backup data in the local backup media will be transferred to the other local center for off-site backup and avoid potential data loss.

‘Disaster recovery-Disaster recovery drill’ feature periodically restores important backup data to the server to verify its availability.

A user-friendly interface and a wizard to allow users to deploy within 6 steps.

Achieved automated backup and disaster recovery of databases

Achieved efficient real-time backup and recovery

Achieved off-site disaster recovery

Achieved disaster recovery drills to verify data availability

Uses domestic hardware/software to ensure independence and controllability

An efficient and robust big data solution for public security should be able to integrate all kinds of data sources to assist the bureau in terms of information detection, analysis and decision-making, command and management, investigation and society service. Scutech, a safeguard of your public security data system, is there for you!