Shulou(Shulou.com)06/01 Report--

This article mainly analyzes the relevant knowledge of what the billion-level API gateway in big data architecture refers to, the content is detailed and easy to understand, the operation details are reasonable, and has a certain reference value. If you are interested, you might as well follow the editor to have a look. Let's follow the editor to learn more about "what is the 100 million-level API gateway in big data's architecture".

The basic functions of API gateway include unified access, protocol adaptation, traffic control and fault tolerance, and security protection.

2. Common tools of API: RAML, Swagger

The life cycle of API: Design (design), Build (build), Test (test), Document (document), Share (release), run (run), DownLine (offline).

Seven weapons of gateway system: degradation, current limit, circuit breaker, configuration hot update, asynchronism, thread pool isolation, pipeline technology

Fifth, the bottleneck leading to the performance of traditional gateways:

1 、 CPU

CPU utilization: shows the percentage of CPU load occupied by the program in real time during operation: shows the average number of tasks using and waiting to use CPU over a period of time. You can use the uptime or top command to view the load of the system. The eg: uptime command shows load averages: 1.74 1.58 1.60. The three numbers represent the average load of the system within 1 minute, 5 minutes and 15 minutes, respectively. Note: high CPU load does not mean high CPU utilization, the two are not related, can be compared to the load is queuing, the utilization rate is to see a doctor, can not say queue leader, the speed of the doctor to see a doctor is fast.

2. Memory

Buffering, caching and memory computing. The larger the memory, the faster the computing speed. Using free, top, vmstat and other commands can display the output of system memory utilization.

3. Disk

Disk utilization: disk space usage disk load percentage: the disk frequently executes the number of util O requests. Using the command "iostat-x 1 10", it shows that% use O is close to 100%, which means there are too many Imax O requests, and the disk may have a bottleneck.

4. Network

The quality of the network directly affects the response time of users. If the network becomes slow and the number of requests is too large, the number of requested tasks will be squeezed and all CPU resources will be exhausted.

VI. Other ways to improve performance

1. De-library and multi-level cache

Access to the gateway does not recommend the direct use of traditional databases (eg: mysql), but the use of caching. Cache rating, level 1 cache (JVM cache), level 2 cache (Redis), level 3 cache (Redis persistence)

2. Hot update

The behavior of changing the property values of a program without restarting the server The common hot update methods are MQ, RPC and ZooKeeper. MQ sends the data that needs to be updated to the MQ queue, the consumer updates the attribute values of the object after receiving the data, the RPC connects directly to the corresponding server, and the server updates the data after receiving the RPC request. Zookeeper receives updates to trigger local object data, all of which do not need to restart the server. On "big data architecture 100 million-level API gateway refers to what" is introduced here, more related content can be searched for previous articles, hope to help you answer questions, please support the website!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.