microservices architecture principles

Tags: blog HTTP Io OS ar use for strong sp1. "Open-Close" principle (OCP)Software entities shocould be open for extension, but closed for modification.It is open to extensions and closed to modifications.2. Lishi replacement principle (LSP)Where the base class is applicable, the subclass must be applicable.3. Dependency reversal principle (DIP)Dependency abstraction is required.4. dummit's Law)An object should have as little understanding as possible about other objects.5. Interface isolation pr

Get object content and metadata TD style= "Margin:0px;padding:3px;border-color: #C0C0C0; border-collapse:collapse;" > Create, update, or copy objects Update object metadata Get object metadata The detailed API specification can be consulted in the developer's Guide. Application development can be implemented using Python bindings already contained in the SWIFT project itself, and in other programming languages, you can refer to the Rackspace compatible Swif

deserialization.Function call layer: locate the function to be called and execute the function. Hadoop RPC uses a java reflection mechanism and a dynamic proxy to call functions.Network Transmission Layer: message transmission between the Client and the Server. Hadoop RPC uses a TCP/IP-based Socket mechanism.Server Processing Framework: network I/O model. Describes how the client interacts with the server. Its design directly determines the server's concurrent processing capability. Hadoop RPC

scalability design of Application server cluster 1. HTTP redirect Load balancer (no longer used) 2. DNS domain name resolution load Balancer 3. Reverse Proxy load Balancing 4. IP load Balancing 5, Data link layer load balancing (also known as triangular mode, or Dr mode) 6. Load Balancing algorithm (1), polling (Round Robin RR) (2), weighted polling (Weighted Round Robin WRR) (3), stochastic (random) (4), Minimum link (Least Connections) (5), So

; Multiple from library for data Select)6) CDN AccelerationAdd a layer of reverse proxy server and CDN Accelerator in front of load balancing, and return some cached data to the server early so that the page reacts fasterCDN Accelerator--Reverse proxy--load Balancer Scheduler---> (Multiple consuming programs distributed deployment)---> External distributed cache---> Database server and File server docking7) Distributed database and Distributed File serverWhen the amount of data in the database i

ImportAerospike#Configuration for the clientConfig = { 'hosts': [ ('127.0.0.1', 3000) ] } #Create a client and connect to the databaseClient =aerospike.client (config). Connect ()#Records is addressable via a tuple of (namespace, set, key)Key = ('Test','Demo','Foo') #Write a record Client.put (key, {'name':'John Doe', ' Age': 32 }) #Read a record(Key, metadata, record) =client.get (Key)#Close Connection to ClusterClient.close () Performance comparison See Http://www.infoq.com/cn/news