Apache Geode Distributed Systems Member

 

 
 

       Distributed system members are programs that connect to a Geode distributed system. You can configure members to belong to a single distributed system, and you can optionally configure them to be clients and servers of members in other distributed systems, and to communicate with other distributed systems.

Member Overview

        Connect to the Geode system when a distributed system member (or simply member) creates a Geode data cache. The distributed system this member belongs to is configured through the Geode properties. See gemfire.properties and gfsecurity.properties (Geode properties) . The Geode property specifies all the necessary information for system members to start, initialize, and communicate.

Note: You cannot change a member's properties while the member is already connected to the distributed system. Use these properties to define:

  • How to find and communicate with other system members
  • How to conduct log and statistics activities
  • Which persistence configuration to use when the cache and data area are initialized orcache.xml文件。
  • Other options include event merging, how to handle dropped calls, and security settings.

Distributed System Membership and System Topology

       Each Geode process is a member of a distributed system, even if the distributed system is determined to be independent, with only one member. You can run a standalone distributed system or you can scale the composite system horizontally or vertically. Check out "General Concepts of Topology and Communication"

  • Peer-to-peer distributed systems. Members define the discovery property that the same members belong to the same distributed system and are peers between them.
  • Client/Server installation. A client/server topology uses relationships that you configure among members of multiple distributed systems. You can configure some or all peers in a distributed system to act as cache servers for clients connecting from outside the system. Each server can host multiple client processes, managing cache access for a vertically tiered cache configuration. You use the client cache configuration to configure client applications to connect to the server. Clients operate as independent distributed system members, with no peers, so all data updates and requests reach the server.

Multisite installation

       A multisite topology uses relationships that you configure among members of multiple distributed systems. With this configuration, you can optionally pair two or more distributed members for data distribution. This is usually done by geographically distinct sites. You configure a subset of peer nodes at each distributed system site with gateway senders and/or gateway receivers to manage events distributed across sites.

       In the context of a separate distributed system, unless otherwise specified, remote members refer to other members of the same distributed system. In client/server and multiple site installations, remote usually refers to members in other distributed systems. For example, all servers are remote to clients connecting to them. Each client runs independently and only connects to the server layer, so all servers and their other clients are remote to individual clients. All gateway receivers are remote to the gateway senders that connect to them from other distributed systems, and are also remote to the peer nodes of those gateway senders.

Guess you like

Origin http://10.200.1.11:23101/article/api/json?id=327058735&siteId=291194637