Big Data components --nova

nova configuration provides a calculation example, supports the creation of a virtual machine, bare metal server. In nova as a set of daemon running on linux, in order to provide the service

nova node into two categories: one is nova-compute computing node, the control node is further

 

 

 

nova api: responsible for receiving and responding to external requests, external access only way to nova and accept external requests through the Message Queue to send requests to other service components

nova scheduler: cloud host dispatch, the decision should create a virtual machine on which the compute nodes. A decision should be scheduled on a virtual machine which physical node (filtration, weights are calculated)

nova computer: running on a computing node, by receiving and Queue Management Message the KVM life cycle. Virtual machine management core services, by calling the virtual hypervisor API Lifecycle Management

hypervisor: running on the hypervisor node, the bottom of the hypervisor calculations. Different virtualization technology to provide their own hypervisor

nova conductor: computing middleware node to access the database, Nova-Computer often need to update the database (such as virtual machine environment), but does not directly access the database, but this task will be entrusted to the nova-conductor

consoleauth Nova : Authorization Authentication Console, you need to open the vnc need consoleauth authorization certification

nova novncporxy: providing a proxy used to access a running instance

spicehtml5proxy-Nova : Based on html5 SPICE browser access

nova cert : nova cert server daemon to provide services x509 certificate

Message Queue: Delivery Center messages between daemon. nova by message queue as the information service sub-relay station

SQL database: state storage cloud host at build time and running cloud infrastructure, including: the type of available instances, instances in use, available networks project

flavor: specifies the host memory, the size of the cpu, etc., and can be used to limit the number of parameters of the virtual machine, using the same to create virtual machines out of flavor, remained the same in the specifications

availability_zones: available area polymeric host, host group may be considered further subdivide the available area, and only the host group is available to administrators visible area of ​​the user (host may belong to multiple groups, but only a part of the available area)

host aggregates: A host group is a group of hosts set of compute nodes, open only to the default administrator, you can modify the access permissions through policy configuration

aggregates: host group management metadata can be added to the host, or remove a host, you can also manage metadata

Guess you like

Origin www.cnblogs.com/dong-liu/p/11871537.html