Brief introduction of Linux cluster type, System extension mode and scheduling method

Source: Internet
Author: User

Linux Cluster Type:
LB Load Balancer Cluster
Reasonable allocation of user requests to the back-end server processing, the cluster load balance, and reasonable use of computing resources.
HA High Availability Cluster
In the event of a critical single-point host failure in the cluster, which can cause the entire cluster to crash, redundancy of these critical single-point hosts is required to improve cluster availability
The redundant host makes heartbeat detection on the critical single-point host and detects that the failure will take over the service of the original server, which is a highly available ha (high availabllity) cluster
HP High Performance clusters

How to extend the Linux cluster system
When limited to a single host hardware software, or when the application needs exceed the existing hardware resources, there are usually two scenarios to solve:
1. Use better hosts instead of better hosts (scale up)
2. Combine multiple hosts to complete a single task, dispersing the task (extending inward), which is what is commonly said load-balanced cluster (load Balancing cluster) short as a lb cluster
The problems brought by:
1> decentralized user requests are easy to resolve, but tracking or tracking user requests can be a challenge,
For example, a user in a time to browse a product and add a shopping cart, this request is a server in the cluster processing,
While viewing another item in B and adding a shopping cart, this request is likely to be handled by a server in cluster B. How do I sync or track user requests and browsing history?
How do I synchronize the hosts of the same task in the 2> cluster? When a user uploads a file to a host of a cluster, how can I find this file when the user requests the B host?
Solve:
1>
A. The session stickiness can bind a user request to a host, as long as the request sent by the same user is sent only to a fixed host (session stickiness), but there is a problem when this server goes down
Cookies
User Bindings (IP)
B.session replication cluster, the sessions of each host are replicated synchronously, but although the dependency of the session stickiness, but will bring greater problems, each host will store a large number of session, the host will not stop replication, so it occupies a large number of resources, not suitable for large clusters
C.session shared Storage server, the session generated by each server in the cluster is saved on the shared memory storage server
2> shared storage
The scheduling mode of the LVS load Balancing cluster:
Static algorithm: Allocated only according to the scheduling algorithm, regardless of the actual connection situation of the backend
①.RR Polling (Round Robin), as the name implies, means that the request is distributed evenly to the backend host processing in a certain order or sequence, which is suitable for situations where the performance difference between the host and the requesting hosts is small.
②.WRR Weighted Polling (Weighted Round Robin), when the back-end processing requests between the host, the performance difference between the poor performance of the host relatively few processing tasks, high-performance host as much as possible processing tasks, when using weighted polling is a more convenient way to use
③.sh Source Address hash dispatch, with the original address as the key word static hash table to determine the need for Rs.
④.DH the target address hash dispatch, the target address as the key word to do static hash table to determine the need for Rs.
Dynamic algorithm: Distribution according to the actual situation of the back-end RS Connection
①.LC minimum number of connections is dispatched (least connection), the user request is assigned to the least connected RS, and when all RS connections are 0 o'clock, it is scheduled by polling mode.
The ②.WLC weighted minimum number of connections dispatch (weighted least connection) assigns user requests to the RS with the fewest back-end connections and the highest weights.

Brief introduction of Linux cluster type, System extension mode and scheduling method

Related Article

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.