1 / 16

Study of Server Clustering Technology

Study of Server Clustering Technology. By Thao Pham and James Horton For CS526, Dr. Chow. Importance of Server Clustering. Provide better service for end users Better throughput Faster turnaround time Higher availability Better reliability

acorbett
Download Presentation

Study of Server Clustering Technology

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Study of Server Clustering Technology By Thao Pham and James Horton For CS526, Dr. Chow

  2. Importance of Server Clustering • Provide better service for end users • Better throughput • Faster turnaround time • Higher availability • Better reliability • Provide dynamic scalability for businesses based on growing customer base

  3. What is a Cluster? • "a parallel or distributed system that consists of a collection of interconnected whole computers that is utilized as a single, unified computing resource.“ • G. Pfister • One or more computers connected together to provide services from the “middle teir”. The number of computers and what services they provide should be transparent to the end user.

  4. independent cluster methods the member servers are not aware of the other members redundancy is achieved by a third service (can be SPOF) weakness of availability : administrators to manually tell the cluster what to do

  5. Centralized The Name servers keep track of which servers are up. same weakness as the independent, name servers can be SPOF. Servers need to know about additional name servers require server,can prolong the recovery time of the cluster.

  6. global sharing • cluster members know about one another. • The information is presented globally in the cluster and locally in each machine. • instant processing. • takes longer to form due to initial traffic • takes one single network call for a remote lookup.

  7. HP TruCluster Server

  8. TruCluster Componets • Cluster file System (CFS) • Device request dispatcher (DRD) • Connection Manager • Quorum Disk (optional) • Cluster Alias • Cluster Application Availability(CAA)

  9. Cluster file System (CFS) • Shared by all members, including a single /root file system /usr and /var • Each member has it own boot partition

  10. Quorum Disk • Used for greater availability. • has a vote of if present. • when any node goes down, the cluster still maintains enough votes to stay active.

  11. Device request dispatcher (DRD) • controller of all I/O to physical devices and makes disks available to all cluster members.

  12. Cluster Alias • It is an IP address that makes some or all of the systems in a cluster look like a single system to TCP and User UDP applications

  13. Connection Manager • Forms a cluster, adds members to a cluster, and removes members from a cluster • Tracks which members in a cluster are active • Maintains a cluster membership list that is consistent on all cluster members • Provides timely notification of membership changes using Event Manager (EVM) events • Detects and handles possible cluster partitions

  14. Cluster Application Availability(CAA) • provides high availability for applications • monitor applications, and the state of resources. • Load balancing (automatically or manually) • resources, resource profiles and action scripts

  15. Conclusion • We learned quite a bit about a few different implementations of server clustering technologies. • The surprising thing is not how different they are in their implementation, but how similar they are in their architecture. • Ease of use award goes to Windows Server2003, but when things go wrong, it appears that TruCluster is better at reporting and deciphering the problem.

More Related