Video Screencast Help

Whitepaper: NetBackup™ Architecture Overview

Created: 04 Jan 2010 • Updated: 04 Jan 2010 | 2 comments
Language Translations
Swathi Turlapaty's picture
+3 3 Votes
Login to vote

1.0 Introduction

The purpose of this document is to provide an overview of the requirements for a basic NetBackup™ architecture to help customers understand which components are required in a data protection solution. Where possible, an example of a component will be noted; however, that component could be swapped for a similar component with the same bandwidth characteristics based on the vendor of choice. A much more comprehensive document, the Veritas NetBackup Backup Planning and Performance Tuning Guide, is available for free from Symantec at:
http://seer.entsupport.symantec.com/docs/307083.htm

This overview is not meant to replace the above referenced document or other documents that provide a deeper understanding of the best way to create a data protection environment, nor is it meant to replace an onsite Professional Services architecting engagement. It is intended to be a quick overview of how a NetBackup architecture is created to assist customers who are new to NetBackup or who are considering upgrading/scaling an environment. This document is designed to help you understand what will be required from a hardware, software and personnel standpoint when NetBackup is deployed; the above referenced document can be used as more of a deep dive.

With the deployment of newer, multi-core servers and PCI Express™ (PCIe) bus, server choices have become less of an issue than they were even a couple of years ago. Most modern servers are capable of performing the master server and/or media server tasks very well; therefore the recommendation for NetBackup customers is to choose the hardware they are comfortable with, or a vendor they have a relationship with.

Note that not all features described in this document are available with all currently supported versions of NetBackup on all platforms and with all hardware configurations. You can learn more about what is and is not supported in the various hardware compatibility lists (HCLs) on the Symantec Support website. This link provides a gateway to all current HCL documents:
http://seer.entsupport.symantec.com/docs/303344.htm

1.1 Glossary of Terms

The following are some of the terms with brief explanations that are used in this document:

  • Bandwidth—Refers to a component’s physical limitation when sending data from one point to another. It also refers to how much work an individual backup administrator (FTE) is capable of. Bandwidth has many variables. ·
  • Master Server—The controller of the NetBackup environment. The master server schedules the jobs, maintains the policies, allocates the resources, stores the catalog metadata, etc.
  • Media Server—The workhorse of the NetBackup environment. Data passes through the media server to the final repository in a typical scenario.
  • NetBackup Client—Any system that is sending data to a media server. In the case of a SAN media server, the client and media server are the same system.
  • Policy—The “who, what, where, when and how” of a backup. The policy can include multiple clients, multiple schedules, multiple file listings, etc.
  • Storage Unit—A virtual representation for physical storage where NetBackup will send data.
  • Recovery Point Objective or RPO—Determines how much data loss is acceptable. For example, if backups occur only once every 24 hours, then the maximum data loss could be as many as 24 hours. In order to meet a specific RPO, careful planning and testing are required. Short RPOs require backup and recovery scenarios that typically will not involve traditional backup from disk to tape.
  • Recovery Time Objective or RTO—Determines how long the environment has until it needs to be back online. In many instances this could be less than an hour, in which case traditional backup to disk/tape is not the right choice for protection. Similar to the RPO, proper planning and extensive testing are needed to document steps for recovery to shorten the amount of time recovery takes.
  • Service Level Agreement or SLA—The SLA of an environment is usually based on input from those whose data is beingprotected and determines how data should be backed up, the RPO/RTO associated with the data, how long it is stored (retention), etc.

To reaad the complete article, please download the PDF.

Comments 2 CommentsJump to latest comment

Kiran Bandi's picture

Thanks for the useful information...

+1
Login to vote
Baski's picture

Hi Swathi,

 

Is it possible to provide the architech difference between NBU 6.5.X and NBU 7.5.X

0
Login to vote