Tag: performance

The database is a critical and vital part of any business or organization.  The growing trends predict that 82% of enterprises expect the number of databases to increase over the next 12 months. A major challenge of every DBA is to discover how to tackle massive data growth, and this is going to be a most important goal. How can you increase database performance, lower costs, and eliminate downtime to give your users the best experience possible? Is data compression is an option? Let’s get started and see how some of the existing features can be useful to handle such situations.

In this article, we are going to learn how the data compression solution can help us optimize the data management solution. In this guide, we’ll cover the following topics:

• An overview of compression
• Benefits of compression
• An outline about data is compression techniques
• Discussion of various types of data compression
• Facts about data compression
• Implementation considerations
• and more…

Read More

Introduction

There are a number of situations which would warrant the movement of database files or transaction log files from one volume to another on the same server. These may include:

1. The need to format the volume assuming it was not formatted properly when SQL Server was installed. Recall that when installing SQL Server, it is recommended that 64K allocation unit size is used to format the volumes. If this is not done at the point of installation and needs to be done later, it will obviously require preserving a backup of the database fist or creating a new, properly formatted volume and moving the database to this new volume.
2. The need to use a new volume assuming the limits have been reached for the underlying storage. A good example would be the 2TB limit of a VMware Data Store. This is the case as of VSphere 5.0. Higher versions of VSphere have much higher limits.
3. The need to improve performance by managing IO. One more reason you may want to move datafiles is performance. There are cases where a database is created with multiple datafiles all sitting on one disk until it becomes obvious, as the database grows, that the you have created a “hot region” in the storage layer. One solution would be creating new data files and rebuilding clustered indexes, another would be moving data files.

Read More

NOTES:

• Windows Failover Clustering comprising two nodes.
• Two SQL Server Failover Cluster Instances. This configuration optimizes the hardware. IN01 is preferred on Node1 and IN02 is preferred on Node2.
• Port Numbers: IN01 listens on port 1435 and IN02 listens on port 1436.
• High Availability. Both nodes back up each other. Failover is automatic in case of failure.
• Quorum Mode is Node and Disk majority.
• Backup LAN in place and routine backup configured using Veritas

Read More