When any kind of slowdown occurs within a SharePoint system, the immediate solution is to go for addition of more web servers in order to handle the load in a better fashion. However, these front-end servers are dependent upon the same back-end database server. Incidentally, it is the back end where most of the issues crop up. This is because it depends upon SQL as a file as well as a database system. The processing demands of SP brings to light the disk I/O as the most most common bottleneck. Let us discuss the steps that need to be taken for avoiding or mitigating the logjams that can hinder SQL Server throughput.
It is advisable to get the binary large objects such as the file attachments, out from the database. SQL Server as well as SharePoint render support to Microsoft’s RBS ( Remote BLOB Storage) technology. This conveniently allows putting all the Excel, Word or other documents back within the NTFS file system wherein they actually belong.
NTFS has the capability of writing and reading files rapidly. On the other hand, SQL Server needs to manage them along its database pages of 8KB, which can consume a great amount of disk space as well as server resources. To do away with this, files are kept by RBS in NTFS and pointers are also provided. This implies that the file attachments can be written to completely separate disks that the one which contains the chief database. This helps in maintaining speed by reducing disk contention.
Regular Maintenance and Premium Storage
One of the most significant performance management investments with respect to SharePoint is buying fast storage. Initially, it is good for organizations to focus on gaining momentum in planning the purchases since this constitutes one of the most costly aspects of storage. Rapid storage area networks coupled with robust drive caches are capable of rendering high quality storage performance, which is a great need of SharePoint. It is advisable to make sure that adequate amount of storage capacity is available for every data developed with respect to a SP installation.
Regular maintenance is extremely important for ensuring the efficiency of ongoing SQL server. The SP collaboration platform is able to generate considerable amount of data, especially in case of heavily used document stores wherein versioning occurs in order to track changes within documents. Moreover, manifold versions are maintained for reference purpose. Appropriate database maintenance, incorporating reorganizing or rebuilding indexes along with updating of statistics, allows SQL Server to remain clean-cut and suitable for quality performance. Moreover, database files can be defragmented as the fragmentation level starts to go beyond 7% or 8%.
Breaking it up
Owing to the inherent scalability of the web, SharePoint is able to scale without any major issues. SQL server is unable to do it in a similarly convenient manner. Because of this, managers of a number of big SharePoint environments develop manifold server farms, each having a dedicated back-end system. For instance, a company can possess a set of SharePoint servers as well as a related database for the document library across the company, for special projects, for users’ blogs and the like. This yields better results than putting everything into the same SQL Server database.
SharePoint administrators can also consider acquiring certain extra tools in addition to visiting Microsoft’s TechNet website, which presents solutions to particular performance problems and best practices to ensure speedy crawling of SharePoint environments.
Defined Performance Metrics
It is also a significant aspect to develop some measurable and reasonable performance expectations for SharePoint development. The basic formula is to try and base these upon the end-user experience. Simply speaking, acceptable performance can be defined by the maximum time a user has to wait during checking of a document accomplishing a task or accessing a blog. You can then break down these metrics into a number of prospective back-end performance metrics. This is a great means to assess the performance quality.
When user-experience metrics get defined, it is required to measure them regularly in order to identify the trends. In this way, it will be convenient to detect whenever a SharePoint system fails to live up to the organization’s expectations.