Although vendor-written, this contributed piece does not promote a product or service and has been edited and approved by Network World editors.
Black Friday are two words that shoppers can’t wait to hear. The shopping lists are ready, money has been set aside and now all shoppers have to do is sit in front of their computers and start filling up their e-carts. Imagine their anger and disappointment if their transactions fail to go through or if they can’t even access your e-commerce database?! To make sure that Black Friday doesn’t become a nightmare for you and your customers, be well prepared by using load balancers or SQL Server Load Balancing. The following tips should help:
- Review last year’s performance. Last year’s sales figures and shopping trends are a good point to start. Study them and understand them. Include factors such as throughput, number of transactions/second, CPU and memory utilization, application and disk response time, batch requests/second, etc. Based on your observations, forecast the current year’s load and assess if the existing hardware is capable of bearing the forecasted capacity. Adjust the hardware, if necessary. Perform load testing on your servers to simulate the point-of-sale transactions.
- Test high availability solutions with zero downtime. Your database server must be able to handle an estimated number of incoming requests during the peak traffic hours. The application and database should be configured for connection pooling and set to an optimal value. Your database server should be configured for hardware and database failures as well. Test the existing high availability solutions, such as ensuring that SQL server high availability is in place.
- Conducting a health check of your database server. Make sure your database server is healthy and perform a health check on the NIC cards, memory and hard disks. The power connectivity and power backups must be in place and a thorough database configuration check must be performed. Look for all the common configurations and check that the memory settings are properly set. The threads, parallelism and other settings must be in synch with the hardware.
- Archive your old data. Data can never go to waste. If you haven’t been using your old data for a while, don’t delete it, archive it. This will help reduce the time to perform various database activities, such as updating statistics, and will also improve the overall query performance. You should export the archived data to a backup file or another database.
- Designing a solid backup strategy. You must evaluate your backup and failover strategy. Make sure you have a redundant copy of your backup in a data center or at an external site. Perform a disaster recovery test to ensure that your system runs properly on the DR server and check if there are any chances of disaster. Ensure that all your backups can be efficiently restored as and when required. It is essential to perform a dry run of your backup and restore procedure.
- Analyze all errors. Keep track of all the issues and errors that you have recently faced as well as those of a recurring nature. Analyze all such problems, even those that have been fixed recently. Review all the changes that have been made to your database configuration recently. These changes may include those made due to some incident or problem. They may also include patch and application updates, added schedules, structural changes, modified file patterns, etc. By extracting and analyzing these errors, you will be better equipped to respond to future issues when they arise.
Just because your infrastructure has been working well for the past few months or weeks does not mean it is capable of handling the Black Friday rush. For a smooth shopping experience, both for your consumers and your organization, you must make sure that your infrastructure can scale beyond your traffic expectations and incorporate high availability. Be prepared for any instances of server failure with database load balancing software.
A self-proclaimed tech geek, with a passion for ScaleArc’s disruptive technology innovation in database load balancing, the author has a passion for dissecting tech topics such as transparent failover, centralized control, ACID compliance, database scalability and downtime effects.