In this chapter, we discussed several different topics that address architectural concerns, all of which are important topics. Using transactions, you can ensure that the data in your unit of work is persisted correctly, and you can tighten the concurrency boundaries by setting the appropriate parameters, such as isolation levels and database locks. We also discussed compensating transactions to implement the undo functionality, and we covered using custom user locks to offer application users more control over the data.
We showed what to watch for when your application is deployed in a clustered environment and also how to implement an application using multiple database shards. Even though, throughout this book, we have discussed performance concerns, we further addressed other concerns related to performance in this chapter.
We also discussed how to reverse engineer a legacy application and the approaches to modernization efforts. And finally, we discussed licensing concerns, multi...