Big Data and Hadoop: Best Practices

What are the best practices for handling big data with Hadoop? How do you optimize performance in large-scale Hadoop clusters, and what tools complement its functionality for data analysis?