Database Reference
In-Depth Information
CHAPTER 8
Cassandra Performance Tuning
As storage hardware has become cheaper and more efficient in terms of capacity, organ-
izations are able to choose less expensive, lowmaintenance storage options for their tera-
bytes of data. Obviously cost effectiveness has always been an important factor behind
technology selection. Storage and retrieval of large data are requirements, but an eco-
nomical solution should not be at the cost of performance. Data, whether large or small,
isn't good unless it can be used effectively and have analytics applied to it.
With proven opportunities from real-world big data-related use cases, many organiz-
ations and startups are exploring business ideas around large data. Real-time feeds are
very much in demand for companies targeting an online audience. Applications
nowadays deal with gigabytes of data per second to be processed and analyzed, which
clearly shows that performance is an important component for modern big data-based
applications.
In this chapter we will discuss
•
Key performance indicators
•
Cassandra cache configurations
•
Discuss Bloom filters and garbage collection
•
Cassandra stress testing
•
Yahoo Cloud Serving Benchmarking
Understanding the Key Performance In-
dicators
