One of the cornerstones of ScyllaDB is the I/O Scheduler, described in details at the moment of its inception in a two-part series that can be found here (part 1) and here (part 2). In the two years in which ScyllaDB has been powering mission-critical workloads in production the importance of the I/O Scheduler was solidified and as our users have attested themselves, it plays a key part in isolati

Cache control For both read/write and mmap, caching is the responsibility of the kernel. The majority of the system’s memory is given over to the page cache. The kernel decides which pages should be evicted when memory runs low; when pages need to be written back to disk; and controls read-ahead. The application can provide some guidance to the kernel using the madvise(2) and fadvise(2) system cal

In a datastore like ScyllaDB, there are many actors competing for disk I/O. Examples of such actors are data writers (in ScyllaDB’s parlance they can be either memtable or commitlog writers), and a disk reader fetching the data to serve a cache miss. To illustrate the role that competition plays, if we are just issuing disk I/O without resorting to any fairness or balancing consideration, a reader

This is the third part in our blog series on ScyllaDB testing. Part 1 covers Apache Cassandra compatibility testing, and Part 2 covers Jepsen tests. Some of the hardest to test code in any database is the code that handles filesystem errors. Such errors are rare, but handling them is critical. Using a simulated filesystem to inject errors is a good way to make sure that the error-handling code is

Welcome to the asynchronous world In the networking world, asynchronous processing is so common that it is practically taken for granted. An application sends a message to a remote server, and goes off to do something else. When the response arrives, it is picked up and the application continues from where it left off. There are many application frameworks and even languages that make this easy. I

Introduction Seastar provides a programming environment that abstracts away most of the problems of multi-threaded programming using a thread-per-core model. Locks, atomic variables, memory barriers, lock-free programming, and all of the scaling and complexity that come from them are gone. In their place, Seastar provides a single facility for inter-core communications. This is, of course, great f

This is the second post in a series of four about the different compaction strategies available in ScyllaDB. In the previous post, we introduced the Size-Tiered compaction strategy (STCS) and discussed its most significant drawback – its disk-space waste, a.k.a. space amplification. In this post, we will look at Leveled Compaction Strategy (LCS), the first alternative compaction strategy designed

This is the first post in a series of four about the different compaction strategies available in ScyllaDB. The series will look at the good and the bad properties of each compaction strategy, and how to choose the best compaction strategy for your workload. This first post will focus on ScyllaDB’s default compaction strategy, size-tiered compaction strategy. Although we’ll discuss the ScyllaDB da

As most earthlings are aware by now, two severe attacks under the names of Meltdown and Spectre were currently disclosed and it affects pretty much everybody living in modern society. Although there is no defense against the more elaborate Spectre, there is a software defense against the more widespread Meltdown. However, there is a catch that set the Internet on fire over the past few days. To pr