The HotSpot JVM takes some time to profile a running Java application for hot spots in the code and then optimizes by compiling (to assembly) and inlining (when possible) these hot spot methods. That’s great because the JIT (just-in-time) compiler can surgically and aggressively optimize the parts of your application that matter the most instead of taking the AOT (ahead-of-time) approach of compiling and trying to optimize the whole thing beforehand. Continue reading
Exploring the Sequencer Architecture through our SimSequencer
SimSequencer is a framework that lets you simulate through code the majority of the aspects of the sequencer architecture without any networking involved. It can be very useful for prototyping, testing and learning purposes. With SimSequencer you can code your nodes to test all the moving parts of your application as they were running and interacting in a real distributed system. And it has the same API of CoralSequencer. Continue reading
Writing a C++ CoralSequencer Node
Writing some C++ code that gets called by your Java code is trivial through shared libraries but a more interesting project is to do the inverse: to call Java code from a C++ system. In this article we write a C++ CoralSequencer node to perform a latency benchmark test. Continue reading
Performance Analysis: comparing C++ and Java
In this article we write two equivalent programs in C++ and in Java, in exactly the same way to do exactly the same thing: the (in)famous bubble sort algorithm. Then we proceed to measure the latency. On this experiment, Java was faster than C++ even with the -O3 compiler option. Continue reading
Shared Memory Transport x Multicast Transport
A CPU core is a scarce resource in high demand from all the different process running on a computer machine. When you are choosing how to allocate your nodes to available CPU cores, CoralSequencer gives you the option to run several nodes inside the same NioReactor thread, which in turn will be pinned to an isolated CPU core. As the number of nodes inside the same CPU core grows, fan-out might become an issue affecting latency, as now the thread has to cycle through all nodes as it reads multicast messages from the event-stream. In this article we explain (with diagrams) how you can deal with that scenario by choosing CoralSequencer’s shared-memory transport instead of multicast at the node machine level. Continue reading
CoralSequencer’s structured data serialization framework (Version 2.0)
CoralSequencer uses its own binary and garbage-free serialization framework to read and write its internal messages. For your application messages, you are free to use any serialization library or binary data model you choose. The fact that CoralSequencer is message agnostic gives you total flexibility in that decision. But you can also consider using CoralSequencer’s native serialization framework described in this article. Continue reading
Building a first-class exchange architecture with CoralSequencer
In this article we will explore an architecture employed by some of the most sophisticated electronic exchanges, the ones that need to handle millions of orders per day with ultra-low-latency and high-availability. Most of the exchange main architecture components will be presented as CoralSequencer nodes and discussed through diagrams. The goal of this article is to demonstrate how a total-ordered messaging middleware such as CoralSequencer naturally enables the implementation and control of complex distributed system through a tight integration of all its moving parts. This article doesn’t mention specifics about any exchange’s internal systems and instead talks about the big picture and the general concepts, which will vary from exchange to exchange. Continue reading
CoralSequencer’s structured data serialization framework
CoralSequencer uses its own binary and garbage-free serialization framework to read and write its internal messages. For your application messages, you are free to use any serialization library or binary data model you choose. The fact that CoralSequencer is message agnostic gives you total flexibility in that decision. But you can also consider using CoralSequencer’s native serialization framework described in this article. Continue reading
The Diamond Queue (Demultiplexer to set of threads to Multiplexer)
In this article we present the DiamondQueue, which is nothing more than thread A sending through a demultiplexer a bunch of requests to a fixed set of worker threads, then this set of worker threads sending the results to thread B through a multiplexer. It is important to note that thread A and thread B can be the same thread. The DiamondQueue also supports lanes to enforce message order when needed. Continue reading
Is CoralSequencer really deterministic? What about the clock?
In this article we explore the deterministic nature of CoralSequencer and how its total ordered message stream is a natural enabler of high-availability clusters. Continue reading