Todos los libros

Profesional

Apps Acerca de Coach Iniciar sesión Empezar a leer

Quantitative Finance · Glosario

¿Qué es Distributed tracing, span?

También llamado: distributed tracing · span

Definition 28.3 Research, Data and Risk Platforms · Capítulo 28 — Observability and Incident Response

Distributed tracing follows one unit of work — a request, a market-data message, an order — across the services it passes through. Each step is a span, a unit of work with a start, an end, a trace identifier shared by all the steps and the identifier of the step that caused it, so that the path can be rebuilt as a tree.

The spans of one traced market-data message through the chain (log scale). Capture, storage and positions take 0.02, 0.2 and 0.1 ms; the risk service computes in 1 ms. What changes when the risk service falls behind is its queue: 50 ms on a normal day, 60 seconds at the height of the slow-consumer incident. Data: fig_observe.py.
Figure 28.1. The spans of one traced market-data message through the chain (log scale). Capture, storage and positions take 0.02, 0.2 and 0.1 ms; the risk service computes in 1 ms. What changes when the risk service falls behind is its queue: 50 ms on a normal day, 60 seconds at the height of the slow-consumer incident. Data: fig_observe.py.
Leer en el capítulo →