Combing the Communication Hairball: Visualizing Parallel Execution Traces using Logical Time

Abstract

Fig. 1: Logical timeline and clustered logical timeline views from Ravel, a tool for visualizing parallel execution traces. Events are represented by boxes, colored by their wall-clock delay. The use of logical time reveals communication patterns and leverages developers' understanding of their program's structure. We use the logical time structure to cluster on any metric, which allow us to represent large-scale traces using explorable clusters while still depicting messages with full timelines for a subset of processes.Abstract-With the continuous rise in complexity of modern supercomputers, optimizing the performance of large-scale parallel programs is becoming increasingly challenging. Simultaneously, the growth in scale magnifies the impact of even minor inefficiencies -potentially millions of compute hours and megawatts in power consumption can be wasted on avoidable mistakes or sub-optimal algorithms. This makes performance analysis and optimization critical elements in the software development process. One of the most common forms of performance analysis is to study execution traces, which record a history of per-process events and interprocess messages in a parallel application. Trace visualizations allow users to browse this event history and search for insights into the observed performance behavior. However, current visualizations are difficult to understand even for small process counts and do not scale gracefully beyond a few hundred processes. Organizing events in time leads to a virtually unintelligible conglomerate of interleaved events and moderately high process counts overtax even the largest display. As an alternative, we present a new trace visualization approach based on transforming the event history into logical time inferred directly from happened-before relationships. This emphasizes the code's structural behavior, which is much more familiar to the application developer. The original timing data, or other information, is then encoded through color, leading to a more intuitive visualization. Furthermore, we use the discrete nature of logical timelines to cluster processes according to their local behavior leading to a scalable visualization of even long traces on large process counts. We demonstrate our system using two case studies on large-scale parallel codes.

A visual performance analysis framework for task‐based parallel applications running on hybrid clusters

“…8 The tool fully replaces the notion of time (along the X axis of space/time views) by the order and simplicity of logical clocks. However, the interaction among threads, represented by arrows from source to destination, usually makes the representation hard to understand.…”

“…This problem has, for example, been addressed in Ravel. 8 The tool fully replaces the notion of time (along the X axis of space/time views) by the order and simplicity of logical clocks.…”

Multi‐scale navigation of large trace data: A survey

“…Significant efforts are made in visualizing HPC performance data. Isaacs et al organize trace data in a logical time ordering for the ease of understanding performance anomalies . Isaacs et al provide a rich survey of visualization tools for performance analysis.…”

“…Isaacs et al organize trace data in a logical time ordering for the ease of understanding performance anomalies. 91 behavior at the architecture level, but it needs the architecture model of the system to be available. This may not always be feasible for all systems.…”

Visual monitoring of process runs: An application study for stored procedures

“…Other visualizations depict different tasks or processes on a horizontal timeline, for instance, as rows connected by links [25,29] or as bundled bars [15] (more examples are surveyed by Isaacs et al [26]). In this context, however, we only found one approach that discerns and visualizes multiple executions: Trümper et al [33] compare two executions by juxtaposing their hierarchical stack trace visualizations and connecting similar execution phases by bundled links.…”