Search for: All records

he Universal Globally Adaptive Load-balance Routing (UGAL) with global information, referred as UGAL-G, represents an ideal form of adaptive routing on Dragonfly. UGAL-G is impractical to implement, however, since the global information cannot be maintained accurately. Practical adaptive routing schemes, such as UGAL with local information (UGAL-L), performs noticeably worse than UGAL-G. In this work, we investigate a machine learning approach for routing on Dragonfly. Specifically, we develop a machine learning-based routing scheme, called UGAL-ML, that is capable of making routing decisions like UGAL-G based only on the information local to each router. Our preliminary evaluation indicates that UGAL-ML can achieve comparable performance to UGAL-G for some traffic patterns. 

Application performance variability caused by network contention is a major issue on dragonfly based systems. This work-in-progress study makes two contributions. First, we analyze real workload logs and conduct application experiments on the production system Theta at Argonne to evaluate application performance variability. We find a strong correlation between system utilization and performance variability where a high system utilization (e.g., above 95%) can cause up to 21% degradation in application performance. Next, driven by this key finding, we investigate a scheduling policy to mitigate workload interference by leveraging the fact that production systems often exhibit diurnal utilization behavior and not all users are in a hurry for job completion. Preliminary results show that this scheduling design is capable of improving system productivity (measured by scheduling makespan) as well as improving user-level scheduling metrics such as user wait time and job slowdown. 

Dragonfly class of networks are considered as promising interconnects for next-generation supercomputers. While Dragonfly+ networks offer more path diversity than the original Dragonfly design, they are still prone to performance variability due to their hierarchical architecture and resource sharing design. Event-driven network simulators are indispensable tools for navigating complex system design. In this study, we quantitatively evaluate a variety of application communication interactions on a 3,456-node Dragonfly+ system by using the CODES toolkit. This study looks at the impact of communication interference from a user’s perspective. Specifically, for a given application submitted by a user, we examine how this application will behave with the existing workload running in the system under different job placement policies. Our simulation study considers hundreds of experiment configurations including four target applications with representative communication patterns under a variety of network traffic conditions. Our study shows that intra-job interference can cause severe performance degradation for communication-intensive applications. Inter-job interference can generally be reduced for applications with one-toone or one-to-many communication patterns through job isolation. Application with one-to-all communication pattern is resilient to network interference.