Watch out for the bully! job interference study on dragonfly network
X Yang, J Jenkins, M Mubarak… - SC'16: Proceedings of …, 2016 - ieeexplore.ieee.org
High-radix, low-diameter dragonfly networks will be a common choice in next-generation
supercomputers. Preliminary studies show that random job placement with adaptive routing …
supercomputers. Preliminary studies show that random job placement with adaptive routing …
Experience and practice of batch scheduling on leadership supercomputers at argonne
The mission of the DOE Argonne Leadership Computing Facility (ALCF) is to accelerate
major scientific discoveries and engineering breakthroughs for humanity by designing and …
major scientific discoveries and engineering breakthroughs for humanity by designing and …
Evaluation of an interference-free node allocation policy on fat-tree clusters
Interference between jobs competing for network bandwidth on a fat-tree cluster can cause
significant variability and degradation in performance. These performance issues can be …
significant variability and degradation in performance. These performance issues can be …
Performance optimality or reproducibility: that is the question
The era of extremely heterogeneous supercomputing brings with itself the devil of increased
performance variation and reduced reproducibility. There is a lack of understanding in the …
performance variation and reduced reproducibility. There is a lack of understanding in the …
Evaluating quality of service traffic classes on the megafly network
An emerging trend in High Performance Computing (HPC) systems that use hierarchical
topologies (such as dragonfly) is that the applications are increasingly exhibiting high run-to …
topologies (such as dragonfly) is that the applications are increasingly exhibiting high run-to …
Overcoming hadoop scaling limitations through distributed task execution
Data driven programming models like MapReduce have gained the popularity in large-scale
data processing. Although great efforts through the Hadoop implementation and framework …
data processing. Although great efforts through the Hadoop implementation and framework …
Trade-off study of localizing communication and balancing network traffic on a dragonfly system
Dragonfly networks are being widely adopted in high-performance computing systems. On
these networks, however, interference caused by resource sharing can lead to significant …
these networks, however, interference caused by resource sharing can lead to significant …
A convergence of key‐value storage systems from clouds to supercomputers
This paper presents a convergence of distributed key‐value storage systems in clouds and
supercomputers. It specifically presents ZHT, a zero‐hop distributed key‐value store system …
supercomputers. It specifically presents ZHT, a zero‐hop distributed key‐value store system …
Joint effects of application communication pattern, job placement and network routing on fat-tree systems
Among the high-radix and low-diameter networks, fat-tree topology is commonly used in
high-performance computing (HPC) and datacenter systems. Resource and job …
high-performance computing (HPC) and datacenter systems. Resource and job …
An analysis of long-tailed network latency distribution and background traffic on dragonfly+
M Salimi Beni, B Cosenza - International Symposium on Benchmarking …, 2022 - Springer
Modern computing systems are highly affected by large performance variability, resulting in
a long tail in the distribution of the network latency. For communication-intensive …
a long tail in the distribution of the network latency. For communication-intensive …