Given the highly empirical nature of research in cloud computing, networked systems, and
related fields, testbeds play an important role in the research ecosystem. In this paper, we …

Large-scale deep learning requires huge computational resources to train a multi-layer
neural network. Recent systems propose using 100s to 1000s of machines to train networks …

TetriSched is a scheduler that works in tandem with a calendaring reservation system to
continuously re-evaluate the immediate-term scheduling plan for all pending jobs (including …

The growing size of modern storage systems is expected to exceed billions of objects,
making metadata scalability critical to overall performance. Many existing distributed file …

Many modern machine learning (ML) algorithms are iterative, converging on a final solution
via many iterations over the input data. This paper explores approaches to exploiting these …

FlexRR provides a scalable, efficient solution to the straggler problem for iterative machine
learning (ML). The frequent (eg, per iteration) barriers used in traditional BSP-based …

Distributed machine learning has typically been approached from a data parallel
perspective, where big data are partitioned to multiple workers and an algorithm is executed …

The last five years have seen a rise of implementationlevel distributed system model
checkers (dmck) for verifying the reliability of real distributed systems. Existing dmcks …

At the core of Machine Learning (ML) analytics is often an expert-suggested model, whose
parameters are refined by iteratively processing a training dataset until convergence. The …

Load balancing techniques (eg work stealing) are important to obtain the best performance
for distributed task scheduling systems that have multiple schedulers making scheduling …