Data centers are widely used for big data analytics, which often involve data-parallel jobs,
including query and web service. Meanwhile, cluster frameworks are rapidly developed for …

We present FEMU, a QEMU-based flash emulator for fostering future full-stack
software/hardware SSD research, with the following four" CASE" benefits. FEMU is cheap …

Fail-slow hardware is an under-studied failure mode. We present a study of 114 reports of
fail-slow hardware incidents, collected from large-scale cluster deployments in 14 …

Distributed Machine Learning (DML) techniques are widely used to accelerate the training of
large-scale machine learning models. However, during training iterations, gradients need to …

The newly-emerging''fail-slow''failures plague both software and hardware where the victim
components are still functioning yet with degraded performance. To address this problem …

MittOS provides operating system support to cut millisecond-level tail latencies for data-
parallel applications. In MittOS, we advocate a new principle that operating system should …

Distributed file systems often exhibit high tail latencies, especially in large-scale datacenters
and in the presence of competing (and possibly higher priority) workloads. This paper …

We address the problem of “fail-slow” fault, a fault where a hardware or software component
can still function (does not fail-stop) but in much lower performance than expected. To …

Duplication can be a powerful strategy for overcoming stragglers in cloud services, but is
often used conservatively because of the risk of overloading the system. We call for making …

ABSTRACT In the Node-Disjoint Paths problem (NDP), the input is an undirected n-vertex
graph G, and a collection {(s1, t1),...,(sk, tk)} of demand pairs. The goal is to route the largest …