Michael P. Kasick

Learn More
We focus on automatically diagnosing different performance problems in parallel file systems by identifying, gathering and analyzing OS-level, black-box performance metrics on every node in the cluster. Our peercomparison diagnosis approach compares the statistical attributes of these metrics across I/O servers, to identify the faulty node. We develop a(More)
We present a behavior-based problem-diagnosis approach for PVFS that analyzes a novel source of instrumentation—CPU instruction-pointer samples and function-call traces—to localize the faulty server and to enable root-cause analysis of the resource at fault. We validate our approach by injecting realistic storage and network problems into three different(More)
In the emerging cloud computing era, enterprise data centers host a plethora of web services and applications, including those for e-Commerce, distributed multimedia, and social networks, which jointly, serve many aspects of our daily lives and business. For such applications, lack of availability, reliability, or responsiveness can lead to extensive(More)
Performance problems account for a significant percentage of documented failures in large-scale distributed systems, such as Hadoop. Localizing the source of these performance problems can be frustrating due to the overwhelming amount of monitoring information available. We automate problem localization using ASDF an online diagnostic framework that(More)
We present a syscall-based approach to automatically diagnose performance problems, server-to-client propagated errors, and server crash/hang problems in PVFS. Our approach compares the statistical and semantic attributes of syscalls across PVFS servers in order to diagnose the culprit server, under these problems, for different file-system benchmarks—dd,(More)
In the large-scale Emulab distributed system, the many failure reports make skilled operator time a scarce and costly resource, as shown by statistics on failure frequency and root cause. We describe the lessons learned with error reporting in Emulab, along with the design, initial implementation, and results of a new local erroranalysis approach that is(More)
Problem Diagnosis and debugging in concurrent environments such as the cloud and popular distributed systems frameworks has been a traditionally hard problem. We explore an evaluation of a novel way of debugging distributed systems frameworks by using system calls. We focus on Google's MapReduce framework, which enables distributed, data-intensive, parallel(More)
Localizing performance problems (or fingerpointing) is essential for distributed systems such as Hadoop that support long-running, parallelized, data-intensive computations over a large cluster of nodes. Manual fingerpointing does not scale in such environments because of the number of nodes and the number of performance metrics to be analyzed on each node.(More)
Localizing performance problems (or fingerpointing) is essential for distributed systems such as Hadoop that support long-running, parallelized, data-intensive computations over a large cluster of nodes. Manual fingerpointing does not scale in such environments because of the number of nodes and the number of performance metrics to be analyzed on each node.(More)