اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدRevelio: ML-Generated Debugging Queries for Distributed Systems
70
0
0.0
(
0
)
تأليف
Pradeep Dogga
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
A major difficulty in debugging distributed systems lies in manually determining which of the many available debugging tools to use and how to query its logs. Our own study of a production debugging workflow confirms the magnitude of this burden. This paper explores whether a machine-learning model can assist developers in distributed systems debugging. We present Revelio, a debugging assistant which takes user reports and system logs as input, and outputs debugging queries that developers can use to find a bugs root cause. The key challenges lie in (1) combining inputs of different types (e.g., natural language reports and quantitative logs) and (2) generalizing to unseen faults. Revelio addresses these by employing deep neural networks to uniformly embed diverse input sources and potential queries into a high-dimensional vector space. In addition, it exploits observations from production systems to factorize query generation into two computationally and statistically simpler learning tasks. To evaluate Revelio, we built a testbed with multiple distributed applications and debugging tools. By injecting faults and training on logs and reports from 800 Mechanical Turkers, we show that Revelio includes the most helpful query in its predicted list of top-3 relevant queries 96% of the time. Our developer study confirms the utility of Revelio.
قيم البحث
اقرأ أيضاً
Model parameter synchronization across GPUs introduces high overheads for data-parallel training at scale. Existing parameter synchronization protocols cannot effectively leverage available network resources in the face of ever increasing hardware he
terogeneity. To address this, we propose Blink, a collective communication library that dynamically generates optimal communication primitives by packing spanning trees. We propose techniques to minimize the number of trees generated and extend Blink to leverage heterogeneous communication channels for faster data transfers. Evaluations show that compared to the state-of-the-art (NCCL), Blink can achieve up to 8x faster model synchronization, and reduce end-to-end training time for image classification tasks by up to 40%.
We present Kaleidoscope an innovative system that supports live forensics for application performance problems caused by either individual component failures or resource contention issues in large-scale distributed storage systems. The design of Kale
idoscope is driven by our study of I/O failures observed in a peta-scale storage system anonymized as PetaStore. Kaleidoscope is built on three key features: 1) using temporal and spatial differential observability for end-to-end performance monitoring of I/O requests, 2) modeling the health of storage components as a stochastic process using domain-guided functions that accounts for path redundancy and uncertainty in measurements, and, 3) observing differences in reliability and performance metrics between similar types of healthy and unhealthy components to attribute the most likely root causes. We deployed Kaleidoscope on PetaStore and our evaluation shows that Kaleidoscope can run live forensics at 5-minute intervals and pinpoint the root causes of 95.8% of real-world performance issues, with negligible monitoring overhead.
Summary: More sophisticated models are needed to address problems in bioscience, synthetic biology, and precision medicine. To help facilitate the collaboration needed for such models, the community developed the Simulation Experiment Description Mar
kup Language (SED-ML), a common format for describing simulations. However, the utility of SED-ML has been hampered by limited support for SED-ML among modeling software tools and by different interpretations of SED-ML among the tools that support the format. To help modelers debug their simulations and to push the community to use SED-ML consistently, we developed a tool for validating SED-ML files. We have used the validator to correct the official SED-ML example files. We plan to use the validator to correct the files in the BioModels database so that they can be simulated. We anticipate that the validator will be a valuable tool for developing more predictive simulations and that the validator will help increase the adoption and interoperability of SED-ML. Availability: The validator is freely available as a webform, HTTP API, command-line program, and Python package at https://run.biosimulations.org/utils/validate and https://pypi.org/project/biosimulators-utils. The validator is also embedded into interfaces to 11 simulation tools. The source code is openly available as described in the Supplementary data. Contact: [email protected]
Trade-offs between accuracy and efficiency are found in multiple non-computing domains, such as law and public health, which have developed rules and heuristics to guide how to balance the two in conditions of uncertainty. While accuracy-efficiency t
rade-offs are also commonly acknowledged in some areas of computer science, their policy implications remain poorly examined. Drawing on risk assessment practices in the US, we argue that, since examining accuracy-efficiency trade-offs has been useful for guiding governance in other domains, explicitly framing such trade-offs in computing is similarly useful for the governance of computer systems. Our discussion focuses on real-time distributed ML systems; understanding the policy implications in this area is particularly urgent because such systems, which include autonomous vehicles, tend to be high-stakes and safety-critical. We describe how the trade-off takes shape for these systems, highlight gaps between existing US risk assessment standards and what these systems require in order to be properly assessed, and make specific calls to action to facilitate accountability when hypothetical risks become realized as accidents in the real world. We close by discussing how such accountability mechanisms encourage more just, transparent governance aligned with public values.
We present GSPMD, an automatic, compiler-based parallelization system for common machine learning computation graphs. It allows users to write programs in the same way as for a single device, then give hints through a few annotations on how to distri
bute tensors, based on which GSPMD will parallelize the computation. Its representation of partitioning is simple yet general, allowing it to express different or mixed paradigms of parallelism on a wide variety of models. GSPMD infers the partitioning for every operator in the graph based on limited user annotations, making it convenient to scale up existing single-device programs. It solves several technical challenges for production usage, such as static shape constraints, uneven partitioning, exchange of halo data, and nested operator partitioning. These techniques allow GSPMD to achieve 50% to 62% compute utilization on 128 to 2048 Cloud TPUv3 cores for models with up to one trillion parameters. GSPMD produces a single program for all devices, which adjusts its behavior based on a run-time partition ID, and uses collective operators for cross-device communication. This property allows the system itself to be scalable: the compilation time stays constant with increasing number of devices.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
