Related papers: Adapting the DMTCP Plugin Model for Checkpointing …
The accuracy of large language models (LLMs) improves with increasing model size, but increasing model complexity also poses significant challenges to training stability. Periodic checkpointing is a key mechanism for fault recovery and is…
This report investigates the relationship between software refactoring and behavior preservation. Existing behavior preservation analyses often lack comprehensive insights into refactoring rejections and do not provide actionable solutions.…
Various efforts have been devoted to developing stabilizing distributed Model Predictive Control (MPC) schemes for tracking piecewise constant references. In these schemes, terminal sets are usually computed offline and used in the MPC…
In this paper we address the issue of change after deployment in safety-critical embedded system applications. Our goal is to substitute lab-based verification with in-field formal analysis to determine whether an update may be safely…
Multipath TCP (MPTCP) can exploit multiple available interfaces at the end devices by establishing concurrent multiple connections between source and destination. MPTCP is a drop-in replacement for TCP and this makes it an attractive choice…
Grid computing is a collection of computer resources that are gathered together from various areas to give computational resources such as storage, data or application services. This is to permit clients to access this huge measure of…
Stream processing in the last decade has seen broad adoption in both commercial and research settings. One key element for this success is the ability of modern stream processors to handle failures while ensuring exactly-once processing…
Traditionally, when the code of a database management system (DBMS) needs to be updated, the system is restarted and database clients suffer downtime, or the provider instantiates hot-standby instances and rolls over the workload. We…
As LLMs and foundation models scale, checkpoint/restore has become a critical pattern for training and inference. With 3D parallelism (tensor, pipeline, data), checkpointing involves many processes, each managing numerous tensors of varying…
Conformance checking is a key process mining task for comparing the expected behavior captured in a process model and the actual behavior recorded in a log. While this problem has been extensively studied for pure control-flow processes,…
Rollback recovery strategies are well-known in concurrent and distributed systems. In this context, recovering from unexpected failures is even more relevant given the non-deterministic nature of execution, which means that it is…
Over the past years, TCP has gone through numerous updates to provide performance enhancement under diverse network conditions. However, with respect to losses, little can be achieved with legacy TCP detection and recovery mechanisms. Both…
Multiple clustering has gathered significant attention in recent years due to its potential to reveal multiple hidden structures of the data from different perspectives. Most of multiple clustering methods first derive feature…
As large language models scale, training them requires thousands of GPUs over extended durations--making frequent failures an inevitable reality. While checkpointing remains the primary fault-tolerance mechanism, existing methods fall short…
This work presents transparent checkpointing of OpenGL applications, refining the split-process technique[1] for application in GPU-based 3D graphics. The split-process technique was earlier applied to checkpointing MPI and CUDA programs,…
We introduce Torch-Points3D, an open-source framework designed to facilitate the use of deep networks on3D data. Its modular design, efficient implementation, and user-friendly interfaces make it a relevant tool for research and…
We formalize the problem of trading-off DNN training time and memory requirements as the tensor rematerialization optimization problem, a generalization of prior checkpointing strategies. We introduce Checkmate, a system that solves for…
Many machine learning algorithms and classifiers are available only via API queries as a ``black-box'' -- that is, the downstream user has no ability to change, re-train, or fine-tune the model on a particular target distribution. Indeed,…
Event-driven multi-threaded programming is an important idiom for structuring concurrent computations. Stateless Model Checking (SMC) is an effective verification technique for multi-threaded programs, especially when coupled with Dynamic…
With the increase in compute nodes in large compute platforms, a proportional increase in node failures will follow. Many application-based checkpoint/restart (C/R) techniques have been proposed for MPI applications to target the reduced…