Handling faults is a growing concern in HPC. In future exascale systems, it is projected that silent undetected errors will occur several times a day, increasing the occurrence of corrupted results. In this article, we propose SEDAR, which is a methodology that improves system reliability against transient faults when running parallel message-passing applications. Our approach, based on process replication for detection, combined with different levels of checkpointing for automatic recovery, has the goal of helping users of scientific applications to obtain executions with correct results. SEDAR is structured in three levels: (1) only detection and safe-stop with notification; (2) recovery based on multiple system-level checkpoints; and (3) recovery based on a single valid user-level checkpoint. As each of these variants supplies a particular coverage but involves limitations and implementation costs, SEDAR can be adapted to the needs of the system. In this work, a description of the methodology is presented and the temporal behavior of employing each SEDAR strategy is mathematically described, both in the absence and presence of faults. A model that considers all the fault scenarios on a test application is introduced to show the validity of the detection and recovery mechanisms. An overhead evaluation of each variant is performed with applications involving different communication patterns; this is also used to extract guidelines about when it is beneficial to employ each SEDAR protection level. As a result, we show its efficacy and viability to tolerate transient faults in target HPC environments.
Scientific workflows have been predominantly used for complex and large scale data analysis and scientific computation/automation and the need for robust workflow scheduling techniques has grown considerably. But, most of the existing workflow scheduling algorithms do not provide the required reliability and robustness. In this paper, a new fault tolerant workflow scheduling algorithm that learns replication heuristics in an unsupervised manner has been proposed. Furthermore, the use of light weight synchronized checkpointing enables efficient resubmission of failed tasks and ensures workflow …
While checkpointing is typically combined with a restart of the whole application, localized recovery permits all but the affected processes to continue. In task-based cluster programming, for instance, the application can then be finished on the intact nodes, and the lost tasks be reassigned. This extended abstract suggests to adapt a checkpointing and localized recovery technique that has originally been developed for independent tasks to nested fork-join programs. We consider a Cilk-like work stealing scheme with work-first policy in a distributed memory setting, and describe the required a…
ZOCheck: CPU-Shadow Checkpointing for Zeroth-Order LLM Fine-Tuning
University of Delaware（特拉华大学）；RIKEN Center for Computational Science（理化学研究所计算科学中心）；Pacific Northwest National Laboratory（太平洋西北国家实验室）；University of Washington（华盛顿大学）
This paper describes two-fold approach towards utilizing Triple Modular Redundancy (TMR) in Wireless Adhoc Network (AdocNet). A distributed checkpointing and recovery protocol is proposed. The protocol eliminates useless checkpoints and helps in selecting only dependent processes in the concerned checkpointing interval, to recover. A process starts recovery from its last checkpoint only if it finds that it is dependent (directly or indirectly) on the faulty process. The recovery protocol also prevents the occurrence of missing or orphan messages. In AdocNet, a set of three nodes (connected to …
FP-Growth algorithm is a Frequent Pattern Min- ing (FPM) algorithm that has been extensively used to study correlations and patterns in large scale datasets. While several researchers have designed distributed memory FP-Growth algorithms, it is pivotal to consider fault tolerant FP-Growth, which can address the increasing fault rates in large scale systems. In this work, we propose a novel parallel, algorithm-level fault-tolerant FP-Growth algorithm. We leverage algorithmic properties and MPI advanced features to guarantee an O(1) space complexity, achieved by using the dataset memory space it…
The success of Google's Pregel framework in distributed graph processing has inspired a surging interest in developing Pregel-like platforms featuring a user-friendly "think like a vertex" programming model. Existing Pregel-like systems support a fault tolerance mechanism called checkpointing, which periodically saves computation states as checkpoints to HDFS, so that when a failure happens, computation rolls back to the latest checkpoint. However, a checkpoint in existing systems stores a huge amount of data, including vertex states, edges, and messages sent by vertices, which significantly d…
HyCoR is a fully-operational fault tolerance mechanism for multiprocessor workloads, based on container replication, using a hybrid of checkpointing and replay. HyCoR derives from two insights regarding replication mechanisms: 1) deterministic replay can overcome a key disadvantage of checkpointing alone -- unacceptably long delays of outputs to clients, and 2) checkpointing can overcome a key disadvantage of active replication with deterministic replay alone -- vulnerability to even rare replay failures due to an untracked nondeterministic events. With HyCoR, the primary sends periodic checkp…
Scaling supercomputers comes with an increase in failure rates due to the increasing number of hardware components. In standard practice, applications are made resilient through checkpointing data and restarting execution after a failure occurs to resume from the latest check-point. However, re-deploying an application incurs overhead by tearing down and re-instating execution, and possibly limiting checkpointing retrieval from slow permanent storage. In this paper we present Reinit++, a new design and implementation of the Reinit approach for global-restart recovery, which avoids application …
In this paper, we revisit traditional checkpointing and rollback recovery strategies, with a focus on silent data corruption errors. Contrarily to fail-stop failures, such latent errors cannot be detected immediately, and a mechanism to detect them must be provided. We consider two models: (i) errors are detected after some delays following a probability distribution (typically, an Exponential distribution); (ii) errors are detected through some verification mechanism. In both cases, we compute the optimal period in order to minimize the waste, i.e., the fraction of time where nodes do not per…
A mobile computing system is a distributed system in which at least one of the processes is mobile. They are constrained by lack of stable storage, low network bandwidth, mobility, frequent disconnection and limited battery life. Checkpointing is one of the commonly used techniques to provide fault tolerance in mobile computing environment. In order to suit the mobile environment a distance based recovery scheme is proposed which is based on checkpointing and message logging. After the system recovers from failures, only the failed processes rollback and restart from their respective recent ch…
We present a new model for rollback recovery in distributed dataflow systems. We explain existing rollback schemes by assigning a logical time to each event such as a message delivery. If some processors fail during an execution, the system rolls back by selecting a set of logical times for each processor. The effect of events at times within the set is retained or restored from saved state, while the effect of other events is undone and re-executed. We show that, by adopting different logical time "domains" at different processors, an application can adopt appropriate checkpointing schemes fo…
Distributed Stream Processing systems are becoming an increasingly essential part of Big Data processing platforms as users grow ever more reliant on their ability to provide fast access to new results. As such, making timely decisions based on these results is dependent on a system's ability to tolerate failure. Typically, these systems achieve fault tolerance and the ability to recover automatically from partial failures by implementing checkpoint and rollback recovery. However, owing to the statistical probability of partial failures occurring in these distributed environments and the varia…
In order to efficiently use the future generations of supercomputers, fault tolerance and power consumption are two of the prime challenges anticipated by the High Performance Computing (HPC) community. Checkpoint/Restart (CR) has been and still is the most widely used technique to deal with hard failures. Application-level CR is the most effective CR technique in terms of overhead efficiency but it takes a lot of implementation effort. This work presents the implementation of our C++ based library CRAFT (Checkpoint-Restart and Automatic Fault Tolerance), which serves two purposes. First, it p…
Fault-tolerant distributed applications require mechanisms to recover data lost via a process failure. On modern cluster systems it is typically impractical to request replacement resources after such a failure. Therefore, applications have to continue working with the remaining resources. This requires redistributing the workload and that the non-failed processes reload data. We present an algorithmic framework and its C++ library implementation ReStore for MPI programs that enables recovery of data after process failures. By storing all required data in memory via an appropriate data distrib…
Efficient utilization of today's high-performance computing (HPC) systems with complex hardware and software components requires that the HPC applications are designed to tolerate process failures at runtime. With low mean time to failure (MTTF) of current and future HPC systems, long running simulations on these systems require capabilities for gracefully handling process failures by the applications themselves. In this paper, we explore the use of fault tolerance extensions to Message Passing Interface (MPI) called user-level failure mitigation (ULFM) for handling process failures without th…
Application partitioning and code offloading are being researched extensively during the past few years. Several frameworks for code offloading have been proposed. However, fewer works attempted to address issues occurred with its implementation in pervasive environments such as frequent network disconnection due to high mobility of users. Thus, in this paper, we proposed a fault tolerant algorithm that helps in consolidating the efficiency and robustness of application partitioning and offloading frameworks. To permit the usage of different fault tolerant policies such as replication and chec…
In this paper we demonstrate how stateful Byzantine Fault Tolerant services may be hosted on a Chord ring. The strategy presented is fourfold: firstly a replication scheme that dissociates the maintenance of replicated service state from ring recovery is developed. Secondly, clients of the ring based services are made replication aware. Thirdly, a consensus protocol is introduced that supports the serialization of updates. Finally Byzantine fault tolerant replication protocols are developed that ensure the integrity of service data hosted on the ring.
MPI has been ubiquitously deployed in flagship HPC systems aiming to accelerate distributed scientific applications running on tens of hundreds of processes and compute nodes. Maintaining the correctness and integrity of MPI application execution is critical, especially for safety-critical scientific applications. Therefore, a collection of effective MPI fault tolerance techniques have been proposed to enable MPI application execution to efficiently resume from system failures. However, there is no structured way to study and compare different MPI fault tolerance designs, so to guide the selec…
HPC systems are a critical resource for scientific research. The increased demand for computational power and memory ushers in the exascale era, in which supercomputers are designed to provide enormous computing power to meet these needs. These complex supercomputers consist of numerous compute nodes and are consequently expected to experience frequent faults and crashes. Mathematical solvers, in particular, iterative linear solvers are key building block in numerous large-scale scientific applications. Consequently, supporting the recovery of distributed solvers is necessary for scaling scien…
Fault tolerance overhead of high performance computing (HPC) applications is becoming critical to the efficient utilization of HPC systems at large scale. HPC applications typically tolerate fail-stop failures by checkpointing. Another promising method is in the algorithm level, called algorithmic recovery. These two methods can achieve high efficiency when the system scale is not very large, but will both lose their effectiveness when systems approach the scale of Exaflops, where the number of processors including in system is expected to achieve one million. This paper develops a new and eff…
Replication is a standard technique for fault tolerance in distributed systems modeled as deterministic finite state machines (DFSMs or machines). To correct f crash or f/2 Byzantine faults among n different machines, replication requires nf additional backup machines. We present a solution called fusion that requires just f additional backup machines. First, we build a framework for fault tolerance in DFSMs based on the notion of Hamming distances. We introduce the concept of an (f,m)-fusion, which is a set of m backup machines that can correct f crash faults or f/2 Byzantine faults among a g…
We describe the voting farm, a tool which implements a distributed software voting mechanism for a number of parallel message passing systems. The tool, developed in the framework of EFTOS (Embedded Fault-Tolerant Supercomputing), can be used in stand-alone mode or in conjunction with other EFTOS fault tolerance tools. In the former case, we describe how the mechanism can be exploited, e.g., to implement restoring organs ($N\!$-modular redundancy systems with $N\!$-replicated voters); in the latter case, we show how it is possible for the user to implement in an easy and effective way a number…
This work proposes the first strategy to make distributed training of neural networks resilient to computing errors, a problem that has remained unsolved despite being first posed in 1956 by von Neumann. He also speculated that the efficiency and reliability of the human brain is obtained by allowing for low power but error-prone components with redundancy for error-resilience. It is surprising that this problem remains open, even as massive artificial neural networks are being trained on increasingly low-cost and unreliable processing units. Our coding-theory-inspired strategy, "CodeNet," sol…
It is commonly agreed that highly parallel software on Exascale computers will suffer from many more runtime failures due to the decreasing trend in the mean time to failures (MTTF). Therefore, it is not surprising that a lot of research is going on in the area of fault tolerance and fault mitigation. Applications should survive a failure and/or be able to recover with minimal cost. MPI is not yet very mature in handling failures, the User-Level Failure Mitigation (ULFM) proposal being currently the most promising approach is still in its prototype phase. In our work we use GASPI, which is a r…
Towards Resiliency in Large Language Model Serving with KevlarFlow
Department of XXX, University of YYY, Location, Country（XXX系，YYY大学，地点，国家）；School of ZZZ, Institute of WWW, Location, Country（ZZZ学院，WWW研究所，地点，国家）；Department of Computer Science, Purdue University, West Lafayette, IN, United States（计算机科学系，普渡大学，西拉法叶，印第安纳，美国）
