Related papers: Performance and Fault Tolerance in the StoreTorren…

In this paper we present and evaluate a parallel algorithm for solving a minimum spanning tree (MST) problem for supercomputers with distributed memory. The algorithm relies on the relaxation of the message processing order requirement for…

In this paper we are interested in bounding the number of instructions taken to process transactions. The main result is a multiversion transactional system that supports constant delay (extra instructions beyond running in isolation) for…

I describe here the performances of a parallel treecode with individual particle timesteps. The code is based on the Barnes-Hut algorithm and runs cosmological N-body simulations on parallel machines with a distributed memory architecture…

In the recent years it can be observed increasing popularity of parallel processing using multi-core processors, local clusters, GPU and others. Moreover, currently one of the main requirements the IT users is the reduction of maintaining…

In this paper, we present a novel and new file-based communication architecture using the local filesystem for large scale parallelization. This new approach eliminates the issues with filesystem overload and resource contention when using…

Stack Long Short-Term Memory (StackLSTM) is useful for various applications such as parsing and string-to-tree neural machine translation, but it is also known to be notoriously difficult to parallelize for GPU training due to the fact that…

The scaling of large language models (LLMs) is currently bottlenecked by the rigidity of distributed programming. While high-performance libraries like CuBLAS and NCCL provide optimized primitives, they lack the flexibility required for…

Peer-to-peer (P2P) locality has recently raised a lot of interest in the community. Indeed, whereas P2P content distribution enables financial savings for the content providers, it dramatically increases the traffic on inter-ISP links. To…

This work elaborates on a High performance computing (HPC) architecture based on Simple Linux Utility for Resource Management (SLURM) [1] for deploying heterogeneous Large Language Models (LLMs) into a scalable inference engine. Dynamic…

In this paper we introduce and describe the highly concurrent xDFS file transfer protocol and examine its cross-platform and cross-language implementation in native code for both Linux and Windows in 32 or 64-bit multi-core processor…

Diffusion Transformers (DiTs) have gained increasing adoption in high-quality image and video generation. As demand for higher-resolution images and longer videos increases, single-GPU inference becomes inefficient due to increased latency…

Parsing is essential for a wide range of use cases, such as stream processing, bulk loading, and in-situ querying of raw data. Yet, the compute-intense step often constitutes a major bottleneck in the data ingestion pipeline, since parsing…

In large-scale LLM pre-training systems with 100k+ GPUs, failures become the norm rather than the exception, and restart costs can dominate wall-clock training time. However, existing fault-tolerance mechanisms are largely unprepared for…

Practical Byzantine Fault Tolerance (PBFT) is a seminal state machine replication protocol that achieves a performance comparable to non-replicated systems in realistic environments. A reason for such high performance is the set of…

We present MiniTensor, an open source tensor operations library that focuses on minimalism, correctness, and performance. MiniTensor exposes a familiar PyTorch-like Python API while it executes performance critical code in a Rust engine.…

We propose a parallel algorithm for local, on the fly, model checking of a fragment of CTL that is well-suited for modern, multi-core architectures. This model-checking algorithm takes bene t from a parallel state space construction…

Volunteer computing has been known as an alternative solution to solve complex problems. It is acknowledged for its simplicity and its ability to work on multiple operating systems. Nonetheless, setting up a server for volunteer computing…

The microservice software architecture is more scalable and efficient than its monolithic predecessor. Despite its increasing adoption, microservices might expose security concerns and issues that are distinct from those associated with…

We investigate distributed memory parallel sorting algorithms that scale to the largest available machines and are robust with respect to input size and distribution of the input elements. The main outcome is that four sorting algorithms…

Storage systems have not kept the same technology improvement rate as computing systems. As applications produce more and more data, I/O becomes the limiting factor for increasing application performance. I/O congestion caused by concurrent…