Related papers: Shirakami: A Hybrid Concurrency Control Protocol f…
Blockchain networks are facing increasingly heterogeneous computational demands, and in response, protocol designers have started building specialized infrastructure to supply that demand. This paper introduces Resonance: a new kind of…
A reliable network is an absolute requirement for telecommunication networks at this time. This is in line with the growing number of users who need a reliable and uninterrupted connection to the server. Link Aggregation is one of the…
This paper emphasizes the critical role of interoperability in enabling efficient and secure communication for the fragmented distributed ledger ecosystem, particularly within on-chain finance. The purpose of this study is to streamline and…
In this paper, we investigate the outage performance of hybrid automatic repeat request with chase combining (HARQ-CC) assisted downlink non-orthogonal multiple access (NOMA) systems. A closed-form expression of the individual outage…
Synchronization is a fundamental phenomenon in complex systems, observed across a wide range of natural and engineered contexts. The Kuramoto model provides a foundational framework for understanding synchronization among coupled…
Training large language models (LLMs) with increasingly long and varying sequence lengths introduces severe load imbalance challenges in large-scale data-parallel training. Recent frameworks attempt to mitigate these issues through data…
SAKURAONE is a managed high performance computing (HPC) cluster developed and operated by the SAKURA Internet Research Center. It builds on the KOKARYOKU PHY bare metal GPU platform and is optimized for advanced workloads, including large…
Pipeline Parallelism (PP) serves as a crucial technique for training Large Language Models (LLMs), owing to its capability to alleviate memory pressure from model states with relatively low communication overhead. However, in long-context…
Distributed data processing systems have advanced through models that expose more and more opportunities for concurrency within a computation. The scheduling of these increasingly sophisticated models has become the bottleneck for improved…
Sharding is used to address the performance and scalability issues of the blockchain protocols, which divides the overall transaction processing costs among multiple clusters of nodes. Shards require less storage capacity and communication…
CXL-based Computational Memory (CCM) enables near-memory processing within expanded remote memory, presenting opportunities to address data movement costs associated with disaggregated memory systems and to accelerate overall performance.…
Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or…
Today's distributed network control planes support multiple routing protocols, filtering mechanisms, and route selection policies. These protocols operate at different layers, e.g. BGP operates at the EGP layer, OSPF at the IGP layer, and…
Recent networking research has identified that data-driven congestion control (CC) can be more efficient than traditional CC in TCP. Deep reinforcement learning (RL), in particular, has the potential to learn optimal network policies.…
Transactions can simplify distributed applications by hiding data distribution, concurrency, and failures from the application developer. Ideally the developer would see the abstraction of a single large machine that runs transactions…
To utilize the multi-core processors properly concurrent programming is needed. Concurrency control is the main challenge while designing a correct and efficient concurrent program. Software Transactional Memory Systems (STMs) provides ease…
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…
As cross-chain technology continues to advance, the scale of cross-chain transactions is experiencing significant expansion. To improve scalability, researchers have turned to the study of cross-chain state channels. However, most of the…
The training efficiency and scalability of language models on massive clusters currently remain a critical bottleneck. Mainstream approaches like ND parallelism are often cumbersome and complex, while flexible alternatives such as the Zero…
Large language model (LLM) agents demonstrate strong performance in short-text contexts but often underperform in extended dialogues due to inefficient memory management. Existing approaches face a fundamental trade-off between efficiency…