arXiv subjects

Eitan Zahavi

Publications and source records attributed to Eitan Zahavi.

5 recordsLinked to original sources

arXiv2026

Quantum Circuit Cutting: Complexity and Optimization

The current noisy intermediate-scale quantum (NISQ) era is characterized by substantial errors and noise, which limit the practical feasibility of deep, many-qubit circuits. To address these constraints, quantum circuit cutting has emerged as a promising tool. Recently, there has been significant research on methods for performing such cutting effectively. In this work, the duality between quantum circuits and classical graphs - specifically, directed acyclic graphs (dags) - is leveraged to analyze the complexity of finding an optimal circuit-cutting configuration that minimizes the number of cuts. After developing a rigorous graph-theoretic framework, the complexity of identifying cut locations that partition a given quantum circuit into smaller fragments is characterized. The corresponding graph-combinatorial task is then defined, and the resulting partition problem is shown to be NP-complete. Furthermore, even a simplified version of the problem, restricted to circuits composed only of one- and two-qubit gates, is shown to be NP-complete. Finally, based on these constraints, an algorithm grounded in satisfiability modulo theories (SMT) is proposed to find optimal cuts when the number of qubits per partition is bounded. This work therefore provides a complexity-theoretic characterization of cut placement and a practical solver for bounded-size decompositions.

quant-ph↗

arXiv2025

Leveraging InfiniBand Controller to Configure Deadlock-Free Routing Engines for Dragonflies

The Dragonfly topology is currently one of the most popular network topologies in high-performance parallel systems. The interconnection networks of many of these systems are built from components based on the InfiniBand specification. However, due to some constraints in this specification, the available versions of the InfiniBand network controller (OpenSM) do not include routing engines based on some popular deadlock-free routing algorithms proposed theoretically for Dragonflies, such as the one proposed by Kim and Dally based on Virtual-Channel shifting. In this paper we propose a straightforward method to integrate this routing algorithm in OpenSM as a routing engine, explaining in detail the configuration required to support it. We also provide experiment results, obtained both from a real InfiniBand-based cluster and from simulation, to validate the new routing engine and to compare its performance and requirements against other routing engines currently available in OpenSM.

cs.NI↗

arXiv2024

Implementing Reinforcement Learning Datacenter Congestion Control in NVIDIA NICs

As communication protocols evolve, datacenter network utilization increases. As a result, congestion is more frequent, causing higher latency and packet loss. Combined with the increasing complexity of workloads, manual design of congestion control (CC) algorithms becomes extremely difficult. This calls for the development of AI approaches to replace the human effort. Unfortunately, it is currently not possible to deploy AI models on network devices due to their limited computational capabilities. Here, we offer a solution to this problem by building a computationally-light solution based on a recent reinforcement learning CC algorithm [arXiv:2207.02295]. We reduce the inference time of RL-CC by x500 by distilling its complex neural network into decision trees. This transformation enables real-time inference within the $μ$-sec decision-time requirement, with a negligible effect on quality. We deploy the transformed policy on NVIDIA NICs in a live cluster. Compared to popular CC algorithms used in production, RL-CC is the only method that performs well on all benchmarks tested over a large range of number of flows. It balances multiple metrics simultaneously: bandwidth, latency, and packet drops. These results suggest that data-driven methods for CC are feasible, challenging the prior belief that handcrafted heuristics are necessary to achieve optimal performance.

cs.NI↗

arXiv2018

Revisiting Network Support for RDMA

The advent of RoCE (RDMA over Converged Ethernet) has led to a significant increase in the use of RDMA in datacenter networks. To achieve good performance, RoCE requires a lossless network which is in turn achieved by enabling Priority Flow Control (PFC) within the network. However, PFC brings with it a host of problems such as head-of-the-line blocking, congestion spreading, and occasional deadlocks. Rather than seek to fix these issues, we instead ask: is PFC fundamentally required to support RDMA over Ethernet? We show that the need for PFC is an artifact of current RoCE NIC designs rather than a fundamental requirement. We propose an improved RoCE NIC (IRN) design that makes a few simple changes to the RoCE NIC for better handling of packet losses. We show that IRN (without PFC) outperforms RoCE (with PFC) by 6-83% for typical network scenarios. Thus not only does IRN eliminate the need for PFC, it improves performance in the process! We further show that the changes that IRN introduces can be implemented with modest overheads of about 3-10% to NIC resources. Based on our results, we argue that research and industry should rethink the current trajectory of network support for RDMA.

cs.NI↗

arXiv2016

Links as a Service (LaaS): Feeling Alone in the Shared Cloud

The most demanding tenants of shared clouds require complete isolation from their neighbors, in order to guarantee that their application performance is not affected by other tenants. Unfortunately, while shared clouds can offer an option whereby tenants obtain dedicated servers, they do not offer any network provisioning service, which would shield these tenants from network interference. In this paper, we introduce Links as a Service, a new abstraction for cloud service that provides physical isolation of network links. Each tenant gets an exclusive set of links forming a virtual fat tree, and is guaranteed to receive the exact same bandwidth and delay as if it were alone in the shared cloud. Under simple assumptions, we derive theoretical conditions for enabling LaaS without capacity over-provisioning in fat-trees. New tenants are only admitted in the network when they can be allocated hosts and links that maintain these conditions. Using experiments on real clusters as well as simulations with real-life tenant sizes, we show that LaaS completely avoids the performance degradation caused by traffic from concurrent tenants on shared links. Compared to mere host isolation, LaaS can improve the application performance by up to 200%, at the cost of a 10% reduction in the cloud utilization.

cs.DC↗