arXiv subjects

Tal Mizrahi

Publications and source records attributed to Tal Mizrahi.

11 recordsLinked to original sources

arXiv2025

Routing for Large ML Models

Training large language models (LLMs), and other large machine learning models, involves repeated communication of large volumes of data across a data center network. The communication patterns induced by these training process exhibit high regularity and persistence, giving rise to significant opportunities for optimizing the manner in which flows are routed across the network. We present an algorithmic framework for \textit{quantifying} network-wide efficiency in the context of training LLMs (and other large-scale ML models), and for periodically \textit{optimizing} routing with respect to this global metric.

cs.NI↗

arXiv2024

The Observer Effect in Computer Networks

Network measurement involves an inherent tradeoff between accuracy and overhead; higher accuracy typically comes at the expense of greater measurement overhead (measurement frequency, number of probe packets, etc.). Capturing the "right" balance between these two desiderata - high accuracy and low overhead - is a key challenge. However, the manner in which accuracy and overhead are traded off is specific to the measurement method, rendering apples-to-apples comparisons difficult. To address this, we put forth a novel analytical framework for quantifying the accuracy-overhead tradeoff for network measurements. Our framework, inspired by the observer effect in modern physics, introduces the notion of a network observer factor, which formally captures the relation between measurement accuracy and overhead. Using our "network observer framework", measurement methods for the same task can be characterized in terms of their network observer factors, allowing for apples-to-apples comparisons. We illustrate the usefulness of our approach by showing how it can be applied to various application domains and validate its conclusions through experimental evaluation.

cs.NI↗

arXiv2022

Using Internet Measurements to Map the 2022 Ukrainian Refugee Crisis

The conflict in Ukraine, starting in February 2022, began the largest refugee crisis in decades, with millions of Ukrainian refugees crossing the border to neighboring countries and millions of others forced to move within the country. In this paper we present an insight into how Internet measurements can be used to analyze the refugee crisis. Based on preliminary data from the first two months of the war we analyze how measurement data indicates the trends in the flow of refugees from Ukraine to its neighboring countries, and onward to other countries. We believe that these insights can greatly contribute to the ongoing international effort to map the flow of refugees in order to aid and protect them.

cs.NI↗

arXiv2022

Internet Performance in the 2022 Conflict in Ukraine: An Asymmetric Analysis

On 24 February 2022 Russia invaded Ukraine, starting one of the largest military conflicts in Europe in recent years. In this paper we present preliminary findings about the impact of the conflict on the Internet performance in Ukraine and in Russia, introducing an ironically asymmetric picture: the Internet performance in Ukraine has significantly degraded, while the performance in Russia has improved.

cs.NI↗

arXiv2019

Time-Multiplexed Parsing in Marking-based Network Telemetry

Network telemetry is a key capability for managing the health and efficiency of a large-scale network. Alternate Marking Performance Measurement (AM-PM) is a recently introduced approach that accurately measures the packet loss and delay in a network using a small overhead of one or two bits per data packet. This paper introduces a novel time-multiplexed parsing approach that enables a practical and accurate implementation of AM-PM in network devices, while requiring just a single bit per packet. Experimental results are presented, based on a hardware implementation, and a software P4-based implementation.

cs.NI↗

arXiv2019

Timing in Software-Defined and Centrally-Managed Networks

The work described in this paper explores the use of time and synchronized clocks in centrally-managed and Software Defined Networks (SDNs). One of the main goals of this work is to analyze use cases in which explicit use of time is beneficial. Both theoretical and practical aspects of timed coordination and synchronized clocks in centralized environments are analyzed. Some of the products of this work are already incorporated in the OpenFlow specification, and open source prototypes of the main components are publicly available.

cs.NI↗

arXiv2017

FM-Delta: Fault Management Packet Compression

Fault Management (FM) is a cardinal feature in communication networks. One of the most common FM approaches is to use periodic keepalive messages. Hence, switches and routers are required to transmit a large number of FM messages periodically, requiring a hardware-based packet generator that periodically transmits a set of messages that are stored in an expensive on-chip memory. With the rapid growth of carrier networks, and as 5G technologies emerge, the number of users and the traffic rates are expected to significantly increase over the next few years. Consequently, we expect the on-chip memories used for FM to become a costly component in switch and router chips. We introduce a novel approach in which FM messages are stored in compressed form in the on-chip memory, allowing to significantly reduce the memory size. We present FM-Delta, a simple hardware-friendly delta encoding algorithm that allows FM messages to be compressed by a factor of 2.6. We show that this compression ratio is very close to the results of the zlib compression library, which requires much higher implementation complexity.

cs.NI↗

arXiv2016

The Case for Data Plane Timestamping in SDN

This paper presents the case for Data Plane Timestamping (DPT). We argue that in the unique environment of Software-Defined Networks (SDN), attaching a timestamp to the header of all packets is a powerful feature that can be leveraged by various diverse SDN applications. We analyze three key use cases that demonstrate the advantages of using DPT, and show that SDN applications can benefit even from using as little as one bit for the timestamp field.

cs.NI↗

arXiv2016

Time4: Time for SDN

With the rise of Software Defined Networks (SDN), there is growing interest in dynamic and centralized traffic engineering, where decisions about forwarding paths are taken dynamically from a network-wide perspective. Frequent path reconfiguration can significantly improve the network performance, but should be handled with care, so as to minimize disruptions that may occur during network updates. In this paper we introduce Time4, an approach that uses accurate time to coordinate network updates. Time4 is a powerful tool in softwarized environments, that can be used for various network update scenarios. Specifically, we characterize a set of update scenarios called flow swaps, for which Time4 is the optimal update approach, yielding less packet loss than existing update approaches. We define the lossless flow allocation problem, and formally show that in environments with frequent path allocation, scenarios that require simultaneous changes at multiple network devices are inevitable. We present the design, implementation, and evaluation of a Time4-enabled OpenFlow prototype. The prototype is publicly available as open source. Our work includes an extension to the OpenFlow protocol that has been adopted by the Open Networking Foundation (ONF), and is now included in OpenFlow 1.5. Our experimental results show the significant advantages of Time4 compared to other network update approaches, and demonstrate an SDN use case that is infeasible without Time4.

cs.NI↗

arXiv2016

OneClock to Rule Them All: Using Time in Networked Applications

This paper introduces OneClock, a generic approach for using time in networked applications. OneClock provides two basic time-triggered primitives: the ability to schedule an operation at a remote host or device, and the ability to receive feedback about the time at which an event occurred or an operation was executed at a remote host or device. We introduce a novel prediction-based scheduling approach that uses timing information collected at runtime to accurately schedule future operations. Our work includes an extension to the Network Configuration protocol (NETCONF), which enables OneClock in real-life systems. This extension has been published as an Internet Engineering Task Force (IETF) RFC, and a prototype of our NETCONF time extension is publicly available as open source. Experimental evaluation shows that our prediction-based approach allows accurate scheduling in diverse and heterogeneous environments, with various hardware capabilities and workloads. OneClock is a generic approach that can be applied to any managed device: sensors, actuators, Internet of Things (IoT) devices, routers, or toasters.

cs.NI↗

arXiv2015

Timed Consistent Network Updates

Network updates such as policy and routing changes occur frequently in Software Defined Networks (SDN). Updates should be performed consistently, preventing temporary disruptions, and should require as little overhead as possible. Scalability is increasingly becoming an essential requirement in SDN. In this paper we propose to use time-triggered network updates to achieve consistent updates. Our proposed solution requires lower overhead than existing update approaches, without compromising the consistency during the update. We demonstrate that accurate time enables far more scalable consistent updates in SDN than previously available. In addition, it provides the SDN programmer with fine-grained control over the tradeoff between consistency and scalability.

cs.NI↗