Inder Monga

Lawrence Berkeley National Laboratory, ESnet, CTO

Followers

Following

Co-authors

Public Views

Uploads

Papers by Inder Monga

Quantum Networks For Open Science

arXiv (Cornell University), Mar 27, 2019

Download

QUANT-NET: A testbed for quantum networking research over deployed fiber

Proceedings of the 1st Workshop on Quantum Networks and Distributed Quantum Computing

QUANT-NET is a three-node quantum network research testbed funded by the U.S. Department of Energ... more QUANT-NET is a three-node quantum network research testbed funded by the U.S. Department of Energy (DOE). The goal is to establish this network between two sites, Lawrence Berkeley National Lab (LBNL) and University of California, Berkeley, connected with an entanglement swapping substrate over optical fiber and managed by a quantum network protocol stack. On top of this entanglement swapping substrate the research team will implement the most basic building blocks of distributed quantum computing and quantum repeater by teleporting a controlled-NOT gate between two far trapped-ion quantum computation nodes. This paper presents QUANT-NET, its design, key technologies, and progress. CCS Concepts • Networks → Network architectures; • Computer systems organization → Quantum computing; • Hardware → Quantum technologies.

Download

Quantum Networks for High Energy Physics

arXiv (Cornell University), Mar 31, 2022

Quantum networks of quantum objects promise to be exponentially more powerful than the objects co... more Quantum networks of quantum objects promise to be exponentially more powerful than the objects considered independently. To live up to this promise will require the development of error mitigation and correction strategies to preserve quantum information as it is initialized, stored, transported, utilized, and measured. The quantum information could be encoded in discrete variables such as qubits, in continuous variables, or anything in-between. Quantum computational networks promise to enable simulation of physical phenomena of interest to the HEP community. Quantum sensor networks promise new measurement capability to test for new physics and improve upon existing measurements of fundamental constants. Such networks could exist at multiple scales from the nano-scale to a global-scale quantum network. a

Download

Quantum Networks for High Energy Physics

Nicholas A. Peters,1 Muneer Alshowkan,1 Katherine J. Evans,2 Phil Evans,1 Marcel Demarteau,3 Trav... more Nicholas A. Peters,1 Muneer Alshowkan,1 Katherine J. Evans,2 Phil Evans,1 Marcel Demarteau,3 Travis Humble,1 Joseph M. Lukens,1 Raphael C. Pooser,1 Bing Qi,1 Nageswara S. Rao,4 and Brian Williams1 1Quantum Information Science Group, Computational Sciences and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee USA 37831 2Computational Sciences and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee USA 37831 3Physics Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee USA 37831 4Autonomous and Complex Systems Group, Computational Sciences and Engineering Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee USA 37831

Download

Quantum Networks For Open Science

arXiv: Quantum Physics, 2019

The United States Department of Energy convened the Quantum Networks for Open Science (QNOS) Work... more The United States Department of Energy convened the Quantum Networks for Open Science (QNOS) Workshop in September 2018. The workshop was primarily focused on quantum networks optimized for scientific applications with the expectation that the resulting quantum networks could be extended to lay the groundwork for a generalized network that will evolve into a quantum internet.

Download

Machine learning-based analysis of COVID-19 pandemic impact on US research networks

Computer Communication Review, Oct 24, 2021

This study explores how fallout from the changing public health policy around COVID-19 has change... more This study explores how fallout from the changing public health policy around COVID-19 has changed how researchers access and process their science experiments. Using a combination of techniques from statistical analysis and machine learning, we conduct a retrospective analysis of historical network data for a period around the stay-at-home orders that took place in March 2020. Our analysis takes data from the entire ESnet infrastructure to explore DOE high-performance computing (HPC) resources at OLCF, ALCF, and NERSC, as well as User sites such as PNNL and JLAB. We look at detecting and quantifying changes in site activity using a combination of t-Distributed Stochastic Neighbor Embedding (t-SNE) and decision tree analysis. Our findings bring insights into the working patterns and impact on data volume movements, particularly during late-night hours and weekends.

Download

INDIRA: ‘Application Intent’ network assistant to configure SDN-based High Performance Scientific Networks

Optical Fiber Communication Conference, 2017

We demonstrate INDIRA ((Intelligent Network Deployment Intent Renderer Application)), an interact... more

Effectiveness and predictability of in-network storage cache for Scientific Workflows

2023 International Conference on Computing, Networking and Communications (ICNC)

Large scientific collaborations often have multiple scientists accessing the same set of files wh... more Large scientific collaborations often have multiple scientists accessing the same set of files while doing different analyses, which create repeated accesses to the large amounts of shared data located far away. These data accesses have long latency due to distance and occupy the limited bandwidth available over the wide-area network. To reduce the wide-area network traffic and the data access latency, regional data storage caches have been installed as a new networking service. To study the effectiveness of such a cache system in scientific applications, we examine the Southern California Petabyte Scale Cache for a high-energy physics experiment. By examining about 3TB of operational logs, we show that this cache removed 67.6% of file requests from the wide-area network and reduced the traffic volume on wide-area network by 12.3TB (or 35.4%) an average day. The reduction in the traffic volume (35.4%) is less than the reduction in file counts (67.6%) because the larger files are less likely to be reused. Due to this difference in data access patterns, the cache system has implemented a policy to avoid evicting smaller files when processing larger files. We also build a machine learning model to study the predictability of the cache behavior. Tests show that this model is able to accurately predict the cache accesses, cache misses, and network throughput, making the model useful for future studies on resource provisioning and planning.

Download

Studying Scientific Data Lifecycle in On-demand Distributed Storage Caches

Fifth International Workshop on Systems and Network Telemetry and Analytics

The XRootD system is used to transfer, store, and cache large datasets from high-energy physics (... more The XRootD system is used to transfer, store, and cache large datasets from high-energy physics (HEP). In this study we focus on its capability as distributed on-demand storage cache. Through exploring a large set of daily log files between 2020 and 2021, we seek to understand the data access patterns that might inform future cache design. Our study begins with a set of summary statistics regarding file read operations, file lifetimes, and file transfers. We observe that the number of read operations on each file remains nearly constant, while the average size of a read operation grows over time. Furthermore, files tend to have a consistent length of time during which they remain open and are in use. Based on this comprehensive study of the cache access statistics, we developed a cache simulator to explore the behavior of caches of different sizes. Within a certain size range, we find that increasing the XRootD cache size improves the cache hit rate, yielding faster overall file access. In particular, we find that increase the cache size from 40TB to 56TB could increase the hit rate from 0.62 to 0.89, which is a significant increase in cache effectiveness for modest cost. CCS CONCEPTS • Information systems → Information lifecycle management; Hierarchical storage management; • Computing methodologies → Model development and analysis.

Download

Access Trends of In-network Cache for Scientific Data

Fifth International Workshop on Systems and Network Telemetry and Analytics

Scientific collaborations are increasingly relying on large volumes of data for their work and ma... more Scientific collaborations are increasingly relying on large volumes of data for their work and many of them employ tiered systems to replicate the data to their worldwide user communities. Each user in the community often selects a different subset of data for their analysis tasks; however, members of a research group often are working on related research topics that require similar data objects. Thus, there is a significant amount of data sharing possible. In this work, we study the access traces of a federated storage cache known as the Southern California Petabyte Scale Cache. By studying the access patterns and potential for network traffic reduction by this caching system, we aim to explore the predictability of the cache uses and the potential for a more general in-network data caching. Our study shows that this distributed storage cache is able to reduce the network traffic volume by a factor of 2.35 during a part of the study period. We further show that machine learning models could predict cache utilization with an accuracy of 0.88. This demonstrates that such cache usage is predictable, which could be useful for managing complex networking resources such as in-network caching. CCS CONCEPTS • Networks → Network performance analysis; • Computing methodologies → Distributed computing methodologies.

Download

Indira: intent-based networking and rendering for networks (Indira) v1

Intelligent Network Deployment Intent Renderer Application an interactive network assistant that ... more Intelligent Network Deployment Intent Renderer Application an interactive network assistant that will help us configure a data path between two scientific end-point groups (EPGs) to optimize the transfer of elephant data flows. This tool is able to talk to scientists gather network requirements and then convert to network programming code and automate the end-to-end provisioning.

HEPiX Network Functions Virtualisation Working Group Report

High Energy Physics (HEP) experiments rely on the networks as one of the critical parts of their ... more High Energy Physics (HEP) experiments rely on the networks as one of the critical parts of their infrastructure both within the participating laboratories and sites as well as globally to interconnect the sites, data centres and experiments instrumentation. Network virtualisation and programmable networks are two key enablers that facilitate agile, fast and more economical network infrastructures as well as service development, deployment and provisioning. Adoption of these technologies by the HEP sites and experiments will allow them to design more scalable and robust networks while decreasing the overall cost and improving the effectiveness of the resource usage. The primary challenge we face is ensuring that WLCG and its constituent collaborations will have the networking capabilities required to most effectively exploit LHC data for the lifetime of the LHC. Network virtualisation and programmable networks are nowadays quite common in the commercial clouds and telco deployments a...

Download

INDIRA: ‘Application Intent’ network assistant to configure SDN-based High Performance Scientific Networks

Optical Fiber Communication Conference, 2017

We demonstrate INDIRA ((Intelligent Network Deployment Intent Renderer Application)), an interact... more

ROBIN (RuciO/BIgData Express/SENSE) A Next-Generation High-Performance Data Service Platform

2020 IEEE/ACM Innovating the Network for Data-Intensive Science (INDIS), 2020

Increasingly, scientific research discovery depends on the analysis of extremely large amounts of... more Increasingly, scientific research discovery depends on the analysis of extremely large amounts of data. This data is gathered through very large scientific instruments, very large numbers of small devices, and mid-scale instruments. These devices generate exponentially increasing volumes of data. Such data is often obtained, analyzed, stored, visualized, and transferred at and among sites worldwide by large collaborations of science communities. Managing and moving extremely large volumes of data across today's networks is a special multidimensional challenge. Also, existing data management and movement services and tools often are inadequate to address all specific requirements. This paper describes the various components of this challenge and how those issues can be addressed by ROBIN, a unique comprehensive set of integrated services designed specifically for managing and moving extremely large amounts of data over long distances, e.g., thousands of miles around the globe. This paper also describes the results of initial experiments using that set of integrated services.

Download

Enhancing ESnet's OSCARS Path Computation Engine

GLOBECOM 2017 - 2017 IEEE Global Communications Conference, 2017

The Department of Energy (DOE) supports scientific collaboration in six program areas, ranging fr... more The Department of Energy (DOE) supports scientific collaboration in six program areas, ranging from nuclear physics, to biological and environmental research. To enable these data intensive large scale collaborations, the Energy Sciences Network (ESnet) is used to move upwards of 50 Petabytes a month between sites within the US and Europe. Bandwidth within ESnet can be requested, reserved, and utilized through the On-demand Secure Circuits and Advance Reservation System (OSCARS), which provisions network resources with guaranteed bandwidth over a known reservation schedule. Traditionally, OSCARS has only supported simple point-to-point connections between endpoints (e.g. universities, research laboratories), which has limited how efficiently the network may be harnessed by users. This paper details recent extensive enhancements prototyped for OSCARS, to be incorporated into a future release, which enable users to select novel service types including survivability, asymmetric bandwid...

Machine learning-based analysis of COVID-19 pandemic impact on US research networks

ACM SIGCOMM Computer Communication Review, 2021

Download

Analyzing Scientific Data Sharing Patterns for In-network Data Caching

Proceedings of the 2021 on Systems and Network Telemetry and Analytics, 2020

The volume of data moving through a network increases with new scientific experiments and simulat... more The volume of data moving through a network increases with new scientific experiments and simulations. Network bandwidth requirements also increase proportionally to deliver data within a certain time frame. We observe that a significant portion of the popular dataset is transferred multiple times to different users as well as to the same user for various reasons. In-network data caching for the shared data has shown to reduce the redundant data transfers and consequently save network traffic volume. In addition, overall application performance is expected to improve with in-network caching because access to the locally cached data results in lower latency. This paper shows how much data was shared over the study period, how much network traffic volume was consequently saved, and how much the temporary in-network caching increased the scientific application performance. It also analyzes data access patterns in applications and the impacts of caching nodes on the regional data repository. From the results, we observed that the network bandwidth demand was reduced by nearly a factor of 3 over the study period. CCS CONCEPTS • Networks → Network performance analysis; • Computing methodologies → Distributed computing methodologies.

Download

Nuclear Physics Network Requirements Review: One-Year Update

Download

FABRIC: A National-Scale Programmable Experimental Network Infrastructure

IEEE Internet Computing, 2019

FABRIC is a unique national research infrastructure to enable cutting-edge and exploratory resear... more FABRIC is a unique national research infrastructure to enable cutting-edge and exploratory research at-scale in networking, cybersecurity, distributed computing and storage systems, machine learning, and science applications. It is an everywhere programmable nationwide instrument comprised of novel extensible network elements equipped with large amounts of compute and storage, interconnected by high speed, dedicated optical links. It will connect a number of specialized testbeds for cloud research (NSF Cloud testbeds CloudLab and Chameleon), for research beyond 5G technologies (Platforms for Advanced Wireless research or PAWR), as well as production high-performance computing facilities and science instruments to create a rich fabric for a wide variety of experimental activities.

Download

Enabling intent to configure scientific networks for high performance demands

Future Generation Computer Systems, 2017

Globally distributed scientific experiments usually involve massive data volumes and distributed ... more Globally distributed scientific experiments usually involve massive data volumes and distributed data analysis being done by many collaborators. With complex workloads and heterogeneous resources, each user may require certain characteristics for their network paths. In this paper, we present the iNDIRA tool, which interacts with SDN northbound interfaces to enable intent-based networking. It provides reliable, simple, and technology-agnostic communication between users and networks. Focusing particularly on science applications, iNDIRA uses natural language processing to construct semantic RDF graphs to understand, interact, and create the required network services. The technical challenges addressed by iNDIRA are: (1) development of a high-level descriptive language to query for network-application requirements, (2) keyword identification and condition checking based on user profiles and topology details, (3) allow user negotiation based on the current network state, (4) provide network provisioning guidance, and finally, (5) automatically provision and monitor a layer-2 network path for use by the application. iNDIRA is implemented on the ESnet network, where it interacts with OpenNSA (aka the NSI client) and Globus data transfer tools, to build complex cross-domain network paths for heterogeneous science applications and perform secure data transfer. We describe our implementation of iNDIRA running on ESnet's production network. We present results of iNDIRA's query processing mechanism, evaluate iNDIRA's intent language evaluation with other approaches, and describe future enhancements for iNDIRA.

Download

Inder Monga

Uploads

Papers by Inder Monga

Log In