ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers

Nov 1, 20160 likes968 views

Data centers offer computational resources with various levels of guaranteed performance to the tenants, through differentiated Service Level Agreements (SLA). Typically, data center and cloud providers do not extend these guarantees to the networking layer. Since communication is carried over a network shared by all the tenants, the performance that a tenant application can achieve is unpredictable and depends on factors often beyond the tenant’s control. We propose ViTeNA, a Software-Defined Networking-based virtual network embedding algorithm and approach that aims to solve these problems by using the abstraction of virtual networks. Virtual Tenant Networks (VTN) are isolated from each other, offering virtual networks to each of the tenants, with bandwidth guarantees. Deployed along with a scalable OpenFlow controller, ViTeNA allocates virtual tenant networks in a work-conservative system. Preliminary evaluations on data centers with tree and fat-tree topologies indicate that ViTeNA achieves both high consolidation on the allocation of virtual networks and high data center resource utilization.

ViTeNA: An SDN-Based
Virtual Network Embedding Algorithm for
Multi-Tenant Data Centers
Daniel Caixinha, Pradeeban Kathiravelu, Lu s Veigaıı
Presented by: André Negrão
INESC-ID Lisboa / Instituto Superior Técnico
Universidade de Lisboa, Portugal
The 15th IEEE International Symposium on Network Computing and Applications (NCA 2016)
November 1st
, 2016. Cambridge, MA.

2
Introduction
● Differentiated SLAs for data center tenants.
● Lack of guarantees in bandwidth.
● Shared bandwidth → Unpredictable performance.
● Software-Defined Networking (SDN) offers unified
and enhanced control to the network.
– From higher levels.

3
Motivation
● Virtual Network Embedding (VNE) aims to
completely virtualize the network.
– Performance isolation among tenants in the
network level.
– Major challenge in network virtualization.
● Can we leverage SDN for a better VNE
approach?

4
Contributions
● A practical solution for the virtual network
embedding problem.
● High consolidation within the placement of
virtual networks
● High utilization of physical resources
– Servers and network.

5
ViTeNA
● A Virtual Network Embedding Algorithm
– For Multi-Tenant Data Centers
– Leveraging SDN.
● Tenants’ bandwidth requirements
– Enforced through virtual networks.

6
Deployment Landscape
● Reduce number of hops

7
Deployment Landscape
● Reduce number of hops
– Increase locality.
– Reduce communication delays.

8
ViTeNA Architecture
● Tenant demands as an XML file.
● Allocation based on the network state.

9
Implementation
● Floodlight 1.1 as the OpenFlow controller.
● Mininet 2.2.1 and Open vSwitch 2.3.1 to
emulate the data center.

10
Evaluation Deployment
● A computer with Intel ® Quad-Core i7 870 @
2.93 GHz processor
– 12 GB DDR3 @ 1333 MHz RAM
– 450 GB Serial ATA @ 7200 rpm hard disk
– Ubuntu 14.04.3 LTS (Linux Kernel 3.13.0).
● Stop an experiment when the controller returns
false to an experiment.
● Experiments run 1000 times.

11
Emulated System
● A tree topology (depth = 3; fanout = 5)
– with 125 servers
– 31 switches and 155 links
● A fat-tree topology
– factor k = 32, i.e. switches consist of 32 ports
– with 128 servers
– 160 switches and 384 links

12
Evaluation Approach
● Scalability
● High consolidation
● High resource utilization.
● Bandwidth guarantees in a work-conservative
system

13
Scalability to data center scale
● Allocation time with tree topology.

14
Scalability to data center scale
● Allocation time with fat-tree topology.

15
High Consolidation
● Allocate the VMs of a virtual network as close
as possible.
● Tree topology
● Fat-Tree topology

16
High Resource Utilization
● Server and network utilization (%)
– For tree and fat-tree.

17
Conclusion
● Conclusions
– ViTeNA addresses the unpredictable performance of the
applications.
● Using the abstraction of virtual networks.
– Evaluations confirm
● low execution time
● high consolidation on the virtual network allocation.
● high data center resource utilization.
● Future Work
– Reliability and isolation guarantees

18
Thank you!
● Questions?
pradeeban.kathiravelu@tecnico.ulisboa.pt

Cloud network systems and applications are tested in simulation and emulation environments prior to physical deployments, at different stages of development. Software-Defined Networking (SDN) enables separating logic and execution from the data plane consisting of switches and hosts, to a logically centralized control plane. The global view and control available to the controller enable incremental updates, management, and allocation of resources to the networks. However, unlike the physical networks or the networks emulated by the emulators, current network simulators still lack integration with the SDN controllers. Hence, currently it is impossible to efficiently orchestrate a simulated network through a centralized controller, or realistically model the controller algorithms and SDN architectures without having the resources for a one-to-one emulation. To address this, this paper presents SDNSim, an SDN simulation middleware, which leverages the principles of SDN for continuous development of cloud and data center networks. SDNSim is an “SDN-aware” network simulator that integrates with the controller through plugins for southbound protocols such as OpenFlow, to execute the algorithms incrementally thus deployed in the control plane.

SD-CPS: Taming the Challenges of Cyber-Physical Systems with a Software-Defin...Pradeeban Kathiravelu, Ph.D.

This document proposes a software-defined approach called SD-CPS to address the challenges of cyber-physical systems (CPS). SD-CPS introduces a centralized control plane that manages both the physical devices and a virtual representation of the CPS. This provides unified control, improved quality of service and resilience, and reduces effort in modeling complex CPS. The prototype implementation demonstrates increased controller performance and orchestration capabilities. Future work aims to improve resource efficiency, security, and cost management of large-scale CPS using this approach.

Selective Redundancy in Network-as-a-Service: Differentiated QoS in Multi-Ten...Pradeeban Kathiravelu, Ph.D.

Data centers consist of various users with multiple roles and differentiated levels of access. Tenant execution flows can be of different priorities based on the role of the tenant and the nature of the process. Traditionally enterprise network optimizations are made at each specific layer, from the physical layer to the application layer. However, a cross-layer optimization of cloud networks would utilize the data available to each of the layers in a more efficient manner. This paper proposes an approach and architecture for differentiated quality of service (QoS). By employing a selective redundancy in a controlled manner, end-to-end delivery is guaranteed for priority tenant application flows despite congestion. The architecture, in a higher level, focuses on exploiting the global knowledge of the underlying network readily available to the Software-Defined Networking (SDN) controller to cater the requirements of the tenant applications. QoS is guaranteed to the critical tenant flows in multi-tenant clouds by cross-layer enhancements across the network and application layers.

Building Blocks of Mayan: Componentizing the eScience Workflows Through Softw...Pradeeban Kathiravelu, Ph.D.

eScience consists of computation-intensive workflows executing on highly distributed networks. Service compositions aggregate web services to automate scientific and enterprise business processes. Along with the increased demand for data quality and Quality of Service (QoS) for an accurate outcome in a shorter completion time, execution of the eScience workflows and service compositions are also required to be distributed efficiently across various geo-distributed nodes. This paper presents Mayan, a Software-Defined Networking (SDN) based approach for service composition. Mayan i) facilitates an adaptive execution of scientific workflows, ii) offers a more efficient service composition by leveraging distributed execution frameworks, in addition to the traditional web service engines, and iii) enables a very large-scale reliable service composition by finding and consuming the current best-fit among the multiple implementations or deployments of the same service.

SDN-Based Enhancements to QoS and Data Quality in Multi-Tenanted Data Center ...Pradeeban Kathiravelu, Ph.D.

Pradeeban Kathiravelu is a PhD student researching how to use SDN to improve QoS and data quality in multi-tenant data center networks. His approach is to deploy an extended SDN controller architecture to increase QoS and enhance data quality for stored and processed data. He has published several papers on this topic and ongoing work includes the SMART project, with promising results showing improved data isolation, quality, and QoS guarantees through the use of SDN.

Software-Defined Approach for QoS and Data Quality in Multi-Tenant CloudsPradeeban Kathiravelu, Ph.D.

CHIEF: Controller Farm for Clouds of Software-Defined Community NetworksPradeeban Kathiravelu, Ph.D.

The document proposes CHIEF, a networking platform that uses a federation of SDN controllers to orchestrate multi-tenant community networks and enable interoperability between community network clouds. CHIEF offers scalability, privacy, and control for software-defined community networks through its controller farm architecture and auxiliary services like monitoring, throttling, and metering. Preliminary assessments of CHIEF prototypes show the feasibility of its complimentary monitoring and billing capabilities for community networks.

$SENDIM for Incremental Development of Cloud Networks: Simulation, Emulation \...$ $SENDIM for Incremental Development of Cloud Networks: Simulation, Emulation \...$

SENDIM for Incremental Development of Cloud Networks: Simulation, Emulation \...Pradeeban Kathiravelu, Ph.D.

This document describes SENDIM, a middleware for integrating network simulation, emulation, and deployment. SENDIM uses Software-Defined Networking (SDN) principles to enable incremental development from simulation to emulation to live deployment. It leverages configuration management tools and checkpointing/versioning through OSGi to minimize repetition across stages. The goal is to streamline the process of developing and testing cloud networks.

Scalability and Resilience of Multi-Tenant Distributed Clouds in the Big Serv...Pradeeban Kathiravelu, Ph.D.

This document summarizes the PhD work of Pradeeban Kathiravelu on improving scalability and resilience in multi-tenant distributed clouds. It describes two approaches: 1) SMART uses SDN to provide differentiated quality of service and service level agreements by dynamically diverting and cloning priority network flows. 2) Mayan componentizes big data services as microservices that can be executed in a network-aware and scalable way across distributed clouds. Evaluation shows these approaches improve speedup and ensure SLAs for critical flows compared to network-agnostic distributed execution.

Rain technology seminar Mufeedh Muhammed

This document summarizes a seminar on RAIN technology. RAIN (Reliable Array of Independent Nodes) originated at Caltech to create a redundant network of processors that could automatically recover from failures. It aims to minimize connections between clients and servers and make nodes robust and independent. The architecture includes reliable transport, consistent global state sharing, and local/global fault monitors. Key features are scalability, high availability, fault tolerance for nodes, networks and data storage. Applications include high availability video and web servers. Future work involves further developing APIs and a distributed file system.

Standardising the compressed representation of neural networksFörderverein Technische Fakultät

Artificial neural networks have been adopted for a broad range of tasks in multimedia analysis and processing, such as visual and acoustic classification, extraction of multimedia descriptors or image and video coding. The trained neural networks for these applications contain a large number of parameters (weights), resulting in a considerable size. Thus, transferring them to a number of clients using them in applications (e.g., mobile phones, smart cameras) benefits from a compressed representation of neural networks. MPEG Neural Network Coding and Representation is the first international standard for efficient compression of neural networks (NNs). The standard is designed as a toolbox of compression methods, which can be used to create coding pipelines. It can be either used as an independent coding framework (with its own bitstream format) or together with external neural network formats and frameworks. For providing the highest degree of flexibility, the network compression methods operate per parameter tensor in order to always ensure proper decoding, even if no structure information is provided. The standard contains compression-efficient quantization and an arithmetic coding scheme (DeepCABAC) as core encoding and decoding technologies, as well as neural network parameter pre-processing methods like sparsification, pruning, low-rank decomposition, unification, local scaling, and batch norm folding. NNR achieves a compression efficiency of more than 97% for transparent coding cases, i.e. without degrading classification quality, such as top-1 or top-5 accuracies. This talk presents an overview of the context, technical features, and characteristics of the NN coding standard, and discusses ongoing topics such as incremental neural network representation.

2019-06-14:7 - Neutral Network Compressionuninfoit

This document discusses approaches to neural network compression, which is important for enabling AI capabilities on edge devices with limited resources. It describes how compression techniques like quantization, pruning and sharing, tensor decomposition, and network transformation can reduce model size and computational requirements by approximating weights and filters. The goal is to compress pre-trained models for more efficient inference while maintaining accuracy. Combining multiple compression methods can achieve greater reductions in model complexity.

Device Data Directory and Asynchronous execution: A path to heterogeneous com...LEGATO project

Moldable pipelines for CNNs on heterogeneous edge devicesLEGATO project

The LEGaTO project received funding from the European Union to develop a framework for efficiently running CNNs on heterogeneous edge devices. The framework implements a brief online training to find a near-optimal pipeline configuration that balances performance across different compute resources. It generates high-throughput CNN pipelines for edge devices containing variable core configurations on a single chip by leveraging computational hints during interface-guided partitioning and online adaptation of pipeline stages.

M tech-2015 vlsi-newAditya Undralla

QoS-Aware Data Replication for Data-Intensive Applications in Cloud Computing...Papitha Velumani

Fast aggregation scheduling in wireless sensor networksLogicMindtech Nologies

Resume_Mahadevan_new (2)Mahadevan N

Mahadevan Nagarajan is seeking a full-time position in VLSI design utilizing his Master's degree in Electrical Engineering from the University of Minnesota and Bachelor's degree from Anna University, India. He has experience with VLSI CAD tools, programming languages, and HDL synthesis tools. Current projects include designing a 128kb SRAM and 16-bit adder using the NCSU 45nm process in Cadence. Previous experience includes power analysis internship at Intel and systems engineering role at Tata Consultancy Services.

Ieee 2015 16 vlsi @dreamweb techno solutions-trichysubhu8430

Update on the Mont-Blanc Project for ARM-based HPCinside-BigData.com

In this deck from the GoingARM workshop at SC17, Filippo Mantovani describes the contributions of the Barcelona Supercomputing Center to the European Mont-Blanc project. "Since 2011, Mont-Blanc has pushed the adoption of Arm technology in High Performance Computing, deploying Arm-based prototypes, enhancing system software ecosystem and projecting performance of current systems for developing new, more powerful and less power hungry HPC computing platforms based on Arm SoC. In this talk, Filippo introduces the last Mont-Blanc system, called Dibona, designed and integrated by the coordinator and industrial partner of the project, Bull/ATOS. He also talks about tests performed at BSC of the Arm software tools (HPC compiler and mathematical libraries) as well as the Dynamic Load Balancing (DLB) technique and the Multiscale Simulator Architecture (MUSA)." Watch the video: https://wp.me/p3RLHQ-i6o Learn more: http://www.goingarm.com/ Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter

Rain Technologykamini chaudhary

Rain Technology is a disk subsystem that provides distributed data storage and protection in network architecture by integrating inexpensive hardware and management software. This research is focused on - high performance, - fault tolerance , - portable clustering technology. It consist of heterogeneous cluster of computing nodes connected via multiple interfaces to networks configured in fault tolerant topologies.

18IMPULSE_TECHNOLOGY

OpenACC Monthly Highlights Summer 2019OpenACC

Efficient Data Center Virtualization with QLogic 10GbE Solutions from HPJone Smith

An Introduction to Google Summer of Code 2015Pradeeban Kathiravelu, Ph.D.

powerpoint febimu409

The document proposes using an ensemble of K-nearest neighbor classifiers optimized with genetic programming for intrusion detection. It trains multiple K-NN classifiers on subsets of the KDD Cup 1999 intrusion detection dataset and then uses genetic programming to combine the classifiers to improve performance. Results show the ensemble approach reduces error rates compared to individual classifiers and the genetic programming-based ensemble achieves an area under the ROC curve of 0.99976, outperforming the component classifiers.

∂u∂u Multi-Tenanted Framework: Distributed Near Duplicate Detection for Big DataPradeeban Kathiravelu, Ph.D.

Near duplicate detection algorithms have been proposed and implemented in order to detect and eliminate duplicate entries from massive datasets. Due to the differences in data representation (such as measurement units) across different data sources, potential duplicates may not be textually identical, even though they refer to the same real-world entity. As data warehouses typically contain data coming from several heterogeneous data sources, detecting near duplicates in a data warehouse requires a considerable memory and processing power. Traditionally, near duplicate detection algorithms are sequential and operate on a single computer. While parallel and distributed frameworks have recently been exploited in scaling the existing algorithms to operate over larger datasets, they are often focused on distributing a few chosen algorithms using frameworks such as MapReduce. A common distribution strategy and framework to parallelize the execution of the existing similarity join algorithms is still lacking. In-Memory Data Grids (IMDG) offer a distributed storage and execution, giving the illusion of a single large computer over multiple computing nodes in a cluster. This paper presents the research, design, and implementation of ∂u∂u, a distributed near duplicate detection framework, with preliminary evaluations measuring its performance and achieved speed up. ∂u∂u leverages the distributed shared memory and execution model provided by IMDG to execute existing near duplicate detection algorithms in a parallel and multi-tenanted environment. As a unified near duplicate detection framework for big data, ∂u∂u efficiently distributes the algorithms over utility computers in research labs and private clouds and grids.

Near Duplicate Detection for Medical Imaging Data Warehouse ConstructionPradeeban Kathiravelu, Ph.D.

Adaptive Intrusion Detection Using Learning ClassifiersPatrick Nicolas

The document discusses using learning classifiers and genetic algorithms to implement an adaptive intrusion detection system. Traditional data mining techniques are limited in their ability to adapt to changing environments, but learning classifiers systems combine genetic algorithms and reinforcement learning to discover and evolve security policies and rules from real-time data. The rules are represented as genes and evolved over time through processes of crossover, mutation, and selection to accurately identify threats.

More Related Content

What's hot (16)

Scalability and Resilience of Multi-Tenant Distributed Clouds in the Big Serv...Pradeeban Kathiravelu, Ph.D.

Rain technology seminar Mufeedh Muhammed

Standardising the compressed representation of neural networksFörderverein Technische Fakultät

2019-06-14:7 - Neutral Network Compressionuninfoit

Device Data Directory and Asynchronous execution: A path to heterogeneous com...LEGATO project

Moldable pipelines for CNNs on heterogeneous edge devicesLEGATO project

M tech-2015 vlsi-newAditya Undralla

QoS-Aware Data Replication for Data-Intensive Applications in Cloud Computing...Papitha Velumani

Fast aggregation scheduling in wireless sensor networksLogicMindtech Nologies

Resume_Mahadevan_new (2)Mahadevan N

Ieee 2015 16 vlsi @dreamweb techno solutions-trichysubhu8430

Update on the Mont-Blanc Project for ARM-based HPCinside-BigData.com

Rain Technologykamini chaudhary

18IMPULSE_TECHNOLOGY

OpenACC Monthly Highlights Summer 2019OpenACC

Scalability and Resilience of Multi-Tenant Distributed Clouds in the Big Serv...Pradeeban Kathiravelu, Ph.D.

Rain technology seminar Mufeedh Muhammed

Standardising the compressed representation of neural networksFörderverein Technische Fakultät

2019-06-14:7 - Neutral Network Compressionuninfoit

Device Data Directory and Asynchronous execution: A path to heterogeneous com...LEGATO project

Moldable pipelines for CNNs on heterogeneous edge devicesLEGATO project

M tech-2015 vlsi-newAditya Undralla

QoS-Aware Data Replication for Data-Intensive Applications in Cloud Computing...Papitha Velumani

Fast aggregation scheduling in wireless sensor networksLogicMindtech Nologies

Resume_Mahadevan_new (2)Mahadevan N

Ieee 2015 16 vlsi @dreamweb techno solutions-trichysubhu8430

Update on the Mont-Blanc Project for ARM-based HPCinside-BigData.com

Rain Technologykamini chaudhary

18IMPULSE_TECHNOLOGY

OpenACC Monthly Highlights Summer 2019OpenACC

Viewers also liked (20)

Efficient Data Center Virtualization with QLogic 10GbE Solutions from HPJone Smith

An Introduction to Google Summer of Code 2015Pradeeban Kathiravelu, Ph.D.

powerpoint febimu409

∂u∂u Multi-Tenanted Framework: Distributed Near Duplicate Detection for Big DataPradeeban Kathiravelu, Ph.D.

Near Duplicate Detection for Medical Imaging Data Warehouse ConstructionPradeeban Kathiravelu, Ph.D.

Adaptive Intrusion Detection Using Learning ClassifiersPatrick Nicolas

Application scheduling in cloud simPradeeban Kathiravelu, Ph.D.

This document evaluates scheduling algorithms for applications in a cloud environment. It compares strict matchmaking-based algorithms like minimum execution time, minimum completion time, and maximum resource utilization to utility-driven algorithms that consider user satisfaction and partial requirement satisfaction. The evaluations are conducted using CloudSim, a cloud simulation tool, by modeling cloud resources, applications, and scheduling various workloads under different algorithms to analyze metrics like completion time and resource utilization. The results show that utility-driven algorithms that take user requirements into account perform better overall.

machine learning in the age of big data: new approaches and business applicat...Armando Vieira

Algorithme DPLLElhem Sassi

Intrusion detection using data miningbalbeerrawat

This document describes a project to develop an intrusion detection system using data mining techniques. It discusses approaches to intrusion detection including signature-based and anomaly-based methods. For the project, a hybrid network-based and host-based intrusion detection system is proposed. Data preprocessing and mining techniques including clustering, outlier detection, and classification are applied to network packet data and system call logs to detect attacks.

Ids presentationSolmaz Salehian

The document discusses energy-efficient intrusion detection techniques for wireless sensor networks. It summarizes routing protocols used in WSNs and proposes using hybrid anomaly and misuse detection at cluster heads in hierarchical routing to increase detection rates while reducing energy consumption. For flat-based routing, it suggests using statistical anomaly detection at each node. For location-based routing, it proposes detecting intrusions based on location and trust information to limit communication between distant nodes and the base station. Simulation results on real sensor network data show the approaches can effectively detect intrusions while preserving energy.

Analysis and Design for Intrusion Detection System Based on Data MiningPritesh Ranjan

This document discusses using data mining techniques to improve intrusion detection systems (IDS). It begins by introducing computer network risks and limitations of existing IDS approaches. It then discusses using data mining algorithms like ID3, k-means clustering, and Apriori pattern mining within a hybrid IDS framework. The framework includes sensors to collect host and network data, a data warehouse for storage, and an analysis engine using misuse detection, anomaly detection and data mining algorithms to detect intrusions. It concludes that data mining allows IDS to detect both known and unknown attacks more efficiently.

2015 01-17 Lambda Architecture with Apache Spark, NextML ConferenceDB Tsai

Lambda architecture is a data-processing architecture designed to handle massive quantities of data by taking advantage of both batch- and stream-processing methods. In Lambda architecture, the system involves three layers: batch processing, speed (or real-time) processing, and a serving layer for responding to queries, and each comes with its own set of requirements. In batch layer, it aims at perfect accuracy by being able to process the all available big dataset which is an immutable, append-only set of raw data using distributed processing system. Output will be typically stored in a read-only database with result completely replacing existing precomputed views. Apache Hadoop, Pig, and HIVE are the de facto batch-processing system. In speed layer, the data is processed in streaming fashion, and the real-time views are provided by the most recent data. As a result, the speed layer is responsible for filling the "gap" caused by the batch layer's lag in providing views based on the most recent data. This layer's views may not be as accurate as the views provided by batch layer's views created with full dataset, so they will be eventually replaced by the batch layer's views. Traditionally, Apache Storm is used in this layer. In serving layer, the result from batch layer and speed layer will be stored here, and it responds to queries in a low-latency and ad-hoc way. One of the lambda architecture examples in machine learning context is building the fraud detection system. In speed layer, the incoming streaming data can be used for online learning to update the model learnt in batch layer to incorporate the recent events. After a while, the model can be rebuilt using the full dataset. Why Spark for lambda architecture? Traditionally, different technologies are used in batch layer and speed layer. If your batch system is implemented with Apache Pig, and your speed layer is implemented with Apache Storm, you have to write and maintain the same logics in SQL and in Java/Scala. This will very quickly becomes a maintenance nightmare. With Spark, we have an unified development framework for batch and speed layer at scale. In this talk, an end-to-end example implemented in Spark will be shown, and we will discuss about the development, testing, maintenance, and deployment of lambda architecture system with Apache Spark.

Using Machine Learning in Networks Intrusion Detection SystemsOmar Shaya

The internet and different computing devices from desktop computers to smartphones have raised many security and privacy concerns, and the need to automate systems that detect attacks on these networks has emerged in order to be able to protect these networks with scale. And while traditional intrusion detection methods may be able to detect previously known attacks, the issue of dealing with new unknown attacks arises and that brings machine learning as a strong candidate to solve these challenges. In this report, we investigate the use of machine learning in detecting network attacks, intrusion detection, by looking at work that has been done in this field. Particularly we look at the work that has been done by Pasocal et al.

Efficient Duplicate Detection Over Massive Data SetsPradeeban Kathiravelu, Ph.D.

Data Café — A Platform For Creating Biomedical Data LakesPradeeban Kathiravelu, Ph.D.

Data Mining and Intrusion Detection amiable_indian

The document discusses using data mining approaches for intrusion detection. It describes current intrusion detection approaches like misuse detection using signatures of known attacks and anomaly detection using deviations from normal behavior profiles. Data mining can help by providing a systematic framework to select relevant audit data features, build and update detection models, and combine multiple models. Relevant techniques include building classifiers from audit data and mining patterns within audit records.

NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique Sujeet Suryawanshi

This document summarizes a presentation given on using decision trees and machine learning techniques for anomaly detection on the NSL KDD Cup 99 dataset. It discusses anomaly detection, machine learning, different machine learning algorithms like decision trees, SVM, Naive Bayes etc. and their application for intrusion detection. It then describes an experiment conducted using the decision tree algorithm on the NSL KDD Cup 99 dataset to classify network traffic as normal or anomalous. The results showed the decision tree model achieved over 98% accuracy on both the full dataset and a reduced feature set.

Intrusion detection and prevention systemNikhil Raj

Intrusion detection system pptSheetal Verma

This document discusses intrusion detection systems (IDS). An IDS monitors network or system activities for malicious activities or policy violations. IDS can be classified based on detection method (anomaly-based detects deviations from normal usage, signature-based looks for known attack patterns) or location (host-based monitors individual systems, network-based monitors entire network traffic). The document outlines strengths and limitations of different IDS types and discusses the future of integrating detection methods.

Efficient Data Center Virtualization with QLogic 10GbE Solutions from HPJone Smith

An Introduction to Google Summer of Code 2015Pradeeban Kathiravelu, Ph.D.

powerpoint febimu409

∂u∂u Multi-Tenanted Framework: Distributed Near Duplicate Detection for Big DataPradeeban Kathiravelu, Ph.D.

Near Duplicate Detection for Medical Imaging Data Warehouse ConstructionPradeeban Kathiravelu, Ph.D.

Adaptive Intrusion Detection Using Learning ClassifiersPatrick Nicolas

Application scheduling in cloud simPradeeban Kathiravelu, Ph.D.

machine learning in the age of big data: new approaches and business applicat...Armando Vieira

Algorithme DPLLElhem Sassi

Intrusion detection using data miningbalbeerrawat

Ids presentationSolmaz Salehian

Analysis and Design for Intrusion Detection System Based on Data MiningPritesh Ranjan

2015 01-17 Lambda Architecture with Apache Spark, NextML ConferenceDB Tsai

Using Machine Learning in Networks Intrusion Detection SystemsOmar Shaya

Efficient Duplicate Detection Over Massive Data SetsPradeeban Kathiravelu, Ph.D.

Data Café — A Platform For Creating Biomedical Data LakesPradeeban Kathiravelu, Ph.D.

Data Mining and Intrusion Detection amiable_indian

NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique Sujeet Suryawanshi

Intrusion detection and prevention systemNikhil Raj

Intrusion detection system pptSheetal Verma

Similar to ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers (20)

The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreePradeeban Kathiravelu, Ph.D.

What is 3d torusEurotech Aurora

Hardware for Deep Learning AI ML CNN.pdfAhmedSaeed115917

Software-Defined Systems for Network-Aware Service Composition and Workflow P...Pradeeban Kathiravelu, Ph.D.

Presentation oracle net servicesxKinAnx

Network-aware Data Management for High Throughput Flows Akamai, Cambridge, ...balmanme

The document discusses Mehmet Balman's work on network-aware data management for large-scale distributed applications. It provides background on Balman, including his employment at VMware and affiliations. The presentation outline discusses VSAN and VVOL storage performance in virtualized environments, data streaming in high-bandwidth networks, the Climate100 100Gbps networking demo, and other topics related to network-aware data management.

Improving performance and efficiency with Network Virtualization OverlaysAdam Johnson

DOE Magellan OpenStack user storylaurabeckcahoon

The document summarizes Magellan's experiences with OpenStack as a cloud computing platform for high performance computing workloads. Initial testing found stability and scalability issues with Eucalyptus, but OpenStack performed much better. A production OpenStack deployment was established to support scientific computing projects. Further work optimized networking performance, achieving near-wire speed wide area data transfers of 95 Gbps using OpenStack. The results demonstrate OpenStack's potential for on-demand high performance computing and data-intensive workloads.

Nginx performance 2015 09 23Bruce Tolley

Making the Web Faster with User Level Networking: 2X NGINX Performance Increa...Bruce Tolley

This presentation explains how three key features of Solarflare's OpenOnload application acceleration stack contribute to increasing Nginx application performance. First, OpenOnload allows network processing at user-level, bypassing the OS kernel. Because OpenOnload includes a TCP stack that runs completely in user space, it avoids the overhead of system calls that consume CPU resources. In addition, application and network performance is improved without sacrificing the security and multiplexing functions that the OS kernel normally provides. Second, OpenOnload now supports the SO_REUSEPORT socket option that allows multiple TCP listening sockets to bind to the same IP address and port and to distribute incoming connection requests between these sockets. This feature can be exploited by applications to increase their scalability. OpenOnload’s implementation of SO_REUSEPORT is superior to that of the Linux kernel because OpenOnload has the ability to instantiate independent, per process, per thread network stacks each of which has direct access to hardware. This eliminates lock contention between processes when sending to and receiving from the network stack. Third, OpenOnload supports socket caching, which avoids the overhead of additional system calls. This improves performance when connections are short-lived.

ddsf-student-presentation_756205.pptxssuser498be2

The document discusses deploying containerized DDS applications on Kubernetes clusters. It explores the feasibility and challenges of using different Kubernetes CNI plugins like Flannel, WeaveNet, and Kube-Router. The document presents results from benchmark tests measuring the performance overhead of container virtualization, different CNIs, multicast vs unicast communication, and different security implementations. It finds that performance is affected by factors like system topology, QoS policies and CNI configurations. The automated benchmark framework provides a tool to evaluate DDS performance on Kubernetes clusters.

Using Galera Cluster to Power Geo-distributed Applications on the WANphilip_stoev

This document discusses using Galera Cluster to power geo-distributed applications across wide area networks (WANs). Some key points: - Galera Cluster allows MySQL databases to be replicated across multiple data centers separated by thousands of kilometers to achieve high availability and disaster recovery. - Replicating data closer to applications improves performance by answering most read queries locally with up-to-date data cached in the InnoDB buffer pool. - Galera is optimized for WAN replication with features like encrypting traffic, automatic detection and removal of unreliable nodes, and minimizing cross-data center traffic. - Basic configuration involves specifying network locations, opening required ports, and configuring IPs for each node. Security must

40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facilityinside-BigData.com

In this deck from the Swiss HPC Conference, Mark Wilkinson presents: 40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facility. "DiRAC is the integrated supercomputing facility for theoretical modeling and HPC-based research in particle physics, and astrophysics, cosmology, and nuclear physics, all areas in which the UK is world-leading. DiRAC provides a variety of compute resources, matching machine architecture to the algorithm design and requirements of the research problems to be solved. As a single federated Facility, DiRAC allows more effective and efficient use of computing resources, supporting the delivery of the science programs across the STFC research communities. It provides a common training and consultation framework and, crucially, provides critical mass and a coordinating structure for both small- and large-scale cross-discipline science projects, the technical support needed to run and develop a distributed HPC service, and a pool of expertise to support knowledge transfer and industrial partnership projects. The on-going development and sharing of best-practice for the delivery of productive, national HPC services with DiRAC enables STFC researchers to produce world-leading science across the entire STFC science theory program." Watch the video: https://wp.me/p3RLHQ-k94 Learn more: https://dirac.ac.uk/ and http://hpcadvisorycouncil.com/events/2019/swiss-workshop/agenda.php Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter

Data Replication In Cloud ComputingRahul Garg

1) The document discusses quality of service (QoS)-aware data replication for data-intensive applications in cloud computing systems. It aims to minimize data replication cost and number of QoS violated replicas. 2) It presents a mathematical model and algorithm to optimally place QoS-satisfied and QoS-violated data replicas. The algorithm uses minimum-cost maximum flow to obtain the optimal placement. 3) The algorithm takes as input a set of requested nodes and outputs the optimal placement for QoS-satisfied and QoS-violated replicas by modeling the problem as a network flow graph and applying existing polynomial-time algorithms.

Everything as a Servicejavicid

The document discusses technologies for infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (SaaS) and how HP is working on solutions in these areas including server cooling technologies to improve energy efficiency for IaaS, a cell-as-a-service model to provide isolated infrastructure for PaaS, and several proposed software services for collective intelligence and collaboration for SaaS. The document concludes that while progress has been made in cloud computing, challenges remain that require continued research and development efforts.

OCRE webinar - April 14 - Cloud_Validation_Suite_Ignacio Peluaga Lozada.pdfOCRE | Open Clouds for Research Environments

The document discusses OCRE validation and testing. It describes the need for a tool to automate deployment of workloads for benchmarking and validating cloud services. The approach taken was to create an automated, abstracted, heterogeneous test suite using open technologies like Terraform, Ansible, Docker and Kubernetes. A variety of workloads were deployed across different domains like compute, networking, storage and machine learning. Results were collected and shared confidentially with cloud providers to evaluate areas like infrastructure, connectivity and APIs. Next steps include expanding the test catalogue and custom workload support.

SoC Solutions Enabling Server-Based NetworkingNetronome

1) Current server-based networking solutions have low throughput and high CPU loads, limiting application performance. Netronome aims to improve efficiency with network flow processors. 2) Netronome's network flow processor uses chip multithreading across many processor engines to distribute packet processing and meet high bandwidth needs for software-defined networking. 3) Netronome provides intelligent server adapters that transparently offload networking functions to improve server efficiency for applications compared to kernel or user-space virtual switches.

Zero Downtime JEE ArchitecturesAlexander Penev

This document discusses zero downtime architectures. It defines zero downtime as services being available to end users at all times. It identifies sources of planned and unplanned downtime. It proposes concepts like independent application groups, redundant infrastructure within and between datacenters, and replicating data between datacenters to reduce downtime. It provides examples of implementing high availability for networks, applications, and databases. It also discusses development guidelines and monitoring to support zero downtime operations.

A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...Tal Lavian Ph.D.

The new architecture is proposed for data intensive enabled by next generation dynamic optical networks Encapsulates “optical network resources” into a service framework to support dynamically provisioned and advanced data-intensive transport services Provides a generalized framework for high performance applications over next generation networks, not necessary optical end-to-end Supports both on-demand and scheduled data retrieval Supports a meshed wavelength switched network capable of establishing an end-to-end lightpath in seconds Supports bulk data-transfer facilities using lambda-switched networks Supports out-of-band tools for adaptive placement of data replicas Offers network resources as Grid services for Grid computing

"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...Edge AI and Vision Alliance

For the full video of this presentation, please visit: http://www.embedded-vision.com/platinum-members/auvizsystems/embedded-vision-training/videos/pages/may-2015-embedded-vision-summit For more information about embedded vision, please visit: http://www.embedded-vision.com Nagesh Gupta, CEO and Founder of Auviz Systems, presents the "Trade-offs in Implementing Deep Neural Networks on FPGAs" tutorial at the May 2015 Embedded Vision Summit. Video and images are a key part of Internet traffic—think of all the data generated by social networking sites such as Facebook and Instagram—and this trend continues to grow. Extracting usable information from video and images is thus a growing requirement in the data center. For example, object and face recognition are valuable for a wide range of uses, from social applications to security applications. Deep neural networks are currently the most popular form of convolutional neural networks (CNN) used in data centers for such applications. 3D convolutions are a core part of CNNs. Nagesh presents alternative implementations of 3D convolutions on FPGAs, and discusses trade-offs among them.

The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreePradeeban Kathiravelu, Ph.D.

What is 3d torusEurotech Aurora

Hardware for Deep Learning AI ML CNN.pdfAhmedSaeed115917

Software-Defined Systems for Network-Aware Service Composition and Workflow P...Pradeeban Kathiravelu, Ph.D.

Presentation oracle net servicesxKinAnx

Network-aware Data Management for High Throughput Flows Akamai, Cambridge, ...balmanme

Improving performance and efficiency with Network Virtualization OverlaysAdam Johnson

DOE Magellan OpenStack user storylaurabeckcahoon

Nginx performance 2015 09 23Bruce Tolley

Making the Web Faster with User Level Networking: 2X NGINX Performance Increa...Bruce Tolley

ddsf-student-presentation_756205.pptxssuser498be2

Using Galera Cluster to Power Geo-distributed Applications on the WANphilip_stoev

40 Powers of 10 - Simulating the Universe with the DiRAC HPC Facilityinside-BigData.com

Data Replication In Cloud ComputingRahul Garg

Everything as a Servicejavicid

OCRE webinar - April 14 - Cloud_Validation_Suite_Ignacio Peluaga Lozada.pdfOCRE | Open Clouds for Research Environments

SoC Solutions Enabling Server-Based NetworkingNetronome

Zero Downtime JEE ArchitecturesAlexander Penev

A Platform for Data Intensive Services Enabled by Next Generation Dynamic Opt...Tal Lavian Ph.D.

"Trade-offs in Implementing Deep Neural Networks on FPGAs," a Presentation fr...Edge AI and Vision Alliance

More from Pradeeban Kathiravelu, Ph.D. (17)

Google Summer of Code_2023.pdfPradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2022Pradeeban Kathiravelu, Ph.D.

This document provides information about Google Summer of Code (GSoC) 2022. It discusses why students should participate in GSoC, the application timeline and process, tips for finding projects and communicating with mentors, expectations during the coding and evaluation periods, and opportunities to continue contributing to open source projects after GSoC. The overall goal is to help potential contributors understand what is required to be accepted into and succeed in GSoC.

Niffler: A DICOM Framework for Machine Learning and Processing Pipelines.Pradeeban Kathiravelu, Ph.D.

Google summer of code (GSoC) 2021Pradeeban Kathiravelu, Ph.D.

A DICOM Framework for Machine Learning Pipelines against Real-Time Radiology ...Pradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2020 for mentorsPradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2020Pradeeban Kathiravelu, Ph.D.

Data Services with Bindaas: RESTful Interfaces for Diverse Data SourcesPradeeban Kathiravelu, Ph.D.

The diversity of data management systems affords developers the luxury of building heterogeneous architectures to address the unique needs of big data. It allows one to mix-n-match systems that can store, query, update, and process data based on specific use cases. However, this heterogeneity brings with it the burden of developing custom interfaces for each data management system. Existing big data frameworks fall short in mitigating these challenges imposed. In this paper, we present Bindaas, a secure and extensible big data middleware that offers uniform access to diverse data sources. By providing a RESTful web service interface to the data sources, Bindaas exposes query, update, store, and delete functionality of the data sources as data service APIs, while providing turn-key support for standard operations involving access control and audit-trails. The research community has deployed Bindaas in various production environments in healthcare. Our evaluations highlight the efficiency of Bindaas in serving concurrent requests to data source instances with minimal overheads.

My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Compos...Pradeeban Kathiravelu, Ph.D.

My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Composi...Pradeeban Kathiravelu, Ph.D.

UCL Ph.D. Confirmation 2018Pradeeban Kathiravelu, Ph.D.

Moving bits with a fleet of shared virtual routersPradeeban Kathiravelu, Ph.D.

Software-Defined Data Services: Interoperable and Network-Aware Big Data Exec...Pradeeban Kathiravelu, Ph.D.

Services that access or process a large volume of data are known as data services. Big data frameworks consist of diverse storage media and heterogeneous data formats. Through their service-based approach, data services offer a standardized execution model to big data frameworks. Software-Defined Networking (SDN) increases the programmability of the network, by unifying the control plane centrally, away from the distributed data plane devices. In this paper, we present Software-Defined Data Services (SDDS), extending the data services with the SDN paradigm. SDDS consists of two aspects. First, it models the big data executions as data services or big services composed of several data services. Then, it orchestrates the services centrally in an interoperable manner, by logically separating the executions from the storage. We present the design of an SDDS orchestration framework for network-aware big data executions in data centers. We then evaluate the performance of SDDS through microbenchmarks on a prototype implementation. By extending SDN beyond data centers, we can deploy SDDS in broader execution environments. https://kkpradeeban.blogspot.com/2018/04/software-defined-data-services.html

On-Demand Service-Based Big Data Integration: Optimized for Research Collabor...Pradeeban Kathiravelu, Ph.D.

Software-Defined Inter-Cloud Composition of Big ServicesPradeeban Kathiravelu, Ph.D.

Componentizing Big Services in the InternetPradeeban Kathiravelu, Ph.D.

Big services need to be componentized into microservices and composed together to execute as distributed applications across organizations and users on the internet. By using software-defined networking and sending the service to the data instead of sending data across the globe, these componentized microservices can be executed in a network-aware manner based on policies that map services to the network. This allows for subscription-based inter-domain communication when composing big services in a software-defined inter-cloud.

Google Summer of Code_2023.pdfPradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2022Pradeeban Kathiravelu, Ph.D.

Niffler: A DICOM Framework for Machine Learning and Processing Pipelines.Pradeeban Kathiravelu, Ph.D.

Google summer of code (GSoC) 2021Pradeeban Kathiravelu, Ph.D.

A DICOM Framework for Machine Learning Pipelines against Real-Time Radiology ...Pradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2020 for mentorsPradeeban Kathiravelu, Ph.D.

Google Summer of Code (GSoC) 2020Pradeeban Kathiravelu, Ph.D.

Data Services with Bindaas: RESTful Interfaces for Diverse Data SourcesPradeeban Kathiravelu, Ph.D.

My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Compos...Pradeeban Kathiravelu, Ph.D.

My Ph.D. Defense - Software-Defined Systems for Network-Aware Service Composi...Pradeeban Kathiravelu, Ph.D.

UCL Ph.D. Confirmation 2018Pradeeban Kathiravelu, Ph.D.

Moving bits with a fleet of shared virtual routersPradeeban Kathiravelu, Ph.D.

Software-Defined Data Services: Interoperable and Network-Aware Big Data Exec...Pradeeban Kathiravelu, Ph.D.

On-Demand Service-Based Big Data Integration: Optimized for Research Collabor...Pradeeban Kathiravelu, Ph.D.

Software-Defined Inter-Cloud Composition of Big ServicesPradeeban Kathiravelu, Ph.D.

Componentizing Big Services in the InternetPradeeban Kathiravelu, Ph.D.

Recently uploaded (20)

How analogue intelligence complements AIPaul Rowe

Artificial Intelligence is providing benefits in many areas of work within the heritage sector, from image analysis, to ideas generation, and new research tools. However, it is more critical than ever for people, with analogue intelligence, to ensure the integrity and ethical use of AI. Including real people can improve the use of AI by identifying potential biases, cross-checking results, refining workflows, and providing contextual relevance to AI-driven results. News about the impact of AI often paints a rosy picture. In practice, there are many potential pitfalls. This presentation discusses these issues and looks at the role of analogue intelligence and analogue interfaces in providing the best results to our audiences. How do we deal with factually incorrect results? How do we get content generated that better reflects the diversity of our communities? What roles are there for physical, in-person experiences in the digital world?

Financial Services Technology Summit 2025Ray Bugg

The FS Technology Summit Technology increasingly permeates every facet of the financial services sector, from personal banking to institutional investment to payments. The conference will explore the transformative impact of technology on the modern FS enterprise, examining how it can be applied to drive practical business improvement and frontline customer impact. The programme will contextualise the most prominent trends that are shaping the industry, from technical advancements in Cloud, AI, Blockchain and Payments, to the regulatory impact of Consumer Duty, SDR, DORA & NIS2. The Summit will bring together senior leaders from across the sector, and is geared for shared learning, collaboration and high-level networking. The FS Technology Summit will be held as a sister event to our 12th annual Fintech Summit.

Generative Artificial Intelligence (GenAI) in BusinessDr. Tathagat Varma

AI Agents at Work: UiPath, Maestro & the Future of DocumentsUiPathCommunity

Do you find yourself whispering sweet nothings to OCR engines, praying they catch that one rogue VAT number? Well, it’s time to let automation do the heavy lifting – with brains and brawn. Join us for a high-energy UiPath Community session where we crack open the vault of Document Understanding and introduce you to the future’s favorite buzzword with actual bite: Agentic AI. This isn’t your average “drag-and-drop-and-hope-it-works” demo. We’re going deep into how intelligent automation can revolutionize the way you deal with invoices – turning chaos into clarity and PDFs into productivity. From real-world use cases to live demos, we’ll show you how to move from manually verifying line items to sipping your coffee while your digital coworkers do the grunt work: 📕 Agenda: 🤖 Bots with brains: how Agentic AI takes automation from reactive to proactive 🔍 How DU handles everything from pristine PDFs to coffee-stained scans (we’ve seen it all) 🧠 The magic of context-aware AI agents who actually know what they’re doing 💥 A live walkthrough that’s part tech, part magic trick (minus the smoke and mirrors) 🗣️ Honest lessons, best practices, and “don’t do this unless you enjoy crying” warnings from the field So whether you’re an automation veteran or you still think “AI” stands for “Another Invoice,” this session will leave you laughing, learning, and ready to level up your invoice game. Don’t miss your chance to see how UiPath, DU, and Agentic AI can team up to turn your invoice nightmares into automation dreams. This session streamed live on May 07, 2025, 13:00 GMT. Join us and check out all our past and upcoming UiPath Community sessions at: 👉 https://community.uipath.com/dublin-belfast/

UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à GenèveUiPathCommunity

Nous vous convions à une nouvelle séance de la communauté UiPath en Suisse romande. Cette séance sera consacrée à un retour d'expérience de la part d'une organisation non gouvernementale basée à Genève. L'équipe en charge de la plateforme UiPath pour cette NGO nous présentera la variété des automatisations mis en oeuvre au fil des années : de la gestion des donations au support des équipes sur les terrains d'opération. Au délà des cas d'usage, cette session sera aussi l'opportunité de découvrir comment cette organisation a déployé UiPath Automation Suite et Document Understanding. Cette session a été diffusée en direct le 7 mai 2025 à 13h00 (CET). Découvrez toutes nos sessions passées et à venir de la communauté UiPath à l’adresse suivante : https://community.uipath.com/geneva/.

Canadian book publishing: Insights from the latest salary survey - Tech Forum...BookNet Canada

Join us for a presentation in partnership with the Association of Canadian Publishers (ACP) as they share results from the recently conducted Canadian Book Publishing Industry Salary Survey. This comprehensive survey provides key insights into average salaries across departments, roles, and demographic metrics. Members of ACP’s Diversity and Inclusion Committee will join us to unpack what the findings mean in the context of justice, equity, diversity, and inclusion in the industry. Results of the 2024 Canadian Book Publishing Industry Salary Survey: https://publishers.ca/wp-content/uploads/2025/04/ACP_Salary_Survey_FINAL-2.pdf Link to presentation recording and transcript: https://bnctechforum.ca/sessions/canadian-book-publishing-insights-from-the-latest-salary-survey/ Presented by BookNet Canada and the Association of Canadian Publishers on May 1, 2025 with support from the Department of Canadian Heritage.

Connect and Protect: Networks and Network SecurityVICTOR MAESTRE RAMIREZ

Reimagine How You and Your Team Work with Microsoft 365 Copilot.pptxJohn Moore

GyrusAI - Broadcasting & Streaming Applications Driven by AI and MLGyrus AI

Gyrus AI: AI/ML for Broadcasting & Streaming Gyrus is a Vision Al company developing Neural Network Accelerators and ready to deploy AI/ML Models for Video Processing and Video Analytics. Our Solutions: Intelligent Media Search Semantic & contextual search for faster, smarter content discovery. In-Scene Ad Placement AI-powered ad insertion to maximize monetization and user experience. Video Anonymization Automatically masks sensitive content to ensure privacy compliance. Vision Analytics Real-time object detection and engagement tracking. Why Gyrus AI? We help media companies streamline operations, enhance media discovery, and stay competitive in the rapidly evolving broadcasting & streaming landscape. 🚀 Ready to Transform Your Media Workflow? 🔗 Visit Us: https://gyrus.ai/ 📅 Book a Demo: https://gyrus.ai/contact 📝 Read More: https://gyrus.ai/blog/ 🔗 Follow Us: LinkedIn - https://www.linkedin.com/company/gyrusai/ Twitter/X - https://twitter.com/GyrusAI YouTube - https://www.youtube.com/channel/UCk2GzLj6xp0A6Wqix1GWSkw Facebook - https://www.facebook.com/GyrusAI

UiPath Agentic Automation: Community Developer OpportunitiesDianaGray10

Viam product demo_ Deploying and scaling AI with hardware.pdfcamilalamoratta

Building AI-powered products that interact with the physical world often means navigating complex integration challenges, especially on resource-constrained devices. You'll learn: - How Viam's platform bridges the gap between AI, data, and physical devices - A step-by-step walkthrough of computer vision running at the edge - Practical approaches to common integration hurdles - How teams are scaling hardware + software solutions together Whether you're a developer, engineering manager, or product builder, this demo will show you a faster path to creating intelligent machines and systems. Resources: - Documentation: https://on.viam.com/docs - Community: https://discord.com/invite/viam - Hands-on: https://on.viam.com/codelabs - Future Events: https://on.viam.com/updates-upcoming-events - Request personalized demo: https://on.viam.com/request-demo

MINDCTI revenue release Quarter 1 2025 PRMIND CTI

Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...Raffi Khatchadourian

Efficiency is essential to support responsiveness w.r.t. ever-growing datasets, especially for Deep Learning (DL) systems. DL frameworks have traditionally embraced deferred execution-style DL code—supporting symbolic, graph-based Deep Neural Network (DNN) computation. While scalable, such development is error-prone, non-intuitive, and difficult to debug. Consequently, more natural, imperative DL frameworks encouraging eager execution have emerged but at the expense of run-time performance. Though hybrid approaches aim for the “best of both worlds,” using them effectively requires subtle considerations to make code amenable to safe, accurate, and efficient graph execution—avoiding performance bottlenecks and semantically inequivalent results. We discuss the engineering aspects of a refactoring tool that automatically determines when it is safe and potentially advantageous to migrate imperative DL code to graph execution and vice-versa.

The Microsoft Excel Parts Presentation.pdfYvonneRoseEranista

AI You Can Trust: The Critical Role of Governance and Quality.pdfPrecisely

Unlocking Generative AI in your Web AppsMaximiliano Firtman

Slides for the session delivered at Devoxx UK 2025 - Londo. Discover how to seamlessly integrate AI LLM models into your website using cutting-edge techniques like new client-side APIs and cloud services. Learn how to execute AI models in the front-end without incurring cloud fees by leveraging Chrome's Gemini Nano model using the window.ai inference API, or utilizing WebNN, WebGPU, and WebAssembly for open-source models. This session dives into API integration, token management, secure prompting, and practical demos to get you started with AI on the web. Unlock the power of AI on the web while having fun along the way!

Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...Markus Eisele

We keep hearing that “integration” is old news, with modern architectures and platforms promising frictionless connectivity. So, is enterprise integration really dead? Not exactly! In this session, we’ll talk about how AI-infused applications and tool-calling agents are redefining the concept of integration, especially when combined with the power of Apache Camel. We will discuss the the role of enterprise integration in an era where Large Language Models (LLMs) and agent-driven automation can interpret business needs, handle routing, and invoke Camel endpoints with minimal developer intervention. You will see how these AI-enabled systems help weave business data, applications, and services together giving us flexibility and freeing us from hardcoding boilerplate of integration flows. You’ll walk away with: An updated perspective on the future of “integration” in a world driven by AI, LLMs, and intelligent agents. Real-world examples of how tool-calling functionality can transform Camel routes into dynamic, adaptive workflows. Code examples how to merge AI capabilities with Apache Camel to deliver flexible, event-driven architectures at scale. Roadmap strategies for integrating LLM-powered agents into your enterprise, orchestrating services that previously demanded complex, rigid solutions. Join us to see why rumours of integration’s relevancy have been greatly exaggerated—and see first hand how Camel, powered by AI, is quietly reinventing how we connect the enterprise.

Play It Safe: Manage Security Risks - Google CertificateVICTOR MAESTRE RAMIREZ

TrsLabs Consultants - DeFi, WEb3, Token ListingTrs Labs

Webinar - Top 5 Backup Mistakes MSPs and Businesses Make .pptxMSP360

Data loss can be devastating — especially when you discover it while trying to recover. All too often, it happens due to mistakes in your backup strategy. Whether you work for an MSP or within an organization, your company is susceptible to common backup mistakes that leave data vulnerable, productivity in question, and compliance at risk. Join 4-time Microsoft MVP Nick Cavalancia as he breaks down the top five backup mistakes businesses and MSPs make—and, more importantly, explains how to prevent them.

How analogue intelligence complements AIPaul Rowe

Financial Services Technology Summit 2025Ray Bugg

Generative Artificial Intelligence (GenAI) in BusinessDr. Tathagat Varma

AI Agents at Work: UiPath, Maestro & the Future of DocumentsUiPathCommunity

UiPath Automation Suite – Cas d'usage d'une NGO internationale basée à GenèveUiPathCommunity

Canadian book publishing: Insights from the latest salary survey - Tech Forum...BookNet Canada

Connect and Protect: Networks and Network SecurityVICTOR MAESTRE RAMIREZ

Reimagine How You and Your Team Work with Microsoft 365 Copilot.pptxJohn Moore

GyrusAI - Broadcasting & Streaming Applications Driven by AI and MLGyrus AI

UiPath Agentic Automation: Community Developer OpportunitiesDianaGray10

Viam product demo_ Deploying and scaling AI with hardware.pdfcamilalamoratta

MINDCTI revenue release Quarter 1 2025 PRMIND CTI

Hybridize Functions: A Tool for Automatically Refactoring Imperative Deep Lea...Raffi Khatchadourian

The Microsoft Excel Parts Presentation.pdfYvonneRoseEranista

AI You Can Trust: The Critical Role of Governance and Quality.pdfPrecisely

Unlocking Generative AI in your Web AppsMaximiliano Firtman

Enterprise Integration Is Dead! Long Live AI-Driven Integration with Apache C...Markus Eisele

Play It Safe: Manage Security Risks - Google CertificateVICTOR MAESTRE RAMIREZ

TrsLabs Consultants - DeFi, WEb3, Token ListingTrs Labs

Webinar - Top 5 Backup Mistakes MSPs and Businesses Make .pptxMSP360

ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers

1. ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers Daniel Caixinha, Pradeeban Kathiravelu, Lu s Veigaıı Presented by: André Negrão INESC-ID Lisboa / Instituto Superior Técnico Universidade de Lisboa, Portugal The 15th IEEE International Symposium on Network Computing and Applications (NCA 2016) November 1st , 2016. Cambridge, MA.

2. 2 Introduction ● Differentiated SLAs for data center tenants. ● Lack of guarantees in bandwidth. ● Shared bandwidth → Unpredictable performance. ● Software-Defined Networking (SDN) offers unified and enhanced control to the network. – From higher levels.

3. 3 Motivation ● Virtual Network Embedding (VNE) aims to completely virtualize the network. – Performance isolation among tenants in the network level. – Major challenge in network virtualization. ● Can we leverage SDN for a better VNE approach?

4. 4 Contributions ● A practical solution for the virtual network embedding problem. ● High consolidation within the placement of virtual networks ● High utilization of physical resources – Servers and network.

5. 5 ViTeNA ● A Virtual Network Embedding Algorithm – For Multi-Tenant Data Centers – Leveraging SDN. ● Tenants’ bandwidth requirements – Enforced through virtual networks.

6. 6 Deployment Landscape ● Reduce number of hops

7. 7 Deployment Landscape ● Reduce number of hops – Increase locality. – Reduce communication delays.

8. 8 ViTeNA Architecture ● Tenant demands as an XML file. ● Allocation based on the network state.

9. 9 Implementation ● Floodlight 1.1 as the OpenFlow controller. ● Mininet 2.2.1 and Open vSwitch 2.3.1 to emulate the data center.

10. 10 Evaluation Deployment ● A computer with Intel ® Quad-Core i7 870 @ 2.93 GHz processor – 12 GB DDR3 @ 1333 MHz RAM – 450 GB Serial ATA @ 7200 rpm hard disk – Ubuntu 14.04.3 LTS (Linux Kernel 3.13.0). ● Stop an experiment when the controller returns false to an experiment. ● Experiments run 1000 times.

11. 11 Emulated System ● A tree topology (depth = 3; fanout = 5) – with 125 servers – 31 switches and 155 links ● A fat-tree topology – factor k = 32, i.e. switches consist of 32 ports – with 128 servers – 160 switches and 384 links

12. 12 Evaluation Approach ● Scalability ● High consolidation ● High resource utilization. ● Bandwidth guarantees in a work-conservative system

13. 13 Scalability to data center scale ● Allocation time with tree topology.

14. 14 Scalability to data center scale ● Allocation time with fat-tree topology.

15. 15 High Consolidation ● Allocate the VMs of a virtual network as close as possible. ● Tree topology ● Fat-Tree topology

16. 16 High Resource Utilization ● Server and network utilization (%) – For tree and fat-tree.

17. 17 Conclusion ● Conclusions – ViTeNA addresses the unpredictable performance of the applications. ● Using the abstraction of virtual networks. – Evaluations confirm ● low execution time ● high consolidation on the virtual network allocation. ● high data center resource utilization. ● Future Work – Reliability and isolation guarantees

18. 18 Thank you! ● Questions? [email protected]

ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers

Recommended

More Related Content

What's hot (16)

Viewers also liked (20)

Similar to ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers (20)

More from Pradeeban Kathiravelu, Ph.D. (17)

Recently uploaded (20)

ViTeNA: An SDN-Based Virtual Network Embedding Algorithm for Multi-Tenant Data Centers