Improving Network I/O Virtualization for Cloud Computing.

Slides:



Advertisements
Similar presentations
Towards High-Availability for IP Telephony using Virtual Machines Devdutt Patnaik, Ashish Bijlani and Vishal K Singh.
Advertisements

Abstract Shortest distance query is a fundamental operation in large-scale networks. Many existing methods in the literature take a landmark embedding.
Virtualization for Cloud Computing
CloudMoV: Cloud-based Mobile Social TV
Optimizing Cloud Resources for Delivering IPTV Services Through Virtualization.
Abstract Cloud data center management is a key problem due to the numerous and heterogeneous strategies that can be applied, ranging from the VM placement.
Energy-Optimum Throughput and Carrier Sensing Rate in CSMA-Based Wireless Networks.
Annotating Search Results from Web Databases. Abstract An increasing number of databases have become web accessible through HTML form-based search interfaces.
Abstract Load balancing in the cloud computing environment has an important impact on the performance. Good load balancing makes cloud computing more.
A Secure Protocol for Spontaneous Wireless Ad Hoc Networks Creation.
Back-Pressure-Based Packet-by-Packet Adaptive Routing in Communication Networks.
Personalized QoS-Aware Web Service Recommendation and Visualization.
Abstract Provable data possession (PDP) is a probabilistic proof technique for cloud service providers (CSPs) to prove the clients' data integrity without.
Secure Encounter-based Mobile Social Networks: Requirements, Designs, and Tradeoffs.
Minimum Cost Blocking Problem in Multi-path Wireless Routing Protocols.
Cross-Domain Privacy-Preserving Cooperative Firewall Optimization.
A Survey of Mobile Cloud Computing Application Models
NICE :Network Intrusion Detection and Countermeasure Selection in Virtual Network Systems.
Dynamic Resource Allocation Using Virtual Machines for Cloud Computing Environment.
Fast Nearest Neighbor Search with Keywords. Abstract Conventional spatial queries, such as range search and nearest neighbor retrieval, involve only conditions.
Security Evaluation of Pattern Classifiers under Attack.
Privacy-Preserving Public Auditing for Secure Cloud Storage
BestPeer++: A Peer-to-Peer Based Large-Scale Data Processing Platform.
m-Privacy for Collaborative Data Publishing
PACK: Prediction-Based Cloud Bandwidth and Cost Reduction System
EAACK—A Secure Intrusion-Detection System for MANETs
Combining Cryptographic Primitives to Prevent Jamming Attacks in Wireless Networks.
Optimal Client-Server Assignment for Internet Distributed Systems.
Protecting Sensitive Labels in Social Network Data Anonymization.
Identity-Based Secure Distributed Data Storage Schemes.
Enabling Dynamic Data and Indirect Mutual Trust for Cloud Computing Storage Systems.
Hiding in the Mobile Crowd: Location Privacy through Collaboration.
Cooperative Caching for Efficient Data Access in Disruption Tolerant Networks.
Accuracy-Constrained Privacy-Preserving Access Control Mechanism for Relational Data.
Identity-Based Distributed Provable Data Possession in Multi-Cloud Storage.
Content Sharing over Smartphone-Based Delay- Tolerant Networks.
Abstract Link error and malicious packet dropping are two sources for packet losses in multi-hop wireless ad hoc network. In this paper, while observing.
A System for Denial-of- Service Attack Detection Based on Multivariate Correlation Analysis.
Modeling the Pairwise Key Predistribution Scheme in the Presence of Unreliable Links.
Privacy Preserving Delegated Access Control in Public Clouds.
Scalable Distributed Service Integrity Attestation for Software-as-a-Service Clouds.
Anomaly Detection via Online Over-Sampling Principal Component Analysis.
A Highly Scalable Key Pre- Distribution Scheme for Wireless Sensor Networks.
Towards Online Shortest Path Computation. Abstract The online shortest path problem aims at computing the shortest path based on live traffic circumstances.
Facilitating Document Annotation using Content and Querying Value.
Traffic Pattern-Based Content Leakage Detection for Trusted Content Delivery Networks.
Privacy Preserving Back- Propagation Neural Network Learning Made Practical with Cloud Computing.
Participatory Privacy: Enabling Privacy in Participatory Sensing
Abstract We propose two novel energy-aware routing algorithms for wireless ad hoc networks, called reliable minimum energy cost routing (RMECR) and reliable.
Supporting Privacy Protection in Personalized Web Search.
Twitsper: Tweeting Privately. Abstract Although online social networks provide some form of privacy controls to protect a user's shared content from other.
Full and Para Virtualization
m-Privacy for Collaborative Data Publishing
Attribute-Based Encryption With Verifiable Outsourced Decryption.
A Scalable Two-Phase Top-Down Specialization Approach for Data Anonymization Using MapReduce on Cloud.
A New Algorithm for Inferring User Search Goals with Feedback Sessions.
Harnessing the Cloud for Securely Outsourcing Large- Scale Systems of Linear Equations.
Securing Broker-Less Publish/Subscribe Systems Using Identity-Based Encryption.
Dynamic Control of Coding for Progressive Packet Arrivals in DTNs.
Privacy-Preserving and Content-Protecting Location Based Queries.
Mona: Secure Multi-Owner Data Sharing for Dynamic Groups in the Cloud.
Whole Test Suite Generation. Abstract Not all bugs lead to program crashes, and not always is there a formal specification to check the correctness of.
Distributed Processing of Probabilistic Top-k Queries in Wireless Sensor Networks.
Load Rebalancing for Distributed File Systems in Clouds.
Facilitating Document Annotation Using Content and Querying Value.
Fast Transmission to Remote Cooperative Groups: A New Key Management Paradigm.
Dynamic Query Forms for Database Queries. Abstract Modern scientific databases and web databases maintain large and heterogeneous data. These real-world.
Virtualization for Cloud Computing
Towards Scalable Traffic Management in Cloud Data Centers
Presentation transcript:

Improving Network I/O Virtualization for Cloud Computing

Abstract Virtualization is a key technology to enable cloud computing. It enhances resource availability and offers high flexibility and cost effectiveness. However, the driver domain- based model for network I/O virtualization exhibit poor networking performance. In this paper, we evaluate the networking performance of VMs based on Xen. The I/O channel transferring packets between the driver domain and the VMs is shown to be the bottleneck. To overcome this limitation, we proposed a packet aggregation-based mechanism to transfer packets from the driver domain to the VM. Experimental performance evaluation showed that our proposal allows the VMs throughput to scale up at line rates. Furthermore, we studied the impact of the aggregation on the packets delay. For this purpose, we proposed a system modeling based on queuing theory. The proposed model allowed us to dynamically tune the aggregation mechanism to achieve the best tradeoff between the packets delay and throughput. The proposed I/O virtualization model henceforth satisfies the infrastructure providers to offer cloud computing services.

Existing System CLOUD computing is a novel paradigm that provides infrastructure, platform, and software as a service. A cloud platform can be either virtualized or not. Virtualizing the cloud platform increases the resources availability and the flexibility of their management (allocation, migration, etc.). It also reduces the cost through hardware multi¬plexing and helps energy saving. Virtualization is then a key enabling technology of cloud computing. System virtualization refers to the software and hardware techni¬ques that allow partitioning one physical machine into multiple virtual instances that run concurrently and share the underlying physical resources and devices. The virtual machine monitor (VMM), also called hyper¬visor, is a software layer that presents abstractions of the underlying physical resources. It manages the virtual machines (VM) and allows them to share the physical resources especially the network device. Since the hypervi¬sors have been first implemented to provide server virtualization to support compute intensive applications, the I/O virtualization has not been implemented with network intensive applications requirements in mind (high throughput, low latency, etc.). Then, current hypervisors exhibit poor network performance due to the additional overhead incurred by the virtualization layer.

Architecture Diagram

System Specification HARDWARE REQUIREMENTS Processor : Intel Pentium IV Ram : 512 MB Hard Disk : 80 GB HDD SOFTWARE REQUIREMENTS Operating System : Windows XP / Windows 7 FrontEnd : Java BackEnd : MySQL 5

C ONCLUSION Virtualization is becoming a key technology to enable deploying efficient and cost-effective cloud computing platforms. However, current network I/O virtualization models still suffer from performance and scalability limitations. In this paper, we showed that VMs exhibit poor reception, transmission, and forwarding throughputs. Through profiling the physical resources, we proved that the I/O channels mechanism prevents the guests from achieving line rate throughput. I/O communication be¬tween the driver domain and the guests is shown to be the most expensive in memory transactions. Memory transac¬tions are limited by the long memory latency that represents a hardware bottleneck of the system. To over-come this limitation, we proposed a packet aggregation mechanism that allows to transfer containers of packets at once. This results in a more efficient I/O communication. Experimental evaluation shows that the proposed packet aggregation mechanism significantly improves the net¬working performance.

THANK YOU