SYNOPSIS ON Going Back and Forth: Efficient Multi deployment and Multi snapshotting on Clouds




2. 3 4 5 6 7 8 8 9 11 11 12 15 16 17 References . High Level Design 6. 4. 3. Problem Definition Abstract Aim & Objective Literature Review 4.3 Cloud Infrastructure 4. Plan of Action Software & Hardware Requirements Page No.Table of Contents Title 1. Scope 5.1 Existing System 4.1 Advantage 6.1 Data flow diagram 6.4 Application State 5.2 Disadvantages 4. 8.2 System Architecture 7.

for check pointing or off-line migration) to another cluster or cloud . . Problem Definition To provide the basic functionalities for the use of Virtual Machine’s (VM’s) over the cloud given as: • Multideployment : The operation of Infrastructure As A Service (IaaS) is the need to deploy a large number of VMs on many nodes of a datacentre at the same time..1. starting from a set of VM images previously stored in a persistent fashion • Multisnapshotting : Many VM images that were locally modified need to be concurrently transferred to stable storage with the purpose of capturing the VM’s state for later use (e.g.

it is important to enable efficient concurrent deployment and snapshotting that are at the same time hypervisor independent and ensure a maximum compatibility with different configurations. It is based on a lazy transfer scheme coupled with object versioning that handles snapshotting transparently in a hypervisor-independent fashion.Abstract Infrastructure as a Service (IaaS) cloud computing has transform the way we think of acquiring resources by introducing a simple change: allowing users to lease computational resources from the cloud provider’s datacenter for a short time by deploying virtual machines (VMs) on these resources. Once the VM instances 1are deployed. . such as suspend-resume and migration. With datacenters growing rapidly and configurations becoming heterogeneous. This paper addresses these challenges by proposing a virtual file system specifically optimized for virtual machine image storage.2. another challenge is to simultaneously take a snapshot of many images and transfer them to persistent storage to support management tasks. This new model raises new challenges in the design and development of IaaS middleware. One of those challenges is the need to deploy a large number (hundreds or even thousands) of VM instances simultaneously. ensuring high portability for different configurations.

One resource can be available to the all the users .2. They’re focused on coming up with solutions that serve customer requirements today and anticipate future needs. restart & suspend operation. . we are creating cloud infrastructure which allows users to lease computational resources from cloud provider. Aim and Objectives In this anticipating their requirements we provide them without confliction. Our aim is to create and implement load balancing mechanisms. A focus on the use of reusable frameworks to provide cost and times benefits. for managing numbers of users at the same time and within the context of time slices. We want to reduce the contention on current system & allow maximum number of user to access VM’s with quick resume.

Literature Survey In recent years. Leasing of computation time is accomplished by allowing users to deploy virtual machines (VMs) on the datacentre’s resources[1]. such approaches generate a large number of files and interdependencies among them. which limits the ability to easily migrate VMs among different hypervisors. since it enables users to expand or shrink their resources according to their computational needs. Infrastructure as a Service (IaaS) cloud computing has emerged as a viable alternative to the acquisition and management of physical resources. When taking frequent snapshots for a large number of VMs. not counting the time to boot the operating system itself. We refer to this pattern as multideployment. Furthermore. A typical deployment consists of hundreds or even thousands of such images. with growing datacentre trends and tendencies to federate clouds.g. a process that can take tens of minutes to hours.[1] Once the VM instances are running. For example. Therefore. Custom image formats are not standardized and can be used with specific hypervisors only. This emerging model leads to new challenges relating to the design and development of IaaS systems. a similar challenge applies to snapshotting the deployment: many VM images that were locally modified need to be concurrently transferred to stable storage with the purpose of capturing the VM state for later use (e. this pattern occurs when the user wants to deploy a virtual cluster that executes a distributed application or a set of environments to support a workflow. IaaS leasing is equivalent to purchasing dedicated hardware but without the long-term commitment and cost. Conventional deployment techniques broadcast the images to the nodes before starting the VM instances. Conventional snapshotting techniques rely on custom VM image file formats to store only incremental differences in a new file that depends on the original VM image as the backing file.. users can lease storage and computation time from large datacentre’s. Such a large deployment of many VMs at once can take a long time. One of the commonly occurring patterns in the operation of IaaS is the need to deploy a large number of VMs on many nodes of a datacentre at the same time. starting from a set of VM images previously stored in a persistent fashion. Since the user has complete control over the configuration of the VMs using on-demand deployments.4. for check pointing or off-line migration to another cluster or cloud). This problem is particularly acute for VM images used in scientific computing where image sizes are large (from a few gigabytes up to more than 10 GB). With IaaS. multisnapshotting must be handled in a transparent and portable . This can make the response time of the IaaS installation much longer than acceptable and erase the ondemand benefits of cloud computing. We refer to this pattern as multisnapshotting. configurations are becoming more and more heterogeneous. by using external resources to complement their local resource base. The on-demand nature of IaaS is critical to making such leases attractive. which are difficult to manage and which interfere with the ease-of-use rationale behind clouds.

fashion that hides the interdependencies of incremental differences and exposes standalone VM images. while keeping maximum portability among different hypervisor configurations. Cloud Computing is a “buzz word” around a wide variety of aspects such as deployment. load balancing. Moreover. 4. we investigate them in conjunction. these patterns may also generate high network traffic that interferes with the execution of applications on leased resources and generates high utilization costs for the user. provisioning.BLOBs that can grow to TB) in very large-scale distributed systems while maintaining a very high data throughput for highly concurrent. storage space. Our proposal offers a good balance between performance.1 Existing system The huge computational potential offered by large distributed systems is hindered by poor data sharing scalability. One such requirement is the need to efficiently cope with massive unstructured data (organized as huge sequences of bytes . Network traffic consumption also very high due to non concentrating on application status. In addition to incurring significant delays and raising manageability issues. We addressed several major requirements related to these challenges. 4. Disk storage (cheap hard-drives with . Clouds have been defined just as virtualized hardware and software plus the previous monitoring and provisioning technologies. Since the patterns are complementary. It is not possible to build a scalable. and data and processing outsourcing. This paper proposes a distributed virtual file system specifically optimized for both the multideployment and multisnapshotting patterns. high-performance distributed data-storage service that facilitates data sharing at large scale. while handling snapshotting transparently and exposing standalone.2 Disadvantages To give less performance and storage space. fine-grain data accesses [1]. 4.3 Cloud infrastructure IaaS platforms are typically built on top of clusters made out of loosely-coupled commodity hardware that minimizes per unit cost and favours low power over maximum speed. raw image files (understood by most hypervisors) to the outside[6]. The role of virtualization in Clouds is also emphasized by identifying it as a key component. and network traffic consumption.

in the most general case . in terms of both hardware and software. which is unacceptable for a single one. The machines are configured with proper virtualization technology. a dedicated repository is deployed either as centralized or as distributed storage service running on dedicated storage nodes. Thus. While several methods have been established in the virtualization community to capture the state of a running VM (CPU registers. download. In order to provide persistent storage. Even so.).point-in-time deployment checkpoint. Model 2 can further be simplified such that the VM state is represented only by the virtual disk attached to it (Model 3). Any in-transit network traffic is discarded. and the like does not need to be saved. the general case is usually simplified such that the application state is reduced to the sum of states of the VM instances .). state of devices.4 Application state The state of the VM deployment is defined at each moment in time by two main components: the state of each of the VM instances and the state of the communication channels between them (opened sockets. For example. virtual communication channels and resend lost information[5]. under the assumption that a fault-tolerant networking protocol is used that is able to re. RAM. etc. In order to avoid this issue. delete. saving 2 GB of RAM for 1. in-transit network packets. which is used to store only minimal information about the state. the issue of capturing the global state of the communication channels is difficult and still an open problem. . for VM instances that need large amounts of memory. and so forth. since the contents of RAM. 4. The repository is responsible for storing the VM images persistently in a reliable fashion and provides the means for users to manipulate them: upload. the necessary storage space can explode to huge sizes. such as configuration files that describe the environment and temporary files that were generated by the application. and (ii) Portability since the VM can be restored on another host without having to worry about restoring the state of hardware devices that are not supported or are in. With the recent explosion in cloud computing demands. while the machines are interconnected with standard Ethernet links. Such an approach has two important practical benefits: (i) Huge reductions in the size of the state. there is an acute need for scalable storage.compatible between different hypervisors. Therefore.[5] This information is then later used to reboot and reinitialize the software stack running inside the VM instance.000 VMs consumes 2 TB of space. CPU registers. etc. saving the application state implies saving both the state of all VM instances and the state of all active communication channels among them.capacities in the order of several hundred GB) is attached to each machine. such that they are able to host the VMs.

Scope A distributed virtual file system specifically optimized for both the multi deployment and multi snapshotting patterns. Our proposal offers a good balance between performance. we investigate them in conjunction.1 ADVANTAGE A good balance between performance. and network traffic consumption. raw image files. we describe an implementation on top of Blob Seer. 5. while handling snapshotting transparently and exposing standalone. and network traffic consumption. To illustrate this point. We introduce a series of design principles that optimize multi deployment and multi snapshotting patterns and describe how our design can be integrated with IaaS infrastructures. . We show how to realize these design principles by building a virtual file system that leverages versioningbased distributed storage services. storage space. raw image files (understood by most hypervisors) to the outside[4].5. storage space. Since the patterns are complementary. a versioning storage service specifically designed for high throughput under concurrency. while handling snapshotting transparently and exposing standalone.


and snapshotting individual VM instances or the whole set. and the mirroring module. when to create a new image clone (CLONE) [2]. dynamically adding or removing compute nodes from that set. distributed application needs to be debugged. CLONE and COMMIT can also be exposed by the cloud middleware at the user level through the control API for fine-grained control over snapshotting. Every uploaded image is automatically striped. For ex-ample. This approach enables snapshotting to be leveraged in interesting ways. The cloud client has direct access to the storage service and is allowed to upload and download images from it. Each compute node runs a hypervisor [1] that is responsible for running the VMs. is performed in the following fashion. while the elements that are part of our proposal are highlighted by a darker background.suing each mirroring module a COMMIT to its corresponding clone. including deploying an image on a set of compute nodes. The cloud middleware interacts directly with both the hypervisor. The reads and writes of the hypervisor are trapped by the mirroring module [1]. A global snapshot of the whole application. Furthermore. A distributed versioning storage service that supports cloning and shadowing is deployed on the compute nodes and consolidates parts of their local disks into a common storage pool. The first time the snap-shot is taken. Running the application repeatedly and waiting for it to reach the point where the bug happens might be prohibitively expensive. Once a clone is created for each VM instance. The cloud middleware [2] in turn coordinates the compute nodes to achieve the afore-mentioned management tasks. CLONE is broadcast to all mirroring modules. Since all image snapshots are independent entities.7. which involves taking a snapshot of all VM instances in parallel. the cloud client interacts with the cloud middleware through a control API that enables a variety of management tasks. The typical elements found in the cloud are illustrated with a light background. let’s assume a scenario where a complex. which is responsible for ondemand mirroring and snapshotting and relies on both the local disk and the distributed versioning storage service to do so. the application right before the bug happens. telling it what image to mirror from the repository. However. and when to persistently store its local modifications (COMMIT) [2]. followed by COMMIT. fully independent VM image that is globally accessible through the storage service and can be deployed on other compute nodes or manipulated by the client. Both CLONE and COMMIT are control primitives that result in the generation of a new. telling it when to start and stop VMs. subsequent global snapshots are performed by is.Plan of Action The simplified architecture of a cloud that integrates our approach is depicted in Figure 1. they can be either collectively or independently analyzed and modified in an attempt to .

the approach can continue iteratively until a fix is found. Such an approach is highly useful in practice at large scale because complex synchronization bugs [3] tend to appear only in large deployments and are usually not triggered during the test phase. If the attempt was not successful. Once this fix is made. the application can safely resume from the point where it left. which is usually performed at smaller scale.fix the bug. .

1 Hardware requirement : CPU type Clock speed Ram size Hard disk capacity : Dual Core : 2.8.6) : MS-Access : Ms-Office .65 GHz : 2 GB : 40 GB 8.2 Software requirement : Operating System: Windows Language Back End Documentation : JAVA(JDK-1. Software And Hardware Requirements 8.

A.A. Science clouds: Early experiences in cloud computing for scientific applications.pdf [3]M. 2008. D. Keahey and T. Rabkin. Joseph. the MIT Press ISBNhttp://hal.nimbusproject.Ameet Talwalkar (2012) Foundations of Machine Learning. Richard. Patterson.A. Commun.USENIX Association.pdf [2]Mehryar B. Fox. A view of cloud computing. [6] Hypervisor Alternative: http://siliconangle. Stoica. R. ACM. . In HPDC’09: Proceedings of the 18th ACM International Symposium on High Performance Distributed Computing. In CCA’08: Proceedings of the 1st Conference on Cloud Computing and Its Applications.adaptive deployment of remote executions.53:50 [4]327. Atlanta.References : [1]Going Back and Forth: Efficient Multi deployment and Multi snapshotting on Clouds: www. R. Freeman. G. Armbrust. A. GA. Afshin Rostamizadeh. Huard. Katz.4. and O. pages 91 [5]K. and M. Lee. Taktuk.inria. Claudel. Konwinski.I. Zaharia. G.

left aligned. 2.Guidelines 1. should be at the center of bottom of each page. Headings in the chapters should have size Times New Roman 14. 3. center.25” 5. 6.5 Paragraph spacing – 2 lines 4. Written matter – Times New Roman 12 (justified) Line spacing – 1. Right : 1. Bottom : 1. Page nos.5” . The paper should be A4 size with margins: Top: 1”. bold. Left : 1. Subtitles Times New Roman 12. Number of copies -.2” . Report should be spiral bound with white plastic cover pages only. bold.02 (Project guide + Project Coordinator) .

.7. Seminar report should be as per the guidelines only.

Sign up to vote on this title
UsefulNot useful