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




Plan of Action Software & Hardware Requirements Page No. Scope 5. 8. 3.1 Data flow diagram 6.2 Disadvantages 4. 3 4 5 6 7 8 8 9 11 11 12 15 16 17 References . Problem Definition Abstract Aim & Objective Literature Review 4.Table of Contents Title 1.4 Application State 5. High Level Design 6.2 System Architecture 7.1 Advantage 6.3 Cloud Infrastructure 4.1 Existing System 4. 4. 2.

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. for check pointing or off-line migration) to another cluster or cloud . 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.1..g. .

With datacenters growing rapidly and configurations becoming heterogeneous. ensuring high portability for different configurations. 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. One of those challenges is the need to deploy a large number (hundreds or even thousands) of VM instances simultaneously. This paper addresses these challenges by proposing a virtual file system specifically optimized for virtual machine image storage. Once the VM instances 1are deployed. . such as suspend-resume and migration.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. This new model raises new challenges in the design and development of IaaS middleware. It is based on a lazy transfer scheme coupled with object versioning that handles snapshotting transparently in a hypervisor-independent fashion.2. another challenge is to simultaneously take a snapshot of many images and transfer them to persistent storage to support management tasks.

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

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

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

while the machines are interconnected with standard Ethernet links. such as configuration files that describe the environment and temporary files that were generated by the application. RAM. there is an acute need for scalable communication channels and resend lost information[5]. in-transit network packets. for VM instances that need large amounts of memory.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. saving 2 GB of RAM for 1.000 VMs consumes 2 TB of space. With the recent explosion in cloud computing demands. virtual topology. Any in-transit network traffic is discarded. the general case is usually simplified such that the application state is reduced to the sum of states of the VM instances . In order to avoid this issue. In order to provide persistent storage. . which is unacceptable for a single one. in the most general case .point-in-time deployment checkpoint. since the contents of RAM. a dedicated repository is deployed either as centralized or as distributed storage service running on dedicated storage nodes. For example.[5] This information is then later used to reboot and reinitialize the software stack running inside the VM instance. the issue of capturing the global state of the communication channels is difficult and still an open problem. The machines are configured with proper virtualization technology. in terms of both hardware and software. Such an approach has two important practical benefits: (i) Huge reductions in the size of the state. download. Even so. and the like does not need to be saved. which is used to store only minimal information about the state. Thus.). 4. CPU registers. Model 2 can further be simplified such that the VM state is represented only by the virtual disk attached to it (Model 3). and so forth. Therefore.compatible between different hypervisors. state of devices. saving the application state implies saving both the state of all VM instances and the state of all active communication channels among them.). under the assumption that a fault-tolerant networking protocol is used that is able to re. such that they are able to host the VMs. 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.capacities in the order of several hundred GB) is attached to each machine. delete. etc. While several methods have been established in the virtualization community to capture the state of a running VM (CPU registers. etc. the necessary storage space can explode to huge sizes. The repository is responsible for storing the VM images persistently in a reliable fashion and provides the means for users to manipulate them: upload.

Our proposal offers a good balance between performance. storage space.1 ADVANTAGE A good balance between performance. raw image files. and network traffic consumption. while handling snapshotting transparently and exposing standalone. To illustrate this point. . Since the patterns are complementary. storage space. 5. and network traffic consumption. 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. while handling snapshotting transparently and exposing standalone. raw image files (understood by most hypervisors) to the outside[4]. a versioning storage service specifically designed for high throughput under concurrency.5. we investigate them in conjunction. Scope A distributed virtual file system specifically optimized for both the multi deployment and multi snapshotting patterns. we describe an implementation on top of Blob Seer.


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

Once this fix is made.fix the bug. . If the attempt was not successful. which is usually performed at smaller scale. 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. the application can safely resume from the point where it left.

2 Software requirement : Operating System: Windows Language Back End Documentation : JAVA(JDK-1.1 Hardware requirement : CPU type Clock speed Ram size Hard disk capacity : Dual Core : 2. Software And Hardware Requirements 8.65 GHz : 2 GB : 40 GB 8.6) : MS-Access : Ms-Office .8.

R. Patterson. Rabkin. A. Konwinski. Fox. [6] Hypervisor Alternative: http://siliconangle. Huard. R. Griffith. Armbrust.adaptive deployment of remote executions. the MIT Press ISBNhttp://hal.A. In HPDC’09: Proceedings of the 18th ACM International Symposium on High Performance Distributed Computing. Richard. Taktuk.References : [1]Going Back and Forth: Efficient Multi deployment and Multi snapshotting on Clouds: www. Commun.inria. G. D.pdf [3]M. Claudel. 2000. Atlanta. Joseph. Science clouds: Early experiences in cloud computing for scientific applications. Freeman. Lee.A. . Zaharia. Afshin Rostamizadeh.Ameet Talwalkar (2012) Foundations of Machine Learning. Keahey and T.USENIX Association. GA. 2008. ACM. and O. In CCA’08: Proceedings of the 1st Conference on Cloud Computing and Its Applications. B. pages 91 [5]K.pdf [2]Mehryar G. and M. A view of cloud computing.nimbusproject. [4]327. A.I.

bold. 3. Left : 1. should be at the center of bottom of each page. Report should be spiral bound with white plastic cover pages only. Bottom : 1.Guidelines 1.5 Paragraph spacing – 2 lines 4.5” . left aligned. 6.02 (Project guide + Project Coordinator) . center. Subtitles Times New Roman 12. Page nos.25” 5. The paper should be A4 size with margins: Top: 1”. Number of copies -. Headings in the chapters should have size Times New Roman 14. Written matter – Times New Roman 12 (justified) Line spacing – 1. 2.2” . Right : 1. bold.

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

Sign up to vote on this title
UsefulNot useful