Theses and Dissertations

Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/1

Browse

Subject: search.filters.subject.Grid computing

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Policy based resource allocation on infrastructure as a service cloud
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2011) Vora, Dhairya; Chaudhary, Sanjay; Bise, Minal
    Cloud computing refers to the provision of computational resources on demand. Resource allocation is an important aspect in cloud computing. Cloud user asks for resources in terms of a lease. Lease stores the information about required resources and the time at which these resources are required. Cloud provider accepts the lease if it can provide guarantee for assigning resources at asked time to the cloud user. Better scheduling algorithm can accept more number of leases and hence give better resource utilization. Cloud provides four types of leases: immediate, advance reservation, best effort and deadline sensitive. Immediate allocation policy accepts the lease if resources are available, else it rejects the lease. Advance reservation policy accepts the lease if resources are available at the asked time, else it rejects the lease. Best effort allocation policy accepts the lease as soon as the resources are available. Deadline sensitive leases have parameters like required resources, startTime, endTime and duration. Scheduler can accept such lease by providing required resources for the asked duration of time between given startTime and endTime. Haizea is a resource lease manager which handles the scheduling of the lease. Proposed algorithm extends the current scheduling algorithm of Haizea for deadline sensitive type of leases. Aim of the thesis is to improve resource utilization by extending the current scheduling algorithms of Haizea. Proposed scheduling algorithm accepts more number of leases by dividing a deadline sensitive lease into multiple slots and by back filling already.
  • Thumbnail Image
    ItemOpen Access
    Checkpointing and recovery mechanism in grid
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2007) Mehta, Janki; Chaudhary, Sanjay
    Grid is a collection of distributed computing resources that performs tasks in co-ordination to achieve high-end computational capabilities. Grid Computing is a collective computing of a given task by breaking it into sub-tasks. Each sub-task could be large and run for several hours or days on a number of grid nodes. If a sub-task fails to complete even on a single site, all the computations need to be done again. In scalable distributed systems, individual component failures usually does not result in failure of the entire system. However, a single failure may crash an entire parallel application. Grid is dynamic in nature. Since the probability of a single component failure rises rapidly with the number of components in the system, as system grows in size, efficient recovery mechanism is most important for highly parallel mission critical and long running applications of grid environment. This thesis addresses a recovery mechanism using checkpoints to recover from Grid Service failure resulting in task or transaction failure in Computational Grid and Data Grid which will prevent computations to be restarted from scratch. Grid Service may fail as a result of hardware or software fault. A checkpoint is a point in time snapshot of a grid node in which its state information is stored. It will help in reducing the crash recovery time. This work helps in preserving two main objectives of grid namely optimal resource utilization and speedy computations which can be achieved by using resources in a better way for improving performance of system rather than engaging them in tasks like rollbacks resulting from cascading aborts. The saved state using checkpoints can also be used for job migration using job schedulers of grid on occurrence of critical failures like Operating System failure. Experiments conducted provide integration of proposed mechanism with standard grid Web Service Resource Framework and will aid in future development work.
  • Thumbnail Image
    ItemOpen Access
    Model for grid service instance migration
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2005) Shah, Dhaval Kiritbhai; Chaudhary, Sanjay
    Grid computing, emerging as a new paradigm for next-generation computing, enables the sharing, selection, and aggregation of distributed resources for solving large-scale problems in science, engineering, and commerce. The resources in the Grid are heterogeneous and geographically distributed. The resources in the Grid are dynamic in nature. Resource owners are at discretion to submit/donate the resources in to the Grid environment. A Web Services is a network-enabled entity accessed using the standard web protocols. Web Services are basically stateless in nature and there is a need to maintain state across the transactions based on Web Services. Grid Services is an extension of Web Services in a Grid environment having statefulness as a key feature. State of any Grid Services is exposed with the help of Service Data Elements. Grid Services may fail during its life cycle due to failure of a resource or a withdrawal of a resource by the resource owner. Thus, there is a need to provide a reliable solution in the form of Grid Service instance migration to protect the work of the users, which was carried out. This thesis proposes a model that supports Grid Services instance migration. Migration of an instance can take place based on the failure of resource, increase in load at the resource, change in the policy of the domain in which resource resides, user specified migration, or migration due to withdrawal of a resource by the resource owner. It enables the users to specify the migration if (s) he does not trust the domain in which instance is running. The model includes an incremental checkpointing mechanism to facilitate migration. Thesis shows the study of checkpointing mechanism in various resource managers: condor, condor-G and LSF. It narrates the different economic models prevalent in distributed computing field. The need for dynamic scheduler for Grid is also discussed. A model for instance migration for a task having parallel independent subtask is also proposed. A proposal to modify existing Globus-GRAM protocol to support Grid Service instance migration is also suggested.
  • Thumbnail Image
    ItemOpen Access
    Scheduling in grid: rescheduling MPI applications using a fault-tolerant MPI implementation
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2005) Reddy, M. Vivekananda; Chaudhary, Sanjay
    Due to advancement in grid technologies, resources spread across the globe can be accessed using standard general purpose protocols. Simulations and scientific experiments were earlier restricted due to limited availability of the resources. These are now carried out vigorously in the Grid. Grid environments are dynamic in nature. The resources in a grid are heterogeneous in nature and are not under a central control. So scheduling in grid is complex. The initial schedule obtained for an application may not be good as it involves the selection of resources at a future time. The resource characteristics like CPU availability, memory availability, network bandwidth etc keep changing. Rescheduling becomes necessary under these conditions. There are many rescheduling methods in the literature. Process migration is one of them. The thesis uses the fault-tolerant functionalities of MPICH-V2 to migrate MPI processes. Load balancing modules which make a decision of when and where to migrate a process are added into the MPICH-V2 system. Simulations are done to show that process migration is viable rescheduling technique for computationally intensive applications. The thesis also gives brief descriptions of some existing fault-tolerant MPI implementations.