Optimized pre-copy live migration for memory intensive applications

Ibrahim, Khaled Z.; Hofmeyr, Steven; Iancu, Costin; Roman, Eric

doi:10.1145/2063384.2063437

Cited by 74 publications

(30 citation statements)

References 15 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The study indicated that our approach saves the total migration time by at least 65% and at most 75% in the memory intensive environment. This is in contrast to previous work in which, in the memory intensive environment, migration takes a long time or reaches a deadlock [6], [8]- [10]. In same environment, our approach can reduce more the total migration time compared to MECOM as much as 40%.…”

Section: Resultsmentioning

confidence: 61%

Optimized Pre-copy Migration Using STORD for Memory Intensive Environment

Cha¹,

Shin²,

Kim³

2015

IJCEE

View full text Add to dashboard Cite

Abstract:The conventional pre-copy method for live migration leads to redundant transmission of data transmitted. In addition, the migration efficiency decreases in the memory intensive environment where the memory is frequently modified because the modified memory data should do retransmission. Recently, these problems intensified because of growing complexity of computing environment such as Big-data and Cloud. In order to address these problems, we propose a method that is split transfer and omitting redundant dirty pages (STORD). We compared our approach with the conventional pre-copy method to evaluate the performance. In particular, we used a static bandwidth and generated a same size of dirty pages on split memory data for migration to show the effect of transferring the split memory. Simulation result showed that the proposed method saved the total migration time from 6% to 65% compared to pre-copy method. Also, when the environment has dirty page rate of 90%, our approach took less total migration time than MECOM which is the improvised method of pre-copy. Therefore, the proposed method is able to alleviate system performance degradation because our approach takes less migration time than previous method in Big-data or the multi-core environment.

show abstract

Section: Resultsmentioning

confidence: 61%

Optimized Pre-copy Migration Using STORD for Memory Intensive Environment

Cha¹,

Shin²,

Kim³

2015

IJCEE

View full text Add to dashboard Cite

show abstract

“…Using network attached storage (NAS), downtime can be brought down to the order of milliseconds (Mian et al, 2011) but for a local persistent storage the downtime will be significantly higher. Migrating CPU state, migrating memory content (Ibrahim et al, 2011), migrating storage content (Hirofuchi et al, 2009a(Hirofuchi et al, , 2009b, and migrating network connections are the important steps to be followed in migrating a VM from source host do destination host. Clark et al (2011) proposed pre-copy migration technique.…”

Section: Literature Surveymentioning

confidence: 99%

Live migration of virtual machines with their local persistent storage in a data intensive cloud

Modi

Achar

Thilagam

2017

IJHPCN

View full text Add to dashboard Cite

Abstract:Processing large volumes of data to drive their core business has been the primary objective of many firms and scientific applications in these days. Cloud computing being a large-scale distributed computing paradigm can be used to cater for the needs of data intensive applications. There are various approaches for managing the workload on a data intensive cloud. Live migration of a virtual machine is the most prominent paradigm. Existing approaches to live migration use network attached storage where just the run time state needs to be transferred. Live migration of virtual machines with local persistent storage has been shown to have performance advantages like security, availability and privacy. This paper presents an optimised approach for migration of a virtual machine along with its local storage by considering the locality of storage access. Count map combined with a restricted block transfer mechanism is used to minimise the downtime and overhead. The solution proposed is tested by various parameters like bandwidth, write access patterns and threshold. Results show the improvement in downtime and reduction in overhead.

show abstract

“…As opposed to evacuation in the migratable-objects model, task migration in CoCheck was synchronous with the execution of the system, i.e., before an MPI rank could be migrated, CoCheck had to make sure all communicating ranks would hold back messages until the rank was at its new location. Similar tools have recently used process-level live migration in MPI applications [36], [37], [38]. Those tools combine health monitoring of nodes with live migration to provide proactive fault tolerance.…”

Section: Related Workmentioning

confidence: 99%

Using Migratable Objects to Enhance Fault Tolerance Schemes in Supercomputers

Meneses

Zheng

et al. 2015

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Abstract-Supercomputers have seen an exponential increase in their size in the last two decades. Such a high growth rate is expected to take us to exascale in the timeframe 2018-2022. But, to bring a productive exascale environment about, it is necessary to focus on several key challenges. One of those challenges is fault tolerance. Machines at extreme scale will experience frequent failures and will require the system to avoid or overcome those failures. Various techniques have recently been developed to tolerate failures. The impact of these techniques and their scalability can be substantially enhanced by a parallel programming model called migratable objects. In this paper, we demonstrate how the migratable-objects model facilitates and improves several fault tolerance approaches. Our experimental results on thousands of cores suggest fault tolerance schemes based on migratable objects have low performance overhead and high scalability. Additionally, we present a performance model that predicts a significant benefit of using migratable objects to provide fault tolerance at extreme scale.

show abstract

Optimized pre-copy live migration for memory intensive applications

Cited by 74 publications

References 15 publications

Optimized Pre-copy Migration Using STORD for Memory Intensive Environment

Optimized Pre-copy Migration Using STORD for Memory Intensive Environment

Live migration of virtual machines with their local persistent storage in a data intensive cloud

Using Migratable Objects to Enhance Fault Tolerance Schemes in Supercomputers

Contact Info

Product

Resources

About