Transparent Checkpoint-Restart of Distributed Applications on Commodity Clusters

Laadan, Oren; Phung, Dan B.; Nieh, Jason

doi:10.1109/clustr.2005.347039

Cited by 49 publications

(49 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We point out that distributed snapshot algorithms have long been proposed and applied [24,13,11,18,25,23] and thus are not our contribution. The contribution of VNsnap is the application of a classic snapshot algorithm to the emerging virtualized environments, as well as the proof of its applicability.…”

Section: Overviewmentioning

confidence: 99%

See 1 more Smart Citation

VNsnap: Taking snapshots of virtual networked environments with minimal downtime

Kangarlou

Eugster

2009

2009 IEEE/IFIP International Conference on Dependable Systems &Amp; Networks

View full text Add to dashboard Cite

A virtual networked environment (VNE) consists of virtual machines (VMs) connected by a virtual network. It has been adopted to create "virtual infrastructures" for individual users on a shared cloud computing infrastructure. The ability to take snapshots of an entire VNE -including images of the VMs with their execution, communication and storage states -yields a unique approach to reliability as a snapshot can restore the operation of an entire virtual infrastructure. We present VNsnap, a system that takes distributed snapshots of VNEs. Unlike existing distributed snapshot/checkpointing solutions, VNsnap does not require any modifications to the applications, libraries, or (guest) operating systems running in the VMs. Furthermore, VNsnap incurs only seconds of downtime as much of the snapshot operation takes place concurrently with the VNE's normal operation. We have implemented VNsnap on top of Xen. Our experiments with real-world parallel and distributed applications demonstrate VNsnap's effectiveness and efficiency.

show abstract

Section: Overviewmentioning

confidence: 99%

“…ZapC [18] is a thin virtualization layer that provides checkpoint/restart functionality for a self-contained virtual machine abstraction, namely a pod (PrOcess Domain), that contains a group of processes. Due to the smaller checkpointing granularity (a pod vs. a VM), ZapC is more efficient than VNsnap in checkpointing a group of processes.…”

Section: Related Workmentioning

confidence: 99%

VNsnap: Taking snapshots of virtual networked environments with minimal downtime

Kangarlou

Eugster

2009

2009 IEEE/IFIP International Conference on Dependable Systems &Amp; Networks

View full text Add to dashboard Cite

show abstract

“…This is especially problematic in the field of high performance computing where applications typically are memory demanding. Research has determined that the memory footprint is a major contributor to the checkpoint image size [7,20]. Further, due to the ever-increasing system size and complexity [4], failures occur more frequently than before, thereby making restart latency a critical concern in networked environments.…”

Section: Introductionmentioning

confidence: 99%

A fast restart mechanism for checkpoint/recovery protocols in networked environments

Lan

2008

2008 IEEE International Conference on Dependable Systems and Networks With FTCS and DCC (DSN)

View full text Add to dashboard Cite

show abstract

“…For instance, to support execution rollback, applications are placed inside the Zap [19,15] virtual execution environment, while RP code is injected using Dyninst [4]. Zap is a considerably complex component that is tightly coupled with the Linux kernel, and requires maintenance along with the operating system (OS).…”

Section: Introductionmentioning

confidence: 99%

REASSURE: A Self-contained Mechanism for Healing Software Using Rescue Points

Portokalidis

Keromytis

2011

Advances in Information and Computer Security

View full text Add to dashboard Cite

Abstract. Software errors are frequently responsible for the limited availability of Internet Services, loss of data, and many security compromises. Self-healing using rescue points (RPs) is a mechanism that can be used to recover software from unforeseen errors until a more permanent remedy, like a patch or update, is available. We present RE-ASSURE, a self-contained mechanism for recovering from such errors using RPs. Essentially, RPs are existing code locations that handle certain anticipated errors in the target application, usually by returning an error code. REASSURE enables the use of these locations to also handle unexpected faults. This is achieved by rolling back execution to a RP when a fault occurs, returning a valid error code, and enabling the application to gracefully handle the unexpected error itself. REAS-SURE can be applied on already running applications, while disabling and removing it is equally facile. We tested REASSURE with various applications, including the MySQL and Apache servers, and show that it allows them to successfully recover from errors, while incurring moderate overhead between 1% and 115%. We also show that even under very adverse conditions, like their continuous bombardment with errors, REASSURE protected applications remain operational.

show abstract

Transparent Checkpoint-Restart of Distributed Applications on Commodity Clusters

Cited by 49 publications

References 12 publications

VNsnap: Taking snapshots of virtual networked environments with minimal downtime

VNsnap: Taking snapshots of virtual networked environments with minimal downtime

A fast restart mechanism for checkpoint/recovery protocols in networked environments

REASSURE: A Self-contained Mechanism for Healing Software Using Rescue Points

Contact Info

Product

Resources

About