<?xml version='1.0' encoding='UTF-8'?><?xml-stylesheet href='static/style.xsl' type='text/xsl'?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-22T08:34:41Z</responseDate><request verb="GetRecord" identifier="oai:ecommons.cornell.edu:1813/3489" metadataPrefix="dim">https://ecommons.cornell.edu/server/oai/request</request><GetRecord><record><header><identifier>oai:ecommons.cornell.edu:1813/3489</identifier><datestamp>2026-05-14T13:58:52Z</datestamp><setSpec>com_1813_35</setSpec><setSpec>col_1813_47</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="author">Bronevetsky, Grigory</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2006-08-30T13:44:20Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2006-08-30T13:44:20Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2006-08-30T13:44:20Z</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://hdl.handle.net/1813/3489</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="bibid">6475861</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en_US">High Performance Computing (HPC) systems represent the peak of modern computational capability. As&#xd;
ever-increasing demands for computational power have fuelled the demand for ever-larger computing systems,&#xd;
modern HPC systems have grown to incorporate hundreds, thousands or as many as 130,000 processors. At these&#xd;
scales, the huge number of individual components in a single system makes the probability that a single&#xd;
component will fail quite high, with today's large HPC systems featuring mean times between failures on the&#xd;
order of hours or a few days. As many modern computational tasks require days or months to complete, fault&#xd;
tolerance becomes critical to HPC system design.&#xd;
&#xd;
The past three decades have seen significant amounts of research on parallel system fault tolerance. However,&#xd;
as most of it has been either theoretical or has focused on low-level solutions that are embedded into a&#xd;
particular operating system or type of hardware, this work has had little impact on real HPC systems. This&#xd;
thesis attempts to address this lack of impact by describing a high-level approach for implementing&#xd;
checkpoint/restart functionality that decouples the fault tolerance solution from the details of the&#xd;
operating system, system libraries and the hardware and instead connects it to the APIs implemented by the&#xd;
above components. The resulting solution enables applications that use these APIs to become&#xd;
self-checkpointing and self-restarting regardless of the the software/hardware platform that may implement&#xd;
the APIs.&#xd;
&#xd;
The particular focus of this thesis is on the problem of checkpoint/restart of parallel applications. It&#xd;
presents two theoretical checkpointing protocols, one for the message passing communication model and one for&#xd;
the shared memory model. The former is the first protocol to be compatible with application-level&#xd;
checkpointing of individual processes, while the latter is the first protocol that is compatible with&#xd;
arbitrary shared memory models, APIs, implementations and consistency protocols. These checkpointing&#xd;
protocols are used to implement checkpointing systems for applications that use the MPI and OpenMP parallel&#xd;
APIs, respectively, and are first in providing checkpoint/restart to arbitrary implementations of these&#xd;
popular APIs. Both checkpointing systems are extensively evaluated on multiple software/hardware platforms&#xd;
and are shown to feature low overheads.</dim:field>
   <dim:field mdschema="dc" element="format" qualifier="extent">14956661 bytes</dim:field>
   <dim:field mdschema="dc" element="format" qualifier="mimetype">application/pdf</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en_US">en_US</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Computer Science</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Fault Tolerance</dim:field>
   <dim:field mdschema="dc" element="title" lang="en_US">Portable Checkpointing for Parallel Applications</dim:field>
   <dim:field mdschema="dc" element="type" lang="en_US">dissertation or thesis</dim:field>
   <dim:field mdschema="dc" element="format" qualifier="mimetype">application/pdf</dim:field>
   <dim:field mdschema="dspace" element="entity" qualifier="type">Publication</dim:field>
   <dim:field mdschema="cris" element="virtual" qualifier="collection" authority="https://cornell-ecommons.eks.prod.4science.cloud/handle/1813/47" confidence="600">Cornell Theses and Dissertations</dim:field>
   <dim:field mdschema="cris" element="virtual" qualifier="author">Bronevetsky, Grigory</dim:field>
   <dim:field mdschema="cris" element="virtualsource" qualifier="collection">5893a6ea-7af3-41d7-abc6-04bcd26ab5df</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
   <dim:field mdschema="cerif" element="openaire" authority="" confidence="-1">&lt;Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="850bc1fb-08ff-4688-86c3-14c114d1e97f">
	&lt;Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843&lt;/Type>
	&lt;Language>en_US&lt;/Language>
   	&lt;Title>Portable Checkpointing for Parallel Applications&lt;/Title>
   	&lt;PublishedIn>
    	&lt;Publication>
      	&lt;/Publication>
   	&lt;/PublishedIn>
   	&lt;PublicationDate>2006-08-30T13:44:20Z&lt;/PublicationDate>
   	&lt;Authors>
      	&lt;Author>
        	&lt;DisplayName>Bronevetsky, Grigory&lt;/DisplayName>
         	&lt;Affiliation>
         		&lt;OrgUnit>
         		&lt;/OrgUnit>
         	&lt;/Affiliation>
      	&lt;/Author>
	&lt;/Authors>
   	&lt;Editors>
	&lt;/Editors>
    &lt;Publishers>
        &lt;Publisher>
            &lt;OrgUnit />
        &lt;/Publisher>
    &lt;/Publishers>
    &lt;Keyword>Computer Science&lt;/Keyword>
    &lt;Keyword>Fault Tolerance&lt;/Keyword>
   	&lt;Abstract>High Performance Computing (HPC) systems represent the peak of modern computational capability. As&#xd;
ever-increasing demands for computational power have fuelled the demand for ever-larger computing systems,&#xd;
modern HPC systems have grown to incorporate hundreds, thousands or as many as 130,000 processors. At these&#xd;
scales, the huge number of individual components in a single system makes the probability that a single&#xd;
component will fail quite high, with today&amp;apos;s large HPC systems featuring mean times between failures on the&#xd;
order of hours or a few days. As many modern computational tasks require days or months to complete, fault&#xd;
tolerance becomes critical to HPC system design.&#xd;
&#xd;
The past three decades have seen significant amounts of research on parallel system fault tolerance. However,&#xd;
as most of it has been either theoretical or has focused on low-level solutions that are embedded into a&#xd;
particular operating system or type of hardware, this work has had little impact on real HPC systems. This&#xd;
thesis attempts to address this lack of impact by describing a high-level approach for implementing&#xd;
checkpoint/restart functionality that decouples the fault tolerance solution from the details of the&#xd;
operating system, system libraries and the hardware and instead connects it to the APIs implemented by the&#xd;
above components. The resulting solution enables applications that use these APIs to become&#xd;
self-checkpointing and self-restarting regardless of the the software/hardware platform that may implement&#xd;
the APIs.&#xd;
&#xd;
The particular focus of this thesis is on the problem of checkpoint/restart of parallel applications. It&#xd;
presents two theoretical checkpointing protocols, one for the message passing communication model and one for&#xd;
the shared memory model. The former is the first protocol to be compatible with application-level&#xd;
checkpointing of individual processes, while the latter is the first protocol that is compatible with&#xd;
arbitrary shared memory models, APIs, implementations and consistency protocols. These checkpointing&#xd;
protocols are used to implement checkpointing systems for applications that use the MPI and OpenMP parallel&#xd;
APIs, respectively, and are first in providing checkpoint/restart to arbitrary implementations of these&#xd;
popular APIs. Both checkpointing systems are extensively evaluated on multiple software/hardware platforms&#xd;
and are shown to feature low overheads.&lt;/Abstract>
	&lt;Access xmlns="http://purl.org/coar/access_right" 
    >
    &lt;/Access>
&lt;/Publication>
</dim:field>
</dim:dim>
</metadata></record></GetRecord></OAI-PMH>