NDLI: Data-driven fault tolerance for work stealing computations

Please wait, while we are loading the content...

High performance supercomputers: should the individual processor be more than a brick?

Distributed replay protocol for distributed uniprocessors

Characterizing and improving the use of demand-fetched caches in GPUs

UniFI: leveraging non-volatile memories for a unified fault tolerance and idle power management technique

Overcoming single-thread performance hurdles in the core fusion reconfigurable multicore architecture

HiRe: using hint & release to improve synchronization of speculative threads

Sparse matrix-vector multiply on the HICAMP architecture

Blue Gene/Q: design for sustained multi-petaflop computing

An analysis of computational workloads for the ORNL Jaguar system

Multiple sub-row buffers in DRAM: unlocking performance and energy improvement opportunities

Hardware support for enforcing isolation in lock-based parallel programs

SnuCL: an OpenCL framework for heterogeneous CPU/GPU clusters

One stone two birds: synchronization relaxation and redundancy removal in GPU-CPU translation

Fault tolerant preconditioned conjugate gradient for sparse linear system solution

CVP: an energy-efficient indirect branch prediction with compiler-guided value pattern

Enhancing the performance of assisted execution runtime systems through hardware/software techniques

On the communication complexity of 3D FFTs and its implications for Exascale

Unified memory optimizing architecture: memory subsystem control with a unified predictor

High-performance code generation for stencil computations on GPU architectures

clSpMV: A Cross-Platform OpenCL SpMV Framework on GPUs

Fast loop-level data dependence profiling

Data-driven fault tolerance for work stealing computations

Congestion avoidance on manycore high performance computing systems

CATS: cache aware task-stealing based on online profiling in multi-socket multi-core architectures

Composable, non-blocking collective operations on power7 IH

Locality & utility co-optimization for practical capacity management of shared last level caches

An efficient work-distribution strategy for gridding radio-telescope data on GPUs

Enabling and scaling matrix computations on heterogeneous multi-core and multi-GPU systems

Apricot: an optimizing compiler and productivity tool for x86-compatible many-core coprocessors

Fault resilience of the algebraic multi-grid solver

Channel borrowing: an energy-efficient nanophotonic crossbar architecture with light-weight arbitration

CRQ-based fair scheduling on composable multicore architectures

Collective algorithms for sub-communicators

Exploiting communication and packaging locality for cost-effective large scale networks

GPU merge path: a GPU merging algorithm

An optimized large-scale hybrid DGEMM design for CPUs and ATI GPUs

Quantifying the effectiveness of load balance algorithms

Space-round tradeoffs for MapReduce computations

Data-driven fault tolerance for work stealing computations

Content Provider	ACM Digital Library
Author	Krishnamoorthy, Sriram Ma, Wenjing
Abstract	Work stealing is a promising technique to dynamically tolerate variations in the execution environment, including faults, system noise, and energy constraints. In this paper, we present fault tolerance mechanisms for task parallel computations, a popular computation idiom, employing work stealing. The computation is organized as a collection of tasks with data in a global address space. The completion of data operations, rather than the actual messages, is tracked to derive an idempotent data store. This information is also used to accurately identify the tasks to be re-executed in the presence of random work stealing. We consider three recovery schemes that present distinct trade-offs --- lazy recovery with potentially increased re-execution cost, immediate collective recovery with associated synchronization overheads, and noncollective recovery enabled by additional communication. We employ distributed-memory work stealing to dynamically rebalance the tasks onto the live processes and evaluate the three schemes using candidate application benchmarks. We demonstrate that the overheads (space and time) of the fault tolerance mechanism are low, the costs incurred due to failures are small, and the overheads decrease with per-process work at scale.
Starting Page	79
Ending Page	90
Page Count	12
File Format	PDF
ISBN	9781450313162
DOI	10.1145/2304576.2304589
Language	English
Publisher	Association for Computing Machinery (ACM)
Publisher Date	2012-06-25
Publisher Place	New York
Access Restriction	Subscribed
Subject Keyword	Work stealing Fault tolerance Load balancing
Content Type	Text
Resource Type	Article

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in

A work-stealing scheduling framework supporting fault tolerance

Load balancing using work-stealing for pipeline parallelism in emerging applications

Dynamic circular work-stealing deque

Understanding and improving JVM GC work stealing at the data center scale

A work-stealing scheduling framework supporting fault tolerance

Scalable work stealing

Scalable work stealing

Work stealing and persistence-based load balancers for iterative overdecomposed applications

Lifeline-based global load balancing

Data-driven fault tolerance for work stealing computations

Similar Documents

A work-stealing scheduling framework supporting fault tolerance

Load balancing using work-stealing for pipeline parallelism in emerging applications

Dynamic circular work-stealing deque

Understanding and improving JVM GC work stealing at the data center scale

A work-stealing scheduling framework supporting fault tolerance

Scalable work stealing

Scalable work stealing

Work stealing and persistence-based load balancers for iterative overdecomposed applications

Lifeline-based global load balancing

Data-driven fault tolerance for work stealing computations