Tyr: Blob storage systems meet Built-In transactions

Matri, Pierre and Costan, Alexandru and Antoniu, Gabriel and Montes Sánchez, Jesús and Pérez Hernández, María de los Santos (2016). Tyr: Blob storage systems meet Built-In transactions. In: "International Conference for High Performance Computing, Networking, Storage and Analysis", 13-18 Nov 2016, Salt Lake City, Utah, Estados Unidos. ISBN 978-1-4673-8815-3. pp. 573-584. https://doi.org/doi.ieeecomputersociety.org/10.1109/SC.2016.48.

Description

Title: Tyr: Blob storage systems meet Built-In transactions
Author/s:
  • Matri, Pierre
  • Costan, Alexandru
  • Antoniu, Gabriel
  • Montes Sánchez, Jesús
  • Pérez Hernández, María de los Santos
Item Type: Presentation at Congress or Conference (Article)
Event Title: International Conference for High Performance Computing, Networking, Storage and Analysis
Event Dates: 13-18 Nov 2016
Event Location: Salt Lake City, Utah, Estados Unidos
Title of Book: SC16: International Conference for High Performance Computing, Networking, Storage and Analysis (SC) (2016)
Date: November 2016
ISBN: 978-1-4673-8815-3
Volume: 1
Subjects:
Faculty: E.T.S. de Ingenieros Informáticos (UPM)
Department: Arquitectura y Tecnología de Sistemas Informáticos
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB) | Preview

Abstract

Concurrent Big Data applications often require high-performance storage, as well as ACID (Atomicity, Consistency, Isolation, Durability) transaction support. Although blobs (binary large objects) are an increasingly popular storage model for such applications, state-of-the-art blob storage systems offer no transaction semantics. This demands users to coordinate data access carefully in order to avoid race conditions, inconsistent writes, overwrites and other problems that cause erratic behavior. We argue there is a gap between existing storage solutions and application requirements, which limits the design of transactionoriented applications. We introduce T¿yr, the first blob storage system to provide built-in, multiblob transactions, while retaining sequential consistency and high throughput under heavy access concurrency. T¿yr offers fine-grained random write access to data and in-place atomic operations. Large-scale experiments with a production application from CERN LHC show T¿yr throughput outperforming state-of-the-art solutions by more than 75%.

Funding Projects

TypeCodeAcronymLeaderTitle
Horizon 2020642963BigStorageUniversidad Politécnica de MadridBigStorage: Storage-based Convergence between HPC and Cloud to handle Big Data

More information

Item ID: 45072
DC Identifier: http://oa.upm.es/45072/
OAI Identifier: oai:oa.upm.es:45072
DOI: doi.ieeecomputersociety.org/10.1109/SC.2016.48
Official URL: https://www.computer.org/csdl/proceedings/sc/2016/8815/00/8815a573-abs.html
Deposited by: Memoria Investigacion
Deposited on: 09 Mar 2017 15:52
Last Modified: 09 Mar 2017 15:52
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM