Change search
ReferencesLink to record
Permanent link

Direct link
Scaling HDFS with a Strongly Consistent Relational Model for Metadata
RISE, Swedish ICT, SICS.
RISE, Swedish ICT, SICS.
Number of Authors: 3
2014 (English)Conference paper (Refereed)
Abstract [en]

The Hadoop Distributed File System (HDFS) scales to store tens of petabytes of data despite the fact that the entire file system's metadata must fit on the heap of a single Java virtual machine. The size of HDFS' metadata is limited to under 100 GB in production, as garbage collection events in bigger clusters result in heartbeats timing out to the metadata server(NameNode). In this paper, we address the problem of how to migrate the HDFS' metadata to a relational model, so that we can support larger amounts of storage on a shared nothing, in-memory, distributed database. Our main contribution is that we show how to provide at least as strong consistency semantics as HDFS while adding support for a multiple-writer, multiple-reader concurrency model. We guarantee freedom from deadlocks by logically organizing inodes (and their constituent blocks and replicas) into a hierarchy and having all metadata operations agree on a global order for acquiring both explicit locks and implicit locks on subtrees in the hierarchy. We use transactions with pessimistic concurrency control to ensure the safety and progress of metadata operations. Finally, we show how to improve performance of our solution by introducing a snapshotting mechanism at NameNodes that minimizes the number of roundtrips to the database

Place, publisher, year, edition, pages
2014, 8. 38-51 p.
National Category
Computer and Information Science
URN: urn:nbn:se:ri:diva-15546OAI: diva2:1036863
DAIS 2014
Available from: 2016-10-13 Created: 2016-10-13

Open Access in DiVA

fulltext(591 kB)4 downloads
File information
File name FULLTEXT01.pdfFile size 591 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Other links


Search in DiVA

By author/editor
Hakimzadeh, KamalPeiro Sajjad, Hooman
By organisation
Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 4 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

ReferencesLink to record
Permanent link

Direct link