Loading…
Vault 2016 has ended

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

User Focus [clear filter]
Wednesday, April 20
 

11:30am PDT

Performance improvements in Replication and Erasure coding in GlusterFS - Anuradha Talur, Red Hat
GlusterFS is a scalable network filesystem that enables users to deploy large distributed storage solution using common off-the-shelf hardware. It provides high availability and redundancy through Automatic File Replication(AFR) and Erasure Coding(EC). In this presentation Anuradha Talur will be talking about improvements put forth in GlusterFS around AFR and EC to effectively tackle some of the long standing performance issues faced by users. These enhancements are targeted towards improving performance of small file workloads, and automatic repair/self-heal so that storage is not in degraded state for longer time. The talk also focus on improvements for scaling up self-heal/repair load when hardware can offer more iops, scaling self-heal load down during heavy traffic on storage from applications, and addressing the high consumption of CPU due to non systematic erasure coding.

Speakers
AT

Anuradha Talur

Red Hat
Anuradha Talur has been working at Red Hat as a software developer for close to three years. She has been an active contributor to GlusterFS, an open source, highly scalable distributed filesystem. She has been predominantly involved in the development of Automatic File Replication... Read More →


Wednesday April 20, 2016 11:30am - 12:20pm PDT
State Ballroom E

3:00pm PDT

pNFS for Block Storage - Sridhar Balasubramanian, NetApp
An integrated solution with pNFS block server and backend block storage comprises of highly optimized layout driver that exposes set of block volume LUNs that hosts the filesystem partition to pNFS clients with appropriate locking/release mechanism for concurrent client access to a file. The implementation includes new SCSI layout type and PR based client fencing for SCSI block storage. The solution is intended to address latency sensitive and scalable bandwidth requirements for high performance data processing workloads through cost optimized storage configurations and performance tiers for productization, and complements NetApp’s open source contributions pertaining to pNFS for file. The project is sponsored as an open source initiative under NetApp Hyperscalar Storage Group Advanced Development organization by closely collaborating with Linux NFS and kernel development community.

Speakers
SB

Sridhar Balasubramanian

NetApp
Speaker name: Sridhar Balasubramanian Bio: Sridhar is currently working as Sr. Engineer, Strategic Architecture @ NetApp RTP. With 25 years in software industry, Sridhar is inventor/co-inventor for 15 US Patents and published 4 Conference papers till date. Sridhar's area of expertise... Read More →


Wednesday April 20, 2016 3:00pm - 3:50pm PDT
Congressional Room AB
 
Thursday, April 21
 

2:00pm PDT

Deploying pNFS over Distributed File Storage - Jiffin Tony Thottan, Red Hat
pNFS is the clustered solution provided by standard NFS protocol which allow NFS clients to access directly and parallelly the Storage device. This is achieved by the separation of metadata from the datapath. Therefore pNFS solution provides better bandwidth utilization, loading balancing across storage devices and significant performance improvement for I/O's. Now consider applying the pNFS over a Distributed Storage instead of native NFS. Here instead of talking to single server, pNFS client can interact directly with all the Storage Servers , i.e I/O distribution become much more effective and also avoids performance bottleneck with in a single server. In this session, Jiffin Tony Thottan is planning to give an overview about pNFS, deploying simple pNFS solution on a Distribute File Storage by taking example of glusterfs + NFS-ganesha and challenges involved in while doing so.

Speakers
avatar for Jiffin Tony Thottan

Jiffin Tony Thottan

Senior Software Engineer, Red Hat
Jiffin Tony Thottan is part of the Red Hat Storage Team working as a Senior Software Engineer. Initially part of the NFS team and contributed to GlusterFS , NFS-Ganesha projects. He was an active contributor for both projects worked on implementing different features like nfsv4acls... Read More →


Thursday April 21, 2016 2:00pm - 2:50pm PDT
State Ballroom CD

3:00pm PDT

Accelerating Ceph performance Profiling and Tuning with CeTune - Jian Zhang, Intel
Ceph is an open-source, massively scalable, software-defined storage system, the demand to benchmarking, profiling and tuning Ceph in an easier and efficient way is increasing. However in production environment, customers still face numerous challenges to drive best performance, including how to troubleshooting the performance issues, identify the best tunings and handle the unexpected performance regression between frequent releases.
In this session, Chendi will introduce the components and capabilities of CeTune; present how to use CeTune to identify software stack bottlenecks and analyze unexpected software behavior; and at the end of the presentation, Chendi will present two case studies on how we improve Ceph cluster performance with CeTune.

Speakers
avatar for Jian Zhang

Jian Zhang

Intel
Jian is a senior software engineer of the Cloud Storage Engineer group from Intel Asia Pacific Research & Development Ltd. He is responsible for developmign software storage architecture and solutions based on Intel platforms. He has 7 years of software development, performance analysis... Read More →


Thursday April 21, 2016 3:00pm - 3:50pm PDT
State Ballroom F