In this video from LUG 2014, Amir Shehata from Intel presents: Dynamic LNET Config.
“Compared to frequent developments in CPUs and accelerators, disk and storage communication technologies as well as parallel file systems are not realizing the same rate of innovation (Moore’s Law). However, this is changing as a number of new new and interesting technologies and developments such as HAMR drives, hybrid disks, intelligent flash, and new communication protocols including 12 Gb/s SAS and PCI-E direct attach are becoming available in the industry. This presentation intends to explain these technological directions and explore what these techniques mean for HPC storage over the coming years.”
“The ClusterStor SDA represents an industry first, combining low level data access enforcement and logging mechanisms, purpose-built within a proven HPC and Big Data single-system solution. Since our customers need to know exactly who is doing what on their systems, as well as access to complete and explicit audit trails, the ClusterStor SDA delivers all the necessary tools and support to collaboratively share information, while strictly complying with security data protection standards.”
“Back in July 2012, Whamcloud was awarded the Storage and I/O Research & Development subcontract for the Department of Energy’s FastForward program. Shortly afterward, the company was acquired by Intel. Nearly completed now, the two-year contract scope includes key R&D necessary for a new object storage paradigm for HPC exascale computing, and the developed technology will also address next-generation storage mechanisms required by the Big Data market.”
“LLNL’s largest supercomputer is paired with a 55-petabyte file system, known as Grove, that stores vast quantities of simulation data. Grove must transfer information to and from Sequoia at a minimum of 500 gigabytes per second. To support Grove’s high storage capacity and bandwidth requirements, LC software developers have engaged in a multi-year project to replace much of Grove’s Lustre foundation with ZFS. “