HPC News Blizzard Day 1 – Forecast Calls for SC11

Print Friendly, PDF & Email

SC11 must be around the corner because a blizzard of news releases and vendor pitches hit our inbox today. What to do? Here’s the HPC News with Snark for Nov. 8, 2011.

  • A Day at the Chapel. Multicore wizard Bartosz Milewski writes that the upcoming full-day Chapel tutorial at SC11 is one not to miss. “Chapel does task-based parallelism better than Cilk, TBB or PPL; data-based parallelism better than AMP or ArBB; generic programming better than D.” It happens on Monday, Nov.14.
  • Carlsbad on ICE at NASA. Through something called the Space Act Agreement, NASA AMES is on the fast track to obtain 10 petaflops of peak computing capacity. Today they announced they will upgrade their Pleiades supercomputer with SGI ICE system based on the company’s next-gen Carlsbad 3.0 platform.
  • Throw Deep, Go Long on SGI. In related news, Seeking Alpha is bullish on SGI stock and this article breaks down the company’s financials every which way to back it up.
  • Foundation and Empire. ScaleMP’s new vSMP Foundation Advanced Platform now offers on-demand SMPs, partitioning, and Very Large Virtual Machines of up to 32 systems.
  • Freeze-dried Hadoop . The good folks at StackIQ (ClusterCorps) have released Rocks+ 6, which is designed for environments having hundreds or thousands of servers supporting Big Data, Analytics, or HPC. Want rapid deployment? Rocks+ 6 supports Apache Hadoop with automated provisioning and management of MapReduce, HDFS, HBase, Zookeeper, and Hive. There’s also Beta support for managing Cassandra and MongoDB.

That’s it for now. I’ll be back late today as soon as I find my snow shovel!