About Us

Scalable Logic was founded by the same group of external (i.e. non-Sun) developers who started contributing code to Sun Grid Engine since 2001.

Over a Decade of Grid Engine Dedication
Since chosen by Oracle as the maintainer of the open-source Grid Engine codebase in 2010, we have been keeping the Grid Engine project alive and active in the post-Sun Microsystems world.

We were active when Grid Engine was maintained by Sun Microsystems, and you still can count on us in the post Sun Grid Engine world!

No Vendor Lock-In
We are the maintainer of the official open-source Grid Engine, the Open Grid Scheduler Project, . The major features and enhancements developed for our Grid Engine project and contributed to the open-source world include:
  • Multi-Core Processor Binding using the hwloc topology library
  • cgroups integration for reliable job monitoring and tracking
  • Removal of NFSv4 dependency on BerkeleyDB spooling
  • NVIDIA CUDA GPU monitoring
  • Support for Linux 3.0 and the ARM platform
  • Cygwin port

We Have A Good Track Record That You Can Count On
We started developing and maintaining Grid Engine since 2001. Our major contributions include:
  • Scalability - our code enables a single Grid Engine cluster to have more than 1000 nodes
  • Performance - Over 30% performance gain in the qacct accounting command
  • Platform Support - FreeBSD, NetBSD, Mac OS X, Linux on MIPS, and Linux on PowerPC, IBM POWER, IBM CELL ports
  • Reliable Job Execution - process tracking and job accounting on AIX, HP-UX, FreeBSD, NetBSD, Mac OS X
  • System Load Monitoring - use of the AIX performance library
  • Performance Debugging and Diagnostics - DTrace Support
  • qmake modernization - we upgraded qmake to gmake version 3.80 (the then latest version)
  • Reliable (Tight) Grid Engine SSH Integration - used by many of the Sun HPC supercomputers
  • Bug Fixes in qtcsh, DRMAA library, qstat XML output, Grid Engine daemons
  • Installation Enhancements - support for Linux 2.6, support for FreeBSD multi-architecture, OSX startup scripts
  • And finally, we collaborated with many others on the development mailing list, and together we added many new features, including major performance improvement in parallel job start-up time, the DRMAA Perl binding, I/O accounting on Linux, the Linux on IBM z Series Mainframe port, Tango icons for qmon, and the NEC Super-UX port for the SX vector computer platform.

Did you know? We brought you the option to use classic spooling in Grid Engine.

In 2004, we insisted that Grid Engine needs job data spooling over NFSv2 & NFSv3, and we worked with Sun Microsystems to make sure that Grid Engine could failover in every environment, not just those that have NFSv4 file servers.

Unparalleled Scalability: From 1,000 to 10,000 Nodes

We fixed the Grid Engine 1,000-node limit encountered by Hess Corp back in 2005. In 2012, we set up industries' first 10,000+ nodes HPC cluster in the Amazon EC2 Cloud.

The Road Ahead - With
Open-Source Scalable Grid Engine
We work with many partners and are enhancing Grid Engine for the following environments:
  • Cloud Provisioning
  • GPU Computing
  • Big-Data & Hadoop
  • Native Windows Support
  • And lots more!

Contact Us: info@scalablelogic.com