The PMB-2.2.1 benchmark was completed on POWER4+ platforms p690+ and p655+. Point-to-point performance was measured between two processes within the same node (memory performance), or between two nodes (network performance). The performance was measured in MBytes/s per process (send+recv) in units of microseconds. The MPI-I output results, without modification, are shown in graphical format in this paper.
Did you find this white paper useful?
25 out of 50 users found this white paper useful
Related white papers
Make the Move from UNIX to Linux: Now is the Time
UNIX generally requires expensive proprietary software and hardware as well as budget-draining annual maintenance costs. These expenses are a key factor in recent wide-spread migrations to other platforms, specifically Linux....
SUSE Linux Enterprise: Differentiation Through Interoperability
Heterogeneous IT environments are a fact of life in large organizations. According to recent Gartner research, more than two thirds of all enterprise data centers operate mixed hardware and software...
Omaha Steaks and IBM Serve up Well-Done e-business
From its headquarters in America’s heartland, Omaha Steaks sells a variety of premium steaks, red meats and other gourmet foods. Although the company had established a successful Internet presence, fulfillment...
Deploying an Open Source HPC Environment on an IBM eServer BladeCenter JS20 Using SUSE LINUX Enterprise Server 8 and CSM
Cluster Systems Management (CSM) provides a distributed system management solution that allows a system administrator to set up and maintain a cluster of nodes that run the Linux operating system....
SUSE Linux Enterprise 10 SP2: Virtualization Technology Support
SUSE® Linux Enterprise Server 10 provides integrated Xen-based virtualization, which can be used to provision, de-provision, install, monitor and manage multiple virtual machines (VMs) on a single physical system. Out...
SUSE Linux Enterprise Server: The Solution for Mission-critical Computing
Dealing with costly and inflexible infrastructure that doesn't deliver enough value? Your data center should drive business forward--not derail it. It's time for a better choice in enterprise computing, and...
Introduction to HiperDispatch Management Mode with z10
HiperDispatch was introduced with IBM's z10 server, and is available (via PTFs) on z/OS V1R7, z/OS V1R8, and z/OS V1R9. HiperDispatch was designed to (1) minimize the z10 hardware performance...
