With the advent of BYOD = Bring Your Own Device it was of interest to discover for admittedly extreme application scenarios how well different hardware configurations performed as the number of calculations increased and the number of application threads spawned varied. First up, an Acer Aspire One netbook: 1 Gbyte of RAM; 32-bit Windows 7; Intel Atom N450 CPU (1.66 GHz); L1 Data cache - 24 Kb/ 6-way set associative/64-byte line size; L1 Instruction cache  - 32 Kb/ 8-way set associative/64-byte line size;  L2 cache  - 512 Kb/ 8-way set associative/64-byte line size; 1 core with Hyperthreading; Intel 3150 Graphics Media Accelerator (not exploited this time).

The ORIBI application has a portion that can be abused to test how well a configuration handles a compute-intense workload with heavy disk writing. Number of calculations across the top; number of threads down the left side. A great many intermediate results (3 threads; 5 threads ... 2 million, 3 million ... calculations) have been omitted