Quote: a data parallel algorithm performs simultaneous operations across large sets of data; like APL and FP [»hillWD12_1986]

Quote: data parallel algorithms typically use O(N) processors to solve a problem of size N in O(log N) time

Quote: both CmLisp and the Connection Machine achieves parallelism through simultaneous operations instead of concurrent control [»hillWD_1985]

Quote: NESL is a parallel programming language for expressing nested parallelism and data parallelism [»blelGE3_1996]

Quote: Accelerator uses data parallelism to program GPUs using a conventional imperative language; within 50% of hand-written code; up to 18x faster than C [»tardD10_2006]

Quote: the Connection Machine represents and processes data as 'active data structures' of interconnected processor and memory cells [»hillWD_1985]

Quote: an active data structure is a machine; the host controls the Connection Machine by telling the data (processor/memory) what to do [»hillWD_1985]

Quote: parallel operations in CmLisp via a xector--a set of processors with one value per processor; e.g., xector add [»hillWD_1985]

Quote: a xector element consists of a index (processor name and memory address) and a value; includes sets, index sequence, and constants [»hillWD_1985]

Quote: apply a xector of functions to all tuples with common xector indices; e.g., (.alpha.+ '{a->1 b->2} '{b-3 c->2}) => {b->5} [»hillWD_1985]

Quote: parallel arrays are Accelerator's fundamental data type; homogeneous, no index operator, no side effects; the result of any operation is a new array [»tardD10_2006]

Quote: set union, intersection, and universal qualifier can be unit-time operations on the Connection Machine [»hillWD_1985]

Quote: the Connection Machine can search for substrings in time proportional to the length of the substring [»hillWD_1985]

Quote: search is parallelizable by randomly dividing index into pieces called index shards; a pool of machines for each shard [»barrLA3_2003]