Future-generation space missions across the solar system to the planets, mo
ons, asteroids, and comets may someday incorporate supercomputers both to e
xpand the range of missions being conducted and to significantly reduce the
ir cost. By performing science computation directly on the spacecraft itsel
f, the amount of data required to be downlinked may be reduced by many orde
rs of magnitude, thus greatly reducing the mass of the resources needed for
communication while increasing the quality and quantity of the science ach
ieved. By performing the mission planning in real time directly on the spac
ecraft, complex and highly responsive missions can be conducted out of rang
e of direct human intervention, and the cost of mission management can be r
educed. Through highly replicated computing structures, continued operation
can be maintained in the presence of faults by means of graceful degradati
on. Two classes of systems, reflecting very different strategies of compute
r system architecture, are actively being pursued by the NASA Jet Propulsio
n Laboratory (JPL) to take advantage of the opportunity of embedded high pe
rformance computing on spacecraft for deep-space missions. Commodity off-th
e-shelf (COTS) clusters may permit the direct application of commercial com
puting hardware in loosely coupled ensembles to benefit from the enormous i
nvestment of industry in mass-market components. New processor-in-memory (P
IM) architectures combine multiple nodes on a single chip of processor-memo
ry pairs exposing the full memory bandwidth. This paper examines the drivin
g issues motivating the use of supercomputing for future deep-space mission
s and describes two active research projects at NASA JPL that are pursuing
both the COTS and PIM strategies for next-generation spaceborne computing.