Programming languages researcher Dan Grossman from the University
of Washington works on making programs safe in the context of
shared memory and multi-core processors. Such research operates
both at the language level and at the level of the hardware,
which gives researchers such as Grossman a wide perspective on
the techniques we use in building safe, reliable software systems.
At ooPSLA, Grossman will deliver an essay
titled
The Transactional Memory / Garbage Collection Analogy,
which offers an intriguing analogy: transactional memory is to
shared memory concurrency as garbage collection is to memory
management. An essay is a different sort of paper that
ooPSLA offers. Grossman won't break
new technical ground in his talk; instead he will offer a new
perspective on topics that we all deal with and explore the
implications of thinking in this new way.
Listen to this podcast to hear Markus Völter of SE Radio chat
with Dan about shared memory concurrency, memory management, and
how thinking about the relationship between them can enrich our
understanding of both -- and how this can improve the way we