Richard Alpert


My research interests are in multicomputer communications, particularly low-latency, high-bandwidth message passing. My goal is to meld shared memory and message passing (which I do not see as mutually exclusive) into a single unified paradigm by creating truly efficient message passing systems. I also enjoy related architectural issues, but can be distracted by challenging theoretical questions. I'm examining message-passing costs and functionality in detail with Kai Li, Jim Philbin, and Jaswinder Pal Singh, to design and to build efficient software for machines supporting memory-mapped communication (like Shrimp ).

My pet conferences are ASPLOS (Architectural Support for Programming Languages and Operating Systems), SOSP (Symposium on Operating System Principles), OSDI (Operating Systems Design and Implementation), ISCA (International Symposium on Computer Architecture), and ICPP (International Conference on Parallel Processing).

Selected publications:

Relaxed Synchronization Message Passing.
Pending publication.

An Empirical Study of the Shrimp System.
International Symposium on Computer Architecture, June 1998.

cBSP: Zero-Cost Synchronization in a Modified BSP Model.
NEC Research Institute Technical Report 97-054, February 1997.

Design and Implementation of NX Message Passing using Virtual Memory Mapped Communication.
International Conference on Parallel Processing, August 1996.

Early Experience with Message-Passing on the SHRIMP Multicomputer.
International Symposium on Computer Architecture, May 1996.

Virtual Memory Mapped Network Interface for the SHRIMP Multicomputer.
International Symposium on Computer Architecture, April 1994.

A Low-latency, Virtual Memory Mapped Network Interface.
Third Workshop on Scalable Shared Memory Multiprocessors, May 1993.

Here's a longer list of papers and citations.

And here's an even longer list of papers.

My curriculum vitae: " Been there. Done that. "

If Sisyphus up there bothers you, hold down your escape key for a few seconds.

I love this place!*

la Speranza è l'Ultimo da Morire.

Una nuvola sola ricerca un cielo senza nubi.
Trova nessuna gioia, per sa che
Sa che l'alba non verrà prima che moriate.

* Terms and Restrictions Apply

non ottenere che cosa è che desiderate può essere un'occasione fantastica
ne pas obtenir cela que l'on veut peut porter bonheur fantastique