The Profet Project
Overview:
The goal of the Profet project is to avoid wasting time whenever we access
data. To accomplish this, we proactively manage caches such that the data
that you want is available in the cache before you actually need it.
The name "Profet" comes from "PROactive FETching data". Profet is also
a deliberate misspelling of both "prophet" (as in predicting the future,
which is what we do improve cache performance) and "profit" (as in
optimizing "cash" - sic), and it is pronounced like both of those words.
People Working for Profet:
Project Leader:
Graduate Students:
Undergrads:
Dearly Departed:
Publications:
Improving the Cache Performance of Databases:
Prefetching Data from Memory into Cache:
Prefetching Data from Disk into Memory:
- Taming the Memory Hogs: Using Compiler-Inserted Releases to Manage
Physical Memory Intelligently.
[ps],
[ps.gz]
Angela Demke Brown and Todd C. Mowry.
Proceedings of the Fourth Symposium on Operating Systems Design and
Implementation (OSDI 2000), pages 31-44, October 2000.
-
Automatic Compiler-Inserted I/O Prefetching for Out-of-Core Applications.
[ps.gz]
Todd C. Mowry, Angela K. Demke and Orran Krieger.
Proceedings of the Second Symposium on Operating Systems Design and
Implementation (OSDI '96), pages 3-17, October 1996.
Prefetching Instructions:
-
Cooperative Prefetching: Compiler and Hardware Support for Effective Instruction
Prefetching in Modern Processors.
[ps],
[ps.gz]
Chi-Keung Luk and Todd C. Mowry.
In Proceedings of Micro-31, Nov. 30 - Dec. 2, 1998, Dallas, Texas.
- Earlier Tech Report Version:
Compiler and Hardware Support for Automatic Instruction Prefetching: A Cooperative Approach.
[ps],
[ps.gz]
Chi-Keung Luk and Todd C. Mowry.
Technical Report CMU-CS-98-140, School of Computer Science,
Carnegie Mellon University, June 1998.
Facilitating Data Layout Optimizations:
-
Memory Forwarding: Enabling Aggressive Layout Optimizations
by Guaranteeing the Safety of Data Relocation.
[ps],
[ps.gz]
Chi-Keung Luk and Todd C. Mowry.
In Proceedings of the 26th Annual International Symposium on Computer
Architecture, May 1999.
Predicting When Cache Misses Occur:
-
Understanding Why Correlation Profiling Improves the Predictability of Data
Cache Misses in Nonnumeric Applications.
Todd C. Mowry and
Chi-Keung Luk.
In IEEE Transactions on
Computers, Vol. 49, No. 4, April 2000.
-
Predicting Data Cache Misses in Non-Numeric Applications Through Correlation Profiling.
[ps],
[ps.gz]
Todd C. Mowry and Chi-Keung Luk.
Proceedings of Micro-30, December 1-3, 1997, Research
Triangle Park, North Carolina.
- Extended Tech Report Version:
Predicting Data Cache Misses in Non-Numeric Applications Through Correlation Profiling.
[ps],
[ps.gz]
Todd C. Mowry and Chi-Keung Luk
Technical Report CMU-CS-97-175,
School of Computer Science, Carnegie Mellon University,
Pittsburgh, PA 15213, September 1997.
Informing Memory Operations:
-
Software-Controlled Multithreading Using Informing Memory Operations.
[ps],
[ps.gz]
Todd C. Mowry and Sherwyn R. Ramkissoon.
In Proceedings of the Sixth International Symposium on High-Performance Computer Architecture,
January, 2000.
- Earlier Tech Report Version:
Software-Controlled Multithreading Using Informing Memory Operations.
[ps],
[ps.gz]
Todd C. Mowry and Sherwyn R. Ramkissoon.
Technical Report CMU-CS-98-169, School of Computer Science,
Carnegie Mellon University, October 1998.
-
Informing Memory Operations: Memory Performance Feedback Mechanisms and Their
Applications.
Mark Horowitz, Margaret Martonosi,
Todd C. Mowry,
and Michael D. Smith.
In ACM Transactions
on Computer Systems, 16(2):170-205, May 1998.
- Informing Memory Operations:
Providing Memory Performance Feedback in Modern Processors.
[ps.gz]
Mark Horowitz, Margaret Martonosi, Todd C. Mowry, and Michael D. Smith.
Proceedings of the 23rd Annual International Symposium on Computer
Architecture, pages 260-270, May 1996.
- Earlier Tech Report Version:
Informing Loads: Enabling Software
to Observe and React to Memory Behavior.
[ps.gz]
Mark Horowitz, Margaret Martonosi, Todd C. Mowry, and Michael D. Smith.
Stanford University Technical Report CSL-TR-95-673 (also numbered STAN-CS-95-675),
July 1995.
Tolerating Network Latencies:
-
Comparative Evaluation of Latency Tolerance Techniques for Software Distributed Shared Memory.
[ps],
[ps.gz]
Todd C. Mowry, Charles Q. C. Chan and Adley K. W. Lo.
Proceedings of the Fourth International Symposium on High-Performance Computer
Architecture, February 2-4, 1998, Las Vegas, Nevada.