“-Hoard” — enabling the use of my Hoard memory allocator — is now an officially sanctioned configuration flag for SPEC CPU2006 (the industry-standard way to measure CPU performance)! See the flags in use for the Intel compiler and Open64. My opinions about benchmarking notwithstanding, I am ok with my work being a standard configuration flag
.
Best. SPEC configuration flag. Ever!
Filed under Uncategorized
Cybersecurity interview on PBS
From PBS affiliate WGBY: University of Mass. Prof. of Computer Science Emery Berger discusses cyber security and what we can do to protect our information from computer hackers.
Filed under Uncategorized
Hot off the Presses (Dthreads & Sheriff)
Camera-ready versions of recent pubs from our research group:
SOSP 2011: Dthreads: Efficient and Deterministic Multithreading, Tongping Liu, Charlie Curtsinger, and Emery Berger.
OOPSLA 2011: Sheriff: Precise Detection and Automatic Mitigation of False Sharing, Tongping Liu and Emery Berger.
Filed under Uncategorized
Dthreads: Efficient Deterministic Multithreading
Dthreads: Efficient Deterministic Multithreading,
Tongping Liu, Charlie Curtsinger, Emery D. Berger
SOSP 2011

[paper (PDF)][source code] [YouTube video of presentation][PPT slides]
Multithreaded programming is notoriously difficult to get right. A key problem is non-determinism, which complicates debugging, testing, and reproducing errors. One way to simplify multithreaded programming is to enforce deterministic execution, but current deterministic systems for C/C++ are incomplete or impractical. These systems require program modification, do not ensure determinism in the presence of data races, do not work with general-purpose multithreaded programs, or run up to 8.4× slower than pthreads.
This paper presents Dthreads, an efficient deterministic mul- tithreading system for unmodified C/C++ applications that replaces the pthreads library. Dthreads enforces determinism in the face of data races and deadlocks. Dthreads works by exploding multithreaded applications into multiple processes, with private, copy-on-write mappings to shared memory. It uses standard virtual memory protection to track writes, and deterministically orders updates by each thread. By separating updates from different threads, Dthreads has the additional benefit of eliminating false sharing. Experimental results show that Dthreads substantially outperforms a state-of-the-art deterministic runtime system, and for a majority of the benchmarks evaluated here, matches and occasionally exceeds the performance of pthreads.
Related post: SHERIFF: Precise Detection and Automatic Mitigation of False Sharing
Filed under Concurrency
SHERIFF: Precise Detection and Automatic Mitigation of False Sharing
Sheriff: Precise Detection and Automatic Mitigation of False Sharing
Tongping Liu, Emery D. Berger, OOPSLA 2011
False sharing is an insidious problem for multithreaded programs running on multicore processors, where it can silently degrade performance and scalability. Previous tools for detecting false sharing are severely limited: they cannot distinguish false sharing from true sharing, have high false positive rates, and provide limited assistance to help programmers locate and resolve false sharing.
This paper presents two tools that attack the problem of false sharing: SHERIFF-DETECT and SHERIFF-PROTECT. Both tools leverage a framework we introduce here called SHERIFF. SHERIFF breaks out threads into separate processes, and exposes an API that allows programs to perform per-thread memory isolation and tracking on a per-page basis. We believe SHERIFF is of independent interest.
SHERIFF-DETECT finds instances of false sharing by comparing updates within the same cache lines by different threads, and uses sampling to rank them by performance impact. SHERIFF-DETECT is precise (no false positives), runs with low overhead (on average, 20%), and is accurate, pinpointing the exact objects involved in false sharing. We present a case study demonstrating SHERIFF-DETECT’s effectiveness at locating false sharing in a variety of benchmarks.
Rewriting a program to fix false sharing can be infeasible when source is unavailable, or undesirable when padding objects would unacceptably increase memory consumption or further worsen runtime performance. SHERIFF-PROTECT mitigates false sharing by adaptively isolating shared updates from different threads into separate physical addresses, effectively eliminating most of the performance impact of false sharing. We show that SHERIFF-PROTECT can improve performance for programs with catastrophic false sharing by up to 9x, without programmer intervention.
Filed under Uncategorized
AluPen: Best iPad Stylus Yet (By Far)
If you have an iPad, and (like me) you’d like to be able to take handwritten notes on it, there is finally a decent stylus that does the job: the AluPen. It’s chunky and a little weird, with a round rubber tip that looks like half of a tiny racquetball, but it works great (unlike every other iPad stylus I have tried).
Wired has a nice review. I bought mine on Amazon; it’s available in a wide range of colors, but I like the gold for its retro pencil look. I am using it in combination with NoteTaker HD, an app by Dan Bricklin of VisiCalc fame, though other note taking apps with a magnified writing area would be fine.
Filed under Uncategorized
Boom: Crank up your Mac
I like my MacBook Air (my hands are soaking in it right now), but the speakers suck. I have never been able to get much volume out of them. Until now.
One of the winners of the MacWorld awards this year was Boom, a “volume booster” for your Mac. Works like a charm! It easily doubles the volume of the speakers, making it practical to use the Air to watch, say, The Daily Show, without having to don earphones. Well worth the $5 to crank your Mac up to 11.
Filed under Uncategorized

