Nikos Anastopoulos

Menu:

• Welcome!
• Personal Info
• Research
• Publications
• Links
• Misc
• Contact

Links

---

Libraries for parallelism

Task-based runtime systems

• Threading Building Blocks

Synchronization

• atomic_ops
• Concurrency Kit

Tools

Simulators

Full system, functional, cycle-accurate, for multiprocessor/multithreaded architectures:

• SIMICS
• gem5
• GEMS (gem5 predecessor)

Microarchitectural, application-level simulators:

• Simplescalar LLC
• SESC

Educational:

• WinDLX: MIPS-like, in-order pipeline & visualization.

Compiler frameworks

• Cetus: source to source C compiler.

Profiling and debugging

• Pin: a tool for dynamic binary instrumentation
• Valgrind: a suite of instrumentation tools for debugging, profiling and analyzing Linux programs. Provides a cache profiler, as well.
• Oprofile: a system-wide profiler for Linux systems. Supports performance counters for a wide range of architectures.

Latex and typesetting

• Latex usage notes
• Greek font society

---

Applications and benchmarks

Parallel benchmarks

• NAS Parallel Benchmarks v.2.3: OpenMP C versions, provided by the Omni OpenMP Compiler Project. More info about NAS Parallel Benchmarks here.
• PARSEC: provides Pthreads, OpenMP and TBB implementations
• SPLASH-2 parallel benchmark suite: C-based, parallelized using the PARMACS macros. PARMACS implementations for Pthreads can be found here or here.
• Barcelona OpenMP Task Suite
• Lonestar: a collection of widely-used real-world applications with irregular behaviour
• Graph 500 benchmark suite: a set of large scale benchmarks for data intensive applications.

Serial benchmarks

• Olden benchmark suite: a set of pointer-intensive benchmarks, involving recursive data structures such as trees and linked lists.
• Livermore loops: 24 kernels encountered in many numerical applications.

---

Resources and articles

• "Welcome to the Jungle"
• The microarchitecture of Intel and AMD CPUs: An optimization guide for assembly programmers and compiler makers.
• What every programmer should know about memory
• The Landscape of Parallel Computing Research: A View From Berkeley

---

People

• Herb Sutter
• Bartosz Milewski
• Hans Boehm
• Dave Dice
• Dmitry Vyukov

---

Useful stuff

• Matplotlib: Python-based plotting tool
• JabRef: bibliography reference manager.
• Meld: graphical tool for diffing and merging files.
• Common bugs in writing papers.

---

Recommended books

• Computer Architecture, 4th edition: A Quantitative Approach, by John Hennesy and David Patterson. Morgan Kaufmann.
• C Programming Language, 2nd edition, by Brian Kernighan and Dennis Ritchie. Prentice Hall.
• Accelerated C++: Practical Programming by Example, by Andrew Koenig and Barbara Moo. Addison-Wesley.
• Effective C++, 3rd edition, by Scott Meyers. Addison Wesley.
• The Art of Multiprocessor Programming, by Maurice Herlihy and Nir Shavit. Morgan Kaufmann.
• Understanding the Linux Kernel, 3rd edition, by Daniel Bovet and Marco Cesati. O'Reilly.
• Introduction to Algorithms, 3rd edition, by Thomas Cormen, Charles Leiserson, Ronald Rivest and Clifford Stein. MIT Press.
• Distributed Algorithms, by Nancy Lynch. Morgan Kaufmann.
• Introduction to the Theory of Computation, by Michael Sipser. Course Technology.
• Database System Concepts, by Abraham Silberschatz, Henry Korth and S. Sudarshan. McGraw-Hill.

---