Copyright (C) 1998-2014 by Emery Berger
The Hoard memory allocator is a fast, scalable, and memory-efficient memory allocator that works on a range of platforms, including Linux, Solaris, Mac OS X, and Windows.
Hoard is a drop-in replacement for malloc that can dramatically improve application performance, especially for multithreaded programs running on multiprocessors and multicore CPUs. No source code changes necessary: just link it in or set one environment variable (see Building Hoard, below).
-
"If you'll be running on multiprocessor machines, ... use Emery Berger's excellent Hoard multiprocessor memory management code. It's a drop-in replacement for the C and C++ memory routines and is very fast on multiprocessor machines."
-
"(To improve scalability), consider an open source alternative such as the Hoard Memory Manager..."
-
"Hoard dramatically improves program performance through its more efficient use of memory. Moreover, Hoard has provably bounded memory blowup and low synchronization costs."
Companies using Hoard in their products and servers include AOL, British Telecom, Blue Vector, Business Objects (formerly Crystal Decisions), Cisco, Credit Suisse, Entrust, InfoVista, Kamakura, Novell, Oktal SE, OpenText, OpenWave Systems (for their Typhoon and Twister servers), Pervasive Software, Plath GmbH, Quest Software, Reuters, Royal Bank of Canada, SAP, Sonus Networks, Tata Communications, and Verite Group.
Open source projects using Hoard include the Asterisk Open Source Telephony Project, Bayonne GNU telephony server, the Cilk parallel programming language, the GNU Common C++ system, the OpenFOAM computational fluid dynamics toolkit, and the SafeSquid web proxy.
Hoard is now a standard compiler option for the Standard Performance Evaluation Corporation's CPU2006 benchmark suite for the Intel and Open64 compilers.
Hoard is distributed under the GPL (v2.0), and can also be licensed for commercial use.
Because of the restrictions imposed by the GPL license (you must make your code open-source), commercial users of Hoard can purchase non-GPL licenses through the University of Texas at Austin. Please consult the current Hoard pricing information (updated 2/5/2009), which lists a number of options for purchasing licenses, as well as software license terms and conditions, and the software license agreement: note that the main UT-Austin licensing page always contains the most up-to-date documents.
To obtain a license, please contact Jitendra Jain directly (jjain@otc.utexas.edu) and copy Emery Berger (emery@cs.umass.edu).
Jitendra Jain's full contact information follows:
Jitendra Jain
The University of Texas at Austin
Office of Technology Commercialization
MCC Building, Suite 1.9A
3925 West Braker Lane
Austin, Texas 78759
(512) 471-9055, (512) 475-6894 (fax)
There are a number of problems with existing memory allocators that make Hoard a better choice.
Multithreaded programs often do not scale because the heap is a bottleneck. When multiple threads simultaneously allocate or deallocate memory from the allocator, the allocator will serialize them. Programs making intensive use of the allocator actually slow down as the number of processors increases. Your program may be allocation-intensive without you realizing it, for instance, if your program makes many calls to the C++ Standard Template Library (STL). Hoard eliminates this bottleneck.
System-provided memory allocators can cause insidious problems for multithreaded code. They can lead to a phenomenon known as "false sharing": threads on different CPUs can end up with memory in the same cache line, or chunk of memory. Accessing these falsely-shared cache lines is hundreds of times slower than accessing unshared cache lines. Hoard is designed to prevent false sharing.
Multithreaded programs can also lead the allocator to blowup memory consumption. This effect can multiply the amount of memory needed to run your application by the number of CPUs on your machine: four CPUs could mean that you need four times as much memory. Hoard is guaranteed (provably!) to bound memory consumption.
NOTE: Make sure to invoke git as follows:
% git clone --recursive https://github.com/emeryberger/Hoard.git
To build Hoard on non-Windows platforms, change into the src/
directory and run make followed by the appropriate target. If you
type make, it will present a list of available targets. These
include linux-gcc-x86, solaris-sunw-sparc, macos, windows, and
more.
% make linux-gcc-x86-64
You can then use Hoard by linking it with your executable, or
by setting the LD_PRELOAD environment variable, as in
% export LD_PRELOAD=/path/to/libhoard.so
in Solaris:
% make solaris-sunw-sparc
% export LD_PRELOAD="/path/to/libhoard_32.so:/usr/lib/libCrun.so.1"
(32-bit version)
% export LD_PRELOAD="/path/to/libhoard_64.so:/usr/lib/64/libCrun.so.1"
(64-bit version)
or, in Mac OS X:
% make macos
% export DYLD_INSERT_LIBRARIES=/path/to/libhoard.dylib
Change into the src directory and build the Windows version:
C:\hoard\src> nmake windows
To use Hoard, link your executable with source\uselibhoard.cpp and libhoard.lib.
You must use the /MD flag.
Example:
C:\hoard\src> cl /Ox /MD yourapp.cpp source\uselibhoard.cpp libhoard.lib
To run yourapp.exe, you will need to have libhoard.dll in your path.
The directory benchmarks/ contains a number of benchmarks used to
evaluate and tune Hoard.
Hoard has changed quite a bit over the years, but for technical details of the first version of Hoard, read Hoard: A Scalable Memory Allocator for Multithreaded Applications, by Emery D. Berger, Kathryn S. McKinley, Robert D. Blumofe, and Paul R. Wilson. The Ninth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-IX). Cambridge, MA, November 2000.