Download System and Environments for High Performance Java - BIS-21++

Systems and Environments
for High Performance
Java Computing
Vladimir Getov
4 January 2006
Part 1: Introduction
Relevant publications:
 V. Getov. Java in High-Performance Computing –
Guest Editorial. FGCS, vol. 18(2), v-vi, Oct. 2001.
 M. Philippsen, R. Boisvert, V. Getov, R. Pozo, J.
Moreira, D. Gannon, G. Fox. JavaGrande – High
Performance Computing with Java. Proceedings of
PARA 2000 Conference, LNCS, Springer, vol. 1947,
20-36, 2001.
 High Performance Computing Using Java Technology
– Tutorial lecture. ACM JavaGrande and
JavaOne2001 Conferences, San Francisco, June
2001.
Some Important Facts About
Current Computer Systems
 The gap between processing speed and memory
access speed continues to grow;
 So does the gap between high-level
programming models and underlying hardware
architectures;
 The wide variety of hardware architectures
makes it particularly difficult to achieve portable
high performance;
 The pace of innovation is such that investment in
tuning for one machine may not pay off before
that machine is obsolete.
Pros/Cons of using Java
 Pros: Java offers tremendous potential for
portability and heterogeneous execution Bytecode Representation, RMI, Object
Serialization
 Cons: Java still suffers a significant
performance penalty and as with any new
language, the thought of rewriting existing
codes brings reluctance and lack of
enthusiasm.
Background Observations
 Java thread model is insufficient
 Message Passing model is important to
support
 Performance is critical

Many applications need “high” performance
 Proper numerical computing

Complex, arrays, performance, reproducibility
Part 2: High Performance Java
Relevant publications:
 V. Getov. A Mixed-Language Programming
Methodology for High Performance Java Computing.
In: R. Boisvert and P. Tang (Eds.) The Architecture of
Scientific Software. Kluwer Academic Publishers,
333-347, 2001.
 Q. Lu, V. Getov. Mixed-Language High-Performance
Computing for Plasma Simulations. Scientific
Programming, vol. 11(1), 57-66, 2003.
Mixed-Language Programming
with Java
Java is a highly-portable language
Java adheres to the “Write once, run anywhere” philosophy
Java has a well-established collection of scientific library bindings
Java’s execution speed is suitable for HPC
C/Fortran are highly-portable languages
C/Fortran adhere to the “Write once, run anywhere” philosophy
C/Fortran have well-established scientific libraries
C/Fortran execution speeds are suitable for HPC
So, What Language to Use?
Java is a highly-portable language
Java adheres to the “Write once, run anywhere” philosophy
C/Fortran have well-established scientific library bindings
C/Fortran execution speeds are suitable for HPC
Utilize Java for its portability and standardization,
but focus on using Java as a wrapper for porting
of native code in the form of shared libraries. This
involves the least amount of work and guarantees
maximum performance on different platforms.
Difficulties in binding a native
library to Java
 Data formats in Java and C differ:




sizes of primitive types;
C pointers;
multidimensional arrays;
C structures;
 Still different Java native method interfaces
exist;
 A native interface is inadequate for calling
existing library functions.
JCI Block Diagram
Legacy libraries bound to Java
so far
Library
Lang. Func.
MPI
C
128
BLACS
C
76
BLAS
F77
21
PBLAS
C
22
PB-BLAS
F77
30
LAPACK
F77
14
ScaLAPACK F77
38
C
Java
4434 439
5702 489
2095 169
2567 127
4973 241
765
65
5373 293
Mixed-language programming
based on JVM
NPB EP kernel on IBM SP2
Mixed-language programming
based on HPCJ
NPB IS kernel on IBM SP2
Part 3: Message Passing in Java
Relevant publications:
 B. Carpenter, V. Getov, G. Judd, A. Skjellum, G. Fox.
MPJ: MPI-like Message Passing for Java.
Concurrency: Practice and Experience, vol. 12 (11),
1019-1038, 2000.
 V. Getov, P. Gray, V. Sunderam. Aspects of
Portability and Distributed Execution for JNI-Wrapped
Message Passing Libraries. Concurrency: Practice
and Experience, vol. 12 (11), 1039-1050, 2000.
 V. Getov, M. Philippsen. Java Communications for
Large-Scale Parallel Computing. Proceedings of
SciComp'01 Conference, LNCS, Springer, vol. 2179,
33-45, 2001.
Message Passing - Motivation
 The existing communication packages in Java
- RMI, API to BSD sockets - are optimized for
Client/Server programming
 The symmetric model of communication is
captured in the MPI standard - MPI-1 and
MPI-2
 An MPI-like message-passing API
specification is needed to enable the
development of portable JavaGrande
applications
Early MPI-like Efforts - 1
 mpiJava - Modeled after the C++ binding for MPI.
Implementation through JNI wrappers to native MPI
software.
 JavaMPI - Automatic generation of wrappers to
legacy MPI libraries. C-like implementation based on
the JCI code generator.
 MPIJ - Pure Java implementation of MPI closely
based on the C++ binding. A large subset of MPI is
implemented using native marshaling of primitive
Java types.
Early MPI-like Efforts - 2
 JMPI - MPI Soft Tech Inc. have announced a
commercial effort under way to develop a message
passing environment for Java.
 Others
 Existing ports - Linux, Solaris (both WS clusters
and SMPs), AIX (both WS clusters and SP2),
Windows NT clusters, Origin-2000, Fujitsu AP3000,
and Hitachi SR2201.
 Java + MPI codes - growing variety including
full applications
MPJ API Specification
 Builds on the MPI-1 Specification and the Java
Specification.
 Immediate standardization for common message
passing programs in Java
 Basis for conversion between C, C++, Fortran
and Java.
 Eventually, support for aspects of MPI-2 as well
as possible improvements to the Java language.
Multidimensional arrays
 In Java an “n-dimensional array” is equivalent
to a one-dimensional array of (n - 1)dimensional arrays.
 In MPJ, message buffers are always onedimensional arrays, but element type may be
an object, which may have array type - hence
multidimensional arrays can appear as
message buffers.
Java multidimensional arrays
A
[
0
]
[
0
]
A
[
0
]
[
1
]
A
[
0
]
[
2
]
A
[
0
]
[
3
]
A
[
0
]
A
[
1
]
[
0
]
A
[
1
]
[
1
]
A
[
1
]
[
2
]
A
[
1
]
[
3
]
A
[
1
]
Array of
Arrays
A
[
2
]
[
0
]
A
[
2
]
[
1
]
A
[
2
]
[
2
]
A
[
2
]
[
3
]
A
[
2
]
A
[
3
]
[
0
]
A
[
3
]
[
1
]
A
[
3
]
[
2
]
A
[
3
]
[
3
]
A
[
3
]
Java multidimensional arrays
[
0
]
[
0
]
A
[
0
]
[
1
]
A
[
0
]
[
2
]
A
[
0
]
[
3
]
A
[
0
] A
B
[
0
]
[
1
]
[
0
]
A
[
1
]
[
1
]
A
[
1
] A
B
[
1
]
[
2
]
[
0
]
A
[
2
]
[
1
]
A
[
2
]
[
2
]
A
[
2
]
[
3
]
A
[
2
] A
B
[
2
]
[
3
]
[
0
]
B
[
3
]
[
1
]
B
[
3
]
[
2
]
A
[
3
] B
B
[
3
]
Java multidimensional arrays are not indivisible objects: could
have intra-array aliasing and "partial overlaps" with other arrays
Naming Conventions
 All MPI classes belong to the package mpi.
 Conventions for capitalization, etc, in class
and member names
generally follow the recommendations of
Sun's Java code conventions
 consistent with the MPI C++ binding
Error codes
 Unlike the C and Fortran interfaces, the Java
interfaces to MPI
calls will not return explicit error codes.
 Instead, the Java exception mechanism will
be used to report errors
Ping-Pong Timings
Execution time (sec)
1e-2
Java/LAM-MPI
C/LAM-MPI
C/IBM-MPI
1e-3
1e-4
1e+1
1e+2
1e+3
1e+4
Message length (bytes)
1e+5
1e+6
Part 4: Grid Systems and
Environments
Relevant publications:
 V. Getov, G. von Laszewski, M. Philippsen, I. Foster. Multi-
Paradigm Communications in Java for Grid Computing.
Communications of the ACM, vol. 44(10), 118-125, Oct. 2001.
 V. Getov, M. Gerndt, A. Hoisie, A. Malony, B. Miller (Eds.)
Performance Analysis and Grid Computing. Kluwer Academic
Publishers, 2003.
 V. Getov, S. Newhouse, O. Rana, E. Sharakan. Developing Grid
Services with Jini and JXTA. Proc of ICCC 2004, 1402-1408,
ICCC Press, 2004 (best paper award).
 V. Getov, A. Puliafito, O. Rana. Computational Grid and Web
Services: Concepts, Functionalities, and Comparisons. Proc of
ICCC 2004, 10-15, ICCC Press, 2004.
Roadmap of Communication
Frameworks
What is the Grid ?
“A Grid provides an abstraction for resource sharing and collaboration across
multiple administrative domains…”
(Source: NGG Expert Group, 16 June 2003 “European Grid Research 2005-2010)
Benefits

Increased productivity by reducing the total cost of ownership

Any-type, anywhere, anytime services by/for all

Infrastructure for dynamic virtual organisations

Backbone for the next generation Internet
services
Industry & Business
Grids
e-Science
Java in Grid Computing
 Main motivation - need to solve bigger
problems with resource requirements beyond
the current limits
 Recent advances in computer communications
make it possible to couple geographically
distributed resources - Grid computing
 In contrast with low-level approaches Java can
support a single object-oriented communication
framework for Grande applications
Example Application:
An Advanced Scientific Instrument
Virtual Reality Cave
Advanced Photon Source
Scientist
Avata
r
Supercomputer
Electronic Library
and Databases
Computing Portal
Clients
Lightweight Grid Platform
 New generic approach to designing the next
generation Grid systems with dynamic properties –
components-based design.
 To develop a lightweight Grid platform suitable for
resource limited devices, to support our design.
 To provide a design that will allow for the efficient
integration of mobile devices into the Grid.
 To provide enhanced security, centralized
management and monitoring, roaming, fault tolerance
and a high level of autonomy in this mobile wireless
environment.
Challenges in Mobile Grids
 Limited available resources.
 Increased power consumption sensitivity.
 Increased heterogeneity and software non-
interoperability.
 Unpredictable long periods of complete
disconnectivity.
 Unreliable, low-bandwidth and high latency
communication links.
 Very frequent, dynamic and unpredictable
changes to the network layout.
Hybrid Environment: Virtual
“Cluster” Approach
Clustered Approach Benefits
 Single point of entry to the wireless cluster.
 Centralized cluster management and
monitoring.
 Encapsulation of heterogeneity and
dynamicity.
 Masking of internal failures and silent
recovering locally without affecting the regular
Grid operation.
Ten Reasons to Use Java in
High-Performance Computing
 Language
 Maintenance
 Class Libraries
 Performance
 Components
 Gadgets
 Deployment
 Industry
 Portability
 Education
Acknowledgements











Bryan Carpenter (Uni Syracuse)
Susan Flynn-Hummel (IBM - T.J. Watson)
Gregor von Laszewski (Argonne NL)
Sava Mintchev (Uni Westminster)
Jose Moreira (IBM - T.J. Watson)
Michael Philippsen (Uni Karlsruhe)
Antonio Puliafito (Uni Messina)
Omer Rana (Uni Cardiff)
Eric Sharakan (Sun Microsystems)
Mary Thomas (San Diego Supercomputer Center)
Experiments - CTC, IBM - T.J. Watson, SDSC,
Southampton and Westminster Universities