[Info-area1] AVVISO SEMINARIO: State-of-the-Art in DataFlow SuperComputing for BigData DeepAnalytics (and TensorCalculus)

Gio 22 Nov 2018 12:57:32 CET

Riceviamo e inoltriamo.

===================================================================

Cari Tutti, siete gentilmente invitati al seguente seminario che si
svolgerà il giorno Martedì 4 DICEMBRE 2018 dalle ore 14:00 (precise)
nell'aula 124, edificio S. Niccolò Via Roma 56, Siena.

Il seminario durerà circa 45 minuti e, per gli interessati, sarà
seguito da una sessione di approfondimento con esempi pratici in
collegamento con il cloud di MAXELER di Londra, fino alle ore 18 con
proseguimento il giorno successivo Mercoledì 5 DICEMBRE dalle ore 11:00
alle ore 13:00.

Resto a disposizione per ulteriori informazioni.
Roberto Giorgi
roberto.giorgi a unisi.it

===================================================================

* TITLE:
DataFlow SuperComputing for BigData DeepAnalytics (and TensorCalculus)

* ABSTRACT:
This presentation analyses the essence of DataFlow SuperComputing,
defines its advantages and sheds light on the related programming model.

According to Alibaba and Google, as well as the open literature, the
DataFlow paradigm, compared to the Control Flow paradigm, offers: (a)
Speedups of at least 10x to 100x and sometimes much more (depends on the

algorithmic characteristics of the most essential loops and the
spatial/temporal characteristics of the Big Data Stream, etc.), (b)
Potentials for a better precision (depends on the characteristics of the

optimizing compiler and the operating system, etc.), (c) Power reduction

of at least 10x (depends on the clock speed and the internal
architecture, etc.), and (d) Size reduction of well over 10x (depends on

the chip implementation and the packaging technology, etc.). However,
the programming paradigm is different, and has to be mastered. This
presentation explains the programming paradigm, using Maxeler as an
example and sheds light on the ongoing research, which, in the case of
the speaker, was higlhy influenced by four different Nobel Laureates:
(a) from Richard Feynman it was learned that future computing paradigms
will be successful only if the amount of data communications is
minimized; (b) from Ilya Prigogine it was learned that the entropy of a
computing system would be minimized if spatial and temporal data get
decoupled; (c) from Daniel Kahneman it was learned that the system
software should offer options related to approximate computing; and (d)
from Andre Geim it was learned that the system software should be able
to trade between latency and precision. The approach that satisfies all
the above requirements is referred to as the Ultimate DataFlow. The
existing Maxeler programming model is applicable to Ultimate DataFlow,
too. So, the presentation concludes with the latest achievements of
Maxeler Technologies in the current year, like emulation of
Quark-related processes available thru Amazon AWS and endorsed by the
Nobel Laureate Jerome Friedman (Nobel Prize for the discovery of Quark)
and tensor calculus applicable for emulation of processes related to
QuasiCrystals (discovered by Nobel Laureate Dan Shechtman). It also
includes examples related to finances (JPMorgan and CitiBank) and
trading (Chicago Mercantile Exchange CME and NYSE), as well as those
related to: math algorithms, image processing, machine learning, and
artificial intelligence (both Google and Alibaba recently announced that

DataFlow is better for these applications compared to Control Flow). All

these examples prepare the attendees for utilization of the future
DataFlow engine of Intel, announced through a recent patent by Intel,
which was accompanied by a press release stating that DataFlow
represents the major paradigm shift in computing, in the century after
von Neumann (available on request). This presentation also offers plenty

of hands-on opportunities for attendees, related to all subjects
mentioned above and, if so requested, governed by a dedicated
professional assistant. The first 45-minutes of this presentation
correspond to the invited talk at the International SuperComputing
Conference, ExaScale Track in Frankfurt, Germany, in 2018.

===================================================================

* ABOUT THE SPEAKER:
Prof. Veljko Milutinovic

Life Member of the ACM
Life Fellow of the IEEE
Member of Academia Europaea
Member of the Serbian Academy of Engineering
Senior Advisor to Maxeler Technologies in London
Member of the Advisory Board of the Vienna Congress COMSULT

Prof. Veljko Milutinovic (1951) received his PhD from the University of
Belgrade in Serbia, spent about a decade on various faculty positions in
the
USA (mostly at Purdue University and more recenlty at the Indiana
University in
Bloomington), and was a co-designer of the DARPAs first GaAs RISC
microprocessor and the DARPAs first GaAs Systolic Array (both well
documented
in the open literature). Later, for about two decades, he taught and
conducted
research at the University of Belgrade, in EE, MATH, BA, and PHYS/CHEM.
Now he
serves as the Senior Advisor to Maxeler Technologies in London, UK, and
the
Chairman of the Board of IPSI Belgrade (a spin-off of Fraunhofer IPSI
from
Darmstadt, Germany). His research is mostly in datamining algorithms and

dataflow computing, with the emphasis on mapping of data analytics
algorithms
onto fast energy efficient architectures. For 10 of his books, forewords
were
written by 10 different Nobel Laureates with whom he cooperated on his
past
industry sponsored projects. He has over 100 SCI journal papers (mostly
in IEEE
and ACM journals), well over 1000 Thomson-Reuters citations, well over
1000
SCOPUS citations and about 4000 Google Scholar citations. Short or long
courses
on the subject he delivered so far in a number of universities
worldwide: MIT,
Harvard, Boston, NEU, Dartmouth, U of Massachusetts at Amherst, USC,
UCLA,
Columbia, NYU, Princeton, NJIT, CMU, Temple, Purdue, IU, UIUC, Michigan,

Wisconsin, Minnesota, FAU, FIU, Miami, Central Florida, Alabama,
Tennessee,
GeorgiaTech, OhioState, Imperial, King's, Manchester, Haddersfield,
Cambridge,
Oxford, Dublin, Cork, Cardiff, Edinburgh, EPFL, ETH, TUWIEN, UNIWIE,
Karlsruhe,
Stuttgart, Bonn, Frankfurt, Heidelberg, Aachen, Darmstadt, Dortmund,
KTH,
Uppsala, Karlskrona, Karlstad, Napoli, Salerno, Siena, Pisa, Barcelona,
Madrid,
Valencia, Oviedo, Ankara, Bogazici, Koc, Istanbul, Technion, Haifa,
BerSheba,
Eilat, etc, etc. Also at the World Bank in Washington DC, IMF, the
Telenor Bank
of Norway, the Reiffeisen Bank of Austria, Brookhaven National
Laboratory,
Lawrence Livermore National Laboratory, IBM TJ Watson, HP Encore Labs,
Intel
Oregon, Qualcomm VP, NCR, RCA, Fairchild, Honeywell, Yahoo NY, Google
CA,
Microsoft, Finsoft, ABB Zurich, Oracle Zurich, and many other industrial
labs,
as well as at Tsinghua, Shandong, NIS of Singapore, NTU of Singapore,
Tokyo,
Sendai, Seoul, Pusan, Sydney, Hobart, Auckland, Wellington, Toronto,
Montreal,
MexicoCity, Durango, etc.

About the Professional Assistant:

Milos Kotlar is a Ph.D. student in the School of Electrical Engineering
at the
University of Belgrade in Serbia. He is permanently employed by ABB of
Zurich,
Switzerland. His paper about Tensor Calculus on DataFlow machines, in
comparison with ControlFlow machines, was one of the two student papers
accepted for the ExaScale Workshop at the International SuperComputng
Conference in Frankfurt, Germany, in 2018.

=====================================================================

Accompanying Textbooks and Journal Papers:

Milutinovic, V., et al,
Guide to DataFlow SuperComputing,
Springer,
2015 (one textbook, part I)
and 2017 (two textbooks, parts II and III).

Hurson, A., Milutinovic, V., editors,
Advances in Computers: DataFlow,
Elsevier,
2015 (one SCI textbook)
and 2017 (two SCI textbooks).

Trifunovic, N., Milutinovic, V. et al,
"The AppGallery.Maxeler.com [1] for BigData SuperComputing,"
Journal of Big Data, Springer,
2016.

Trifunovic, N., Milutinovic, V. et al,
"Paradigm Shift in SuperComputing: DataFlow vs ControlFlow,"
Journal of Big Data,
2015.

Milutinovic, V.,
"The HoneyComb Architecture,"
Proceedings of the IEEE, 1989.

Milutinovic, V. et all,
"Splitting Spatial and Temporal Localities for Entropy Minimiation"
Tutorial of the IEEE ISCA, 1995.

Jovanovic, Z., Milutinovic, V.,
"FPGA Accelerator for Floating-Point Matrix Multiplication,"
The IET Computers and Digital Techniques Premium Award for 2014,
IET (formerly IEE), Volume 6, Issue 4,
2012 (pp. 249-256).

Milutinovic, V.,
"A Comparison of Suboptimal Detection Algorithms
(Suboptimal Algorithms for Data Analytics),"
Proceedings of the IEE (now IET),
1988.

Flynn, M., Mencer, O., Milutinovic, V., at al,
Moving from PetaFlops to PetaData,
Communications of the ACM,
May 2013.

Trobec, R. Vasiljevic, R., Tomasevic, M., Milutinovic, V., et al,
"Interconnection Networks for PetaComputing,"
ACM Computing Surveys,
November 2016.

Kotlar, M., Milutinovic, V.,
"The Tensor Calculus Operations for the Data Flow Paradigm,"
The ExaComm Workshop of the International Supercomputing Conference,
Frankfurt, Germany, June 28, 2018.

Milutinovic, V.,
"The Ultimate DataFlow",
Invited Talk at the ExaComm Workshop of the ISC,
Frankfurt, Germany, June 28, 2018.

================================================

-- 

----------------------------------

Q: Why is this email four sentences or less?
A: http://four.sentenc.es
----------------------------------
Roberto Giorgi, PhD --- http://www.dii.unisi.it/~giorgi

Links:
------
[1] http://AppGallery.Maxeler.com